Who is more likely to hesitate to accept COVID-19 vaccine: a cross-sectional survey in China

ABSTRACT

Background

The aim of our study was to identify factors associated with coronavirus disease 2019 (COVID-19)vaccine willingness in China to aid future public health actions to improve vaccination.

Research Design and Methods

This study was conducted in August 2020 using a mixed-method approach, including a cross-sectional self-administered anonymous questionnaire survey and in-depth interviews with community residents in China.

Results

Of the participants, 30.9% showedCOVID-19 vaccine hesitancy. Being female(OR = 1.297), having poor health(OR = 1.312), having non-health or medical-related occupations (OR = 1.129), no COVID-19 infection experience(OR = 1.523), living with vulnerable family members(OR = 1.294), less knowledge(OR = 1.371), less attention to COVID-19 information(OR = 1.430), less trust in official media(OR = 1.336), less perceived susceptibility to COVID-19(OR = 1.367), and less protective behavior(OR = 1.195) were more likely to hesitate. Qualitative research has shown that they doubt the importance and necessity, as well as the effectiveness and safety of the vaccination. The economic and service accessibility of the vaccination was an impediment to their vaccine acceptance.

Conclusion

Nearly one-thirdof people showed hesitancy to accept COVID-19 vaccination in China. Our findings highlight that health communication and publicity should be performed for the targeted population, and immunization programs should be designed to remove underlying barriers to vaccine uptake.

KEYWORDS:

• COVID-19
• China
• vaccination
• vaccine hesitancy
• vaccine willingness
• vaccine acceptance

1. Introduction

The World Health Organization (WHO) declared the coronavirus disease 2019 (COVID-19) as a pandemic on 11 March 2020 [1], which has demonstrated a widespread and profound impact on health and the economy worldwide [2]. Until 14 June 2021, there have been 175,541,600 confirmed cases of COVID-19, including 3,798,361 deaths globally [3].

In the absence of SARS-CoV-2-specific medications, traditional public health strategies and strong social mobilization have benefited many countries, including China, to delay the rapidly increasing number of COVID-19 cases. However, these non-medication strategies render a high economic and social development cost and could be unrealistic if the pandemic continues for years [2]. It is well recognized that the research and development of COVID-19 vaccines, as well as the implementation of its vaccination, are vital in controlling the pandemic [4,5].The success of the COVID-19 vaccine depends not only on the development of the vaccine itself, but also on vaccination coverage. Hesitancy or resistance to COVID-19 vaccination might influence the results of ending or slowing down the pandemic, as expected.

In 2012, the Strategic Advisory Group of Experts (SAGE) on immunization defined ‘vaccine hesitancy’ as the delay in acceptance or refusal of vaccination despite the availability of services [6]. Although vaccination is widely recognized as one of the most successful and cost-benefit health interventions, a growing number of people have suspected the necessity, effectiveness, and safety of vaccinations [7–9]. As a result, the increasing transmission of vaccine-preventable diseases(VPD) has caused community outbreaks and increased morbidity in some countries and areas [6].According to a survey of 194 WHO member countries from 2014 to 2016, vaccine hesitancy has become a problem in at least 90% of countries [10]. Due to the severity of vaccine hesitancy, the WHO has recognized it as one of the ‘ten great threats to global health in 2019 [11].’Recently, cross-sectional analysis has been used to determine the factors associated with vaccine hesitancy, including knowledge, attitudes, practice, and beliefs [12,13]. In the context of new media, the factors associated with vaccine hesitancy have become more complicated. Determinants of vaccine hesitancy may greatly influence the impact of public health interventions [14], and thus impede the control and prevention of infectious diseases.

Given the growing number of vaccine hesitancy toward available vaccines developed for years in many countries, global COVID-19 vaccination might face greater challenges due to the uncertainty of its efficiency and safