Trust in scientists on COVID-19 vaccine hesitancy and vaccine intention in China and the US

Abstract

When we recognize various factors which influence vaccine willingness, it is unclear whether these factors work the same in different countries. This study explored how trust in scientists was related to COVID-19 vaccine hesitancies and vaccine willingness in China and the U.S. We attempted to understand the extent to which the perceived trustworthiness of scientists would predict vaccine hesitancies and intention to get vaccinated differently in these two countries. An online survey with participants in China (N = 391) and in the U.S. (N = 363) was conducted. Participants were asked about their view of scientists' competence, benevolence, and integrity as dimensions of trustworthiness, three types of vaccine hesitancies, as well as their willingness to get vaccinated. The results showed that trust in scientists was negatively related to individual vaccine hesitancy in both China and the U.S., and trust was negatively associated with the level of contextual hesitancy and vaccine-specific hesitancy in the U.S. Higher trust in scientists was also associated with the willingness to get vaccines in the U.S., rather than in China. Results yielded differences in China and the U.S. in how trust in scientists and vaccine hesitancies predicted individuals’ willingness to get vaccinated.

1. Introduction

Despite the increasing availability of COVID-19 vaccines, concerns regarding vaccine safety and other obstacles which prevent people from getting vaccinated have affected the pandemic mitigation movement. Vaccine hesitancy, which is defined as a delay in the acceptance or refusal to vaccinate, has been reported in more than 90% of countries in the world [1]. While vaccine hesitancy is an important indicator of people's attitude and intention of getting vaccinated, it does not appear to be the only factor that determines the actual vaccine decision [2]. Additionally, it is unclear to what extent the relationship between vaccine hesitancy and willingness to get vaccinated varies in different countries. Thus, understanding the factors that contribute to vaccine hesitancy and willingness to get vaccinated could provide better insights to develop vaccine promotion strategies and increase vaccine uptake.

Previous studies showed that many factors could contribute to vaccine hesitancy as well as vaccine uptake, including knowledge and information sources, experiences with vaccination and relevant diseases, health professionals and their recommendations, the public health system, social norms, and parental responsibilities, appraisals of vaccine safety and effectiveness, and religious beliefs [3]. However, little research has been done to explore potential factors outside the health practice circle, such as the relationship with scientists, particularly the trust belief with scientists. Developed over time, trust in the current context reflects the relationship between scientists and the public [4]. Different from health experts such as doctors, who are more visible to the public, scientists have limited direct interactions with the public and the development of trust may be built through media exposure and other experiences [5]. It is unclear how trust in scientists plays a role in individuals' responses to COVID-19 vaccines. Therefore, the present study extends prior research by focusing on trust in scientists as a potential antecedent of COVID-19 vaccine hesitancy to understand how people's beliefs in science may shape their willingness to receive the vaccination.

We also recognize that how trust in scientists relates to vaccine hesitancy and vaccination may vary in different countries, specifically China versus the U.S. To our best knowledge, no research studied vaccine hesitancy between China and the U.S. except one which compared the level of HPV vaccine hesitancy and the specific issues people in both countries are concerned about [6]. Furthermore, prior research suggests that the object of and the reason for mistrust in vaccines differed between western and East Asian countries[7]. Specifically, those demonstrating a high level of vaccine hesitancy in western countries were skeptical of the effect of general vaccination primarily because of their political ideology[7]. Additionally, the high level of mistrust in general vaccination among certain groups in the U.S. could be derived from vaccine-related historical events in Europe and America, in which human rights such as freedom and safety were violated [8]. In other words, vaccine hesitancy among Americans might be driven by political and historical factors and was usually related to the act of vaccination. On the contrary, in China, individuals who exhibited a high level of vaccine hesitancy were more concerned about the safety of specific vaccines [[6], [7], [9]]. This might be because vaccine-related scandals in China undermined public confidence in vaccines [10,11]. As prior research categorized three types of reasons which contributed to vaccine hesitancy [12,13], it is necessary to compare the role that each type of vaccine hesitancy played in shaping individual willingness to get vaccinated in China and the U.S. Specifically, we aim to examine how trust in scientists may be related to the willingness to receive the COVID-19 vaccine via three types of vaccine hesitancy and how the country may moderate this process. Results of this investigation can provide a plausible model which explains vaccine hesitancy in two major countries that have fundamentally different socio-cultural and political systems, thereby offering implications that can guide vaccination campaigns across different cultures. In addition, as the process by which trust in scientists predicts willingness to get vaccinated through vaccine hesitancy was proposed and developed in the western context, it is necessary to test whether this relationship holds in China, where the COVID-19 mitigation movement is highly politicalized [14]. It is worth noting that the purpose of this study is not to explore the cultural or political differences that potentially affect trust in scientists or vaccine hesitancy, but to diversify the findings in different countries.

1.1. Vaccine hesitancy

Vaccine hesitancy is one of the most discussed concepts that reflect vaccine refusal and vaccine uptake. Different from vaccine attitude or belief, which can be either positive or negative, vaccine hesitancy addresses the subjective and objective obstacles that prevent people from accepting vaccination [[2], [15]]. A review of studies showed the difficulty to predict vaccine hesitancy, probably due to different social, cultural, political, and personal factors based on individual studies [13]. However, some commonly agreed factors include knowledge and information, past experiences, perceived importance of vaccination, risk perception and trust, subjective norm, religious or moral convictions [2]. Among them, a key predictor that consistently predicted the willingness to get vaccinated is confidence [[7], [16], [17]], conceptualized as “trust in the effectiveness and safety of vaccines, the system that delivers them, including the reliability and competence of the health services and health professionals and the motivations of policy-makers who decide on the needed vaccines” [17]; p. 1169). In other words, the object of vaccine-related confidence includes the vaccine itself [[9], [10], [11],17,18] and people who advocate for and deliver it [[19], [20], [21]]. For example, vaccine opponents reported a higher level of mistrust in health professionals, scientists, and the government than vaccine supporters [20]. Likewise, public confidence in the safety and effectiveness of flu vaccines enhanced the level of vaccine hesitancy [17].

However, it is worth noting that vaccine hesitancy only reflects individuals' views on a continuum ranging from active request vaccine to complete refusal; hesitancy does not suggest a direct relationship with the final vaccine uptake [2]. Additionally, vaccine hesitancy does not just reflect the concerns about vaccine efficacy or safety (confidence); it also contains whether the person finds it is needed (complacency) or whether it is convenient to get vaccinated (convenience) [13]. Therefore, in the current study, we adopted the three dimensions of vaccine hesitancies from previous research [12,13], including: contextual influences, which occur due to historical, socio-cultural, or environmental factors; individual/group influences, which arise from the personal perception of the vaccine or influences of the peer environment; and vaccine-specific issue, which refers to concerns directly related to vaccine or vaccination. In the current study, besides examining how these vaccine hesitancies affect individuals' willingness to accept the COVID-19 vaccine, we want to further explore the factors contributing to vaccine hesitancy. Thus, the first hypothesis we propose is the negative relationship between vaccine hesitancies and individuals’ willingness to get vaccinated.

H1

(a) contextual vaccine hesitancy, (b) individual hesitancy, and (c) vaccine-specific hesitancy are negatively related to the willingness to get vaccinated.

Although vaccine hesitancy has been extensively discussed, the investigation of the antecedents of vaccine hesitancy still needs to be further explored, as it provides possible pathways for communicators or health decision-makers to reduce vaccine hesitancy and increase vaccine acceptance. Studies that examined the causes of vaccine hesitancy have identified several factors based on the work of the theory of planned behavior and risk decision-making, including perceived susceptibility, severity, attitude towards the vaccine, norm, behavioral control, and anticipated regret [22], as well as social or environmental factors such as parental responsibility and religious beliefs [3].

1.2. Trust in scientists

Few studies have looked at the relationship with scientists as an influencing factor of vaccine hesitancy. The rationale for exploring the role of trust in scientists is that vaccines are the product of scientific findings based on the work of scientists, which can be challenging for the public to fully understand. Thus, what would possibly influence individuals’ confidence in vaccines may result from their opinion of and belief in scientists [4], which can be viewed from the trust perspective.

Trust is described as “the willingness to be vulnerable to the actions of a trustee on the basis of the expectation that the trustee will perform a particular action”[23]. Trust also reflects people's expectations that the interactor is dependable and can be relied on to deliver promises [24]. In the current context, this vulnerability can be interpreted as the willingness that people would believe in the work of scientists, with or without fully understanding how the COVID-19 vaccine works. Trust is not a new concept for understanding individuals' vaccination decisions and vaccine hesitancies, and has been discussed from several perspectives, including trust in science [25], trust in health care providers [26], and trust in scientists [4]. However, it is not yet fully examined, especially the extent of how trust was perceived by the public. Most studies generally asked participants the extent to which the public trusts scientists, where they see trust as a single, manifest factor. However, trust is a complex concept that needs to be broken down into multiple dimensions. More specifically, researchers illustrated that trustworthiness, as the attribute of a trustee that inspires trust from others, is captured via three characteristics: competence or ability, which reflects the skills or efficacy of the trustee; benevolence, which reflects the sense that the trustee cares and wants to “do good”; and integrity, which reflects a principle or ethical value the trustee holds [27]. Researchers found that the public generally perceives high competence from researchers and scientists and moderate levels of warmth [28], but no data are shown on the integrity dimension yet.

Trust in scientists works as an important factor for many science-related decision makings, such as climate change or GMO issues [4], especially when the public does not have the expertise to make the decision. Findings from a limited amount of research showed that trust in scientists provides more confidence for the public and therefore generates a positive impact on vaccine decisions, including reducing vaccine hesitancy [29,30]. In the current study, we argue that trust in scientists would be negatively related to vaccine hesitancy. Trust in scientists can reduce the level of uncertainty individuals have about the vaccine and are likely to cause less vaccine hesitancy. Additionally, high trust in scientists would be more likely to encourage individuals to get vaccinated.

H2

Trust in scientists is negatively associated with vaccine hesitancy.

H3

Trust in scientists is positively associated with willingness to get vaccinated.

H4

Vaccine hesitancies mediate the relationship between trust in scientists and willingness to be vaccinated.

1.3. Vaccine and trust in scientists in China and the U.S.

While we propose that vaccine hesitancy and trust in scientists predict the willingness to get vaccinated, we recognize that both vaccine hesitancy and trust in scientists are rooted in geopolitical and cultural contexts. There has been a wealth of research on vaccine hesitancy in different countries which provided empirical evidence on different reasons for vaccine hesitancy across countries. In the U.S, political orientation [19,31], religious beliefs[3], community norms [3], concerns about vaccine safety and effectiveness [32] were significant factors for vaccine hesitancy. In China, concerns about vaccine safety [[9], [10], [11],18,33] and distrust in the government, healthcare industries, and vaccine manufacturers [11,33] were identified as critical factors of vaccine hesitancy. However, research that compares the effect of trust in scientists on vaccine hesitancy in these two countries is rare [6]. The interference of science literacy, science engagement, or even politics might affect the role that trust in scientists and vaccine hesitancy play in vaccination acceptance.

Specifically, a few studies on students' perception of scientists in China and the U.S. show that Chinese students’ perceptions of scientists were more consistent compared to American students [34,35] because the education system in China allowed for little room for differences [34]. Moreover, research suggests that the public perception of scientists might be more positive in China than in the U.S. [[34], [36], [37]] . Additionally, some media in the U.S., particularly conservative media, lowered public trust in scientists [36] whereas Chinese people tend to consider scientists positively such as being intelligent, hardworking, and making contributions to society [37]. Taken together, Chinese perception of scientists might be more consistent and positive than Americans. Although this argument does not directly pinpoint the potentially different relationships between trust in scientists and vaccine acceptance in these two countries, it provides the rationale to include the country as a moderating variable between trust and hesitancy as well as between trust and willingness to vaccinate.

Based on the literature review and the concepts we introduced above, we propose the following research question on the potential difference between China and US participants on vaccine hesitancy, and hypotheses on the effects of trust in scientists and vaccine hesitancies on willingness to be vaccinated. Fig. 1 presents the conceptual model that we proposed.

RQ1

To what extent does the country difference moderate the relationship between trust in scientists and willingness to get vaccinated?

RQ2

To what extent does the country difference moderate the relationship between trust in scientists and vaccine hesitancies?

Fig. 1.

Conceptual framework of the proposed model.

2. Method

Upon IRB approval at the first author's institution, a survey was conducted in both China and the U.S. at the same time in March 2021. A nationwide probability sample was recruited via Qualtrics, to maximize the representation of the population. More specifically, quotas on participants' age, gender, race, income, education, and region provided by Qualtrics were used in data collection. Data from participants who did not receive any vaccination were selected for the data analysis. A total of 391 participants were surveyed in China, and a total of 363 participants completed the U.S. survey. The survey questions in China were translated into Chinese by the researchers.

2.1. Procedure

The survey was developed online. After receiving informed consent from participants, participants were first asked about their demographic information to fill up the quota. They were then instructed to provide answers to their willingness of getting vaccinated. After indicating their decisions, we showed participants the list of vaccine hesitancies and asked them to rate each of them. Then, participants provided their evaluation of trust (competence, benevolence, and integrity) in scientists in COVID-19 on a 5-point scale. Participants were debriefed with the purpose of this study as well.

2.2. Key measures

In the proposed model, vaccine hesitancies were captured by three categories [12]. Contextual influence (Cronbach alpha = .85) includes 5 items: I do not know where to get good/reliable information, I heard or read negative media reports on vaccines before, Someone else told me that the vaccine is not safe, Someone else told me that they had a bad reaction, Religious reason. Individual hesitancy (Cronbach alpha = .87) includes 5 items: I do not think it is needed, I had a bad experience with a previous vaccinator/health clinic, I had a bad experience or reaction with previous vaccination, fear of needles, other beliefs/traditional medicines. Vaccine-specific hesitancy includes 4 items (Cronbach alpha = .81): I don't know where to get the vaccination, I do not think the vaccine is effective, I do not think the vaccine is safe (concerned about side effects), I do not have time to get vaccinated. The average of these items was computed as the indicator of the level of contextual, individual, and vaccine-specific hesitancy.

Trust in scientists was measured involving three dimensions [27], with one item each: when it comes to handling the pandemic, scientists or researchers are capable at their work (competence), scientists or researchers really look out what is important to my life (benevolence), scientists or researchers have a strong sense of justice (integrity). Participants rated each statement from “never” (1) to “always” (5).

Willingness to get vaccinated was measured by one question “to what extent will you take COVID-19 vaccine when it becomes available to you” (1 = extremely unlikely, 5 = extremely likely).

2.3. Statistical analysis

We conducted an ordinal least squares (OLS) regression to test H1. PROCESS Macro was utilized to test the moderation hypotheses (H2 – H4) with effect-coded message conditions. PROCESS for SPSS can estimate the coefficients of a model using OLS regression as well as generate the conditional effects in moderation. Specifically, trust in scientists was entered as the independent variable, all three types of hesitancies were entered as parallel mediators, the country was entered as the moderator, and willingness to get vaccinated was entered as the dependent variable. Additionally, we included individuals’ age, gender, education, and self-perception of COVID severity as control variables.

3. Results

3.1. Demographics

To ensure the findings can be representative, the researchers requested a national probability sample from Qualtrics with participants over 18 years old. Qualtrics used the most available U.S. census and determined the quota. Because there was no China census data available, Qualtrics used a quota self-determined by the panel company. We used a couple of attention check questions to control the data quality. We also screened out participants who had already been vaccinated, given the study's focus on vaccine hesitancy. In the final sample, the U.S. data included 53.6% female and 46.1% male, with 0.4% of the participants not identifying their biological sex. Over a quarter of the participants (26.7%) were in the 18–34 age group, 34.7% in the 35–54 age group, and 38.6% in the 55+ age group. Over half of the participants (52.9%) received a High School diploma/GED or less, 7% received some college education, about 39.2% received college or higher education, and 3.5% did not indicate their education level.

The Chinese sample included 52.5% female and 42.5% male. Less than one third of the participants (31.5%) were in the 18–34 age group, 43.4% in the 35–54 age group, and 25.1% in the 55+ age group. About 9.7% received a High School diploma/GED or less, 15.9% received some college education, and about 74.4% received college or higher education.

Prior to examining the hypotheses, we first checked how participants in China and the U.S. viewed the three types of COVID-19 vaccine hesitancies and whether the differences between China and the U.S were significant. We conducted an ANOVA test and the results showed no significant differences between these two countries in terms of the three types of vaccine hesitancies. The average mean showed that the level of hesitancy were around midpoint (on a 5-point scale): contextual hesitancies (M _China = 2.45, SD _China = 1.14; M _US = 2.34, SD _US = 0.92), F (1, 752) = 1.97, p = .16); Individual hesitancies (M _China = 2.13, SD _China = 1.29; M _US = 2.04, SD _US = 0.94), F (1, 752) = 1.09, p = .30); vaccine hesitancies (M _China = 2.42, SD _China = 0.06; M _US = 2.36, SD _US = 0.92), F (1, 752) = 0.62, p = .43). Table 1 presents the descriptive of key measures in China and the U.S.

Table 1.

Descriptive results of items in key measures.

	M(SD) – U.S.	M(SD) - China	Significant difference
Perceived COVID severity	3.04 (1.21)	2.03 (1.15)	p < .001
Trust in Scientists	3.63 (.95)	4.30 (.42)	p < .001
Contextual hesitancy	2.35 (.92)	2.45 (1.15)	p = .16
Individual hesitancy	2.05 (.94)	2.13 (1.28)	p = .31
Vaccine hesitancy	2.35 (.91)	2.43 (1.30)	p = .33
Vaccination willingness	3.66 (1.51)	4.36 (.74)	p < .001

3.2. Hypotheses and research questions

H1 proposed that the level of vaccine hesitancies would lead to less willingness to get vaccinated. To examine this hypothesis with data in China and the U.S., we conducted an OLS regression analysis with all three types of hesitancies entered (age, gender, education, and perceived severity were entered as covariates). However, when the three types of vaccine hesitancy were entered the model simultaneously, the variance inflation factor ranged between 4 and 5, indicating the problem of multicollinearity [38]. Therefore, we conducted a series of OLS regression analyses with only one type of vaccine hesitancy included, controlling for age, gender, education, and perceived severity, first with the combined sample and subsequently with Chinese and American samples separated.

In the combined sample, the OLS regression revealed that contextual hesitancy (p = .72; F(5, 727) = 13.41, p < .001; R² = 0.08), individual hesitancy (p = .63; F(5, 727) = 13.43, p < .001; R² = 0.08), and vaccine-specific hesitancy (p = .23; F(5, 727) = 13.70, p < .001, R² = 0.08) were not significantly related to the willingness to get COVID-19 vaccinated. However, in the American sample, contextual hesitancy (β = −0.26, p < .001; F(5, 347) = 24.93, p < .001, R² = 0.26), individual hesitancy (β = −0.26, p < .001; F(5, 347) = 25.19, p < .001, R² = 0.27), and vaccine-specific hesitancy (β = −0.34, p < .001; F(5, 347) = 31.45, p < .001, R² = 0.31) were all negatively related to the willingness to get vaccination. On the contrary, in the Chinese sample, contextual hesitancy (β = 0.30, p < .001; F(5, 374) = 6.46, p < .001, R² = 0.07), individual hesitancy (β = 0.35, p < .001; F(5, 374) = 9.50, p < .001, R² = 0.11), and vaccine-specific hesitancy (β = 0.30, p < .001; F(5, 374) = 7.21, p < .001, R² = 0.08) were all positively associated with the willingness to receive vaccination. Therefore, H1 only receives support in the American sample.

H2 and H3 predicted the effects of trust in scientists on vaccine hesitancies and willingness to get vaccinated. H4 explored the mediation of vaccine hesitancies between trust in scientists and willingness to get vaccinated. RQ1 and RQ2 added the country as the moderator to the model. Therefore, we used PROCESS for SPSS to examine the potential mediating effect of vaccine hesitancy between trust in scientists and willingness, as well as the moderation effect of the country. The results (Table 2 ) first reported the main effects of trust in scientists, country, and the interaction on three types of vaccine hesitancies, then report the effects of vaccine hesitancies on willingness to get vaccines, as well as the significant indirect effects. The results showed that trust in scientists predicted all three types of vaccine hesitancies negatively (B _contextual = −0.76, B _individual = −0.48, B _vaccine = 0.80, p < .001), and trust in scientists had a positive effect on willingness to get vaccinated (b _willingness = 1.28, p < .001). Therefore, H2 and H3 are supported.

Table 2.

Coefficients of the mediated moderation model.

The three types of vaccine hesitancy regressed on trust in scientists
	Contextual hesitancy	Individual hesitancy	Vaccine-specific hesitancy
Trust in scientists (H2)	−.76***	−.48**	−.80***
Country (1 = US, 2 = China)	3.65***	4.24***	3.49***
Trust * Country (RQ1)	.40**	.17	.39*
Severity	.36***	.34***	.34***
Age	−.01	.00	−.00
Gender	−.00	.01	.01
Education	.02	.02	−.01
R2	.17	.12	.14
F (df1, df2)	21.56 (7726)	13.92 (7, 726)	16.64 (7, 726)

Vaccination willingness regressed on trust in scientists & three types of vaccine hesitancy
Variable names	B, p
Trust in scientists (H3)	1.28***
Contextual hesitancy	−.07
Individual hesitancy	.15*
Vaccine-specific hesitancy	−.12
Country (1 = US, 2 = China)	5.29***
Trust * Country (RQ2)	0.56***
Severity	.19***
Age	.00
Gender	.01
Education	.06*
R2	.33
F (df1, df2)	35.01 (10, 723)

H4 tested the mediation effect of hesitancies between trust in scientists and willingness. The results showed that only individual hesitancy served as a significant mediator in the U.S. data (B_{indirect effect}: −0.04; 95% CI [-0.10, −0.01]). In other words, trust in scientists was negatively related to the level of individual hesitancy, which predicted willingness positively. H4 is only partially supported with U.S. participants. Fig. 2 presented the significant paths in mediation analysis with U.S., no significant path is found with China samples.

Fig. 2.

Significant paths of mediation analysis with U.S. sample.

RQ1 and RQ2 proposed the moderating effect of countries with trust in scientists on vaccine hesitancies and willingness to be vaccinated. The results showed that trust in scientists was negatively associated with contextual hesitancy in the U.S. (B = −0.36, 95%CI [-0.57，0.25]), p < .001), but no significant relationship was observed in China (p = .75). Similarly, trust in scientists was negatively related to vaccine-specific hesitancy in the U.S. (B = −0.41, 95%CI [-0.52,.24], p < .001), but no significant result was found among Chinese respondents (p = .90). No significant interaction was observed with individual hesitancy (p = .24). However, trust in scientists predicted individual hesitancy significantly only in the U.S. (B = −0.31, 95%CI [-0.44, −0.19], p < .001) but not in China (p = .27).

Additionally, we found that the country also significantly moderated the relationship between trust in scientists and vaccination willingness. Specifically, trust was a significant positive predictor of vaccination willingness (B = .71, 95%CI [0.59, 0.83], p < .001), but it was not in China (p = .22).

4. Discussion

Although many types of vaccine hesitancies were identified, little was known on how trust in scientists would affect vaccine hesitancies and willingness to vaccinate. This knowledge, however, is important as the interference of politics in COVID-19 mitigation may affect individuals’ perception of science and their trust in scientists [4]. This study adds insights to the current body of literature on vaccine hesitancy by testing the relationship between trust in scientists, vaccine hesitancy, and willingness to get vaccinated in both China and the US. In general, the data showed a clearer relationship between the aforementioned relationship in the U.S., but not many significant relationships were observed with participants in China.

First, as predicted, trust in scientists was negatively associated with the level of three types of vaccine hesitancies and was positively related to the willingness to get vaccinated. However, the results of the interaction analyses reveal that these relationships are more observed with participants in the U.S. There are several possible explanations for these findings. The initial scale of the vaccine hesitancy [12] was developed with participants from multiple countries, but China was not on the list. Thus, it is possible that the current list of vaccine hesitancies was not able to capture all the concerns that individuals in China have. Additionally, as reviewed earlier, Chinese public perception of scientists might be more consistent than Americans' given the more unified educational system in China versus the more polarized media landscape in the U.S. Hence, there might be less variation in terms of Chinese’ trust in scientists than Americans’. This might explain why we did not find trust in scientists was significantly related to vaccine hesitancy and willingness to get vaccinated in the Chinese sample. As there is limited research on vaccine hesitancy in China, our findings present a preliminary effort for this line of scholarship and suggest that the framework applied to the western countries might not explain vaccine hesitancy and vaccination in China. Future research may want to study COVID-19 vaccine hesitancy in China further based on exploratory qualitative studies to understand unique concerns held by Chinese residents.

In addition, how the three types of vaccine hesitancies are related to willingness to get vaccinated differed between China and U.S. Specifically, in the U.S., all three types of vaccine-specific hesitancy are likely to lead to less willingness to get vaccinated. However, in China, higher vaccine hesitancy was associated with more willingness to get vaccinated. This counterintuitive finding suggests that the relationship between vaccine hesitancy and receiving COVID-19 vaccination in China is not as simple as the concept predicts. Notably, some research also found the results in China did not align with the extant studies in western countries [39]. Although Chinese individuals may hold a negative attitude towards COVID-19 vaccines, they are still willing to receive the vaccination, which might result from the unique sociopolitical environment in China. For example, a news article reported China may plan to issue public space bans on individuals who have not been vaccinated [40]. In fact, the COVID-19 mitigation movement in China is highly politicalized [14]. Therefore, fear of repercussions and political ideology may change the relationship between vaccine hesitancy and willingness to vaccinate among Chinese individuals. In other words, it is possible that those feeling more negative about the vaccine were more aware of the negative consequences they might face if they do not get vaccinated, thereby demonstrating more willingness to get vaccinated. Additionally, this study was conducted in March 2021, when COVID-19 vaccines were still scarce in China. As scarcity is usually perceived as high value [41], the perceived scarcity of the vaccine might also change the relationship mentioned above in China. These results need to be interpreted with caution as the Chinese sample may not represent the population. Regardless, more investigations are needed to understand the complex mechanism underlying this finding and develop a theoretical framework that can explain vaccine hesitancy and vaccination in China.

Additionally, the mediation analysis attempted to explore whether vaccine hesitancies explain the relationship between trust in scientists and willingness to get vaccinated. The result suggests that only individual hesitancy with the U.S. sample significantly mediated between trust in scientists and willingness to get vaccinated. In other words, trust in scientists could reduce individual's concern regarding their previous bad experience, or their own value of vaccine, and therefore increase the possibility to accept the vaccine. However, this mediation effect only holds with individual hesitancy, rather than contextual hesitancy and vaccine-specific hesitancy. These nonsignificant results might result from the expedited process of COVID-19 vaccine development [42], which causes concerns regarding vaccine effectiveness, as well as the long-term misinformation about the vaccine [43]. Our findings suggest that in the U.S., trust in scientists could only change the personal perception of the vaccine but might not affect the hesitancies derived from interpersonal contacts and about specific vaccines. As for the nonsignificant mediation result in China, it is possible that other influencing factors overweighed the effects of vaccine hesitancy and trust in scientists such as politically related variables.

4.1. Practice implication

This study provides valuable implications for health caregivers and policy decision-makers. Although our findings showed that acceptance of vaccination requires a climate of trust between science and society, this trust relationship would vary from country to country. The insignificant results in the Chinese sample did not reject the importance of trust in scientists to health promotion because trust in scientists was still negatively related to individual hesitancy of Chinese individuals. Thus, this study underscores the importance of the specific belief in trust in understanding individuals' motivation behind their vaccine decisions. Therefore, scientists may want to conduct more active interactions with the public and focus on nurturing their relationship with laypersons in addition to offering them scientific knowledge. Strategies like more personal language with followers and community maintenance may be useful to enhance the public's trust in scientists and achieve a mutually beneficial relationship, which would help with vaccine uptake, especially from a long-term perspective.

4.2. Limitation and future studies

Although this study provides unique insights into our understanding of vaccine willingness between countries with very different cultures and regulation backgrounds, it is not without limitations. First, although this study yielded many significant results, the sample size is relatively small and it may not generate sufficient statistical power to speak the significant paths preciously. Additionally, while we controlled several important covariates, more can be considered, such as individuals’ occupation (whether in health industry or not), political ideology, or attitude towards other vaccines. Future studies should further expand the consideration of influencing factors, with larger representative sample size. Third, although these two surveys were conducted almost simultaneously, the progress of the pandemic in these two counties was not comparable and at the same stage. Thus, replication might be needed at a similar stage such as the post-pandemic era.

5. Conclusion

The quick and collaborative development of COVID-19 vaccination is the most effective way to solve the current crisis. However, this effort can be hampered by vaccine hesitancy and low trust in scientists. Our study highlighted that trust in scientists is relatively high in both China and the U.S., and trust in scientists is negatively associated with all types of vaccine hesitancy, but this relationship is more observed with U.S. participants. The insignificant data in China suggests that existing models may not apply to the cultural, political, and social environment in China.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.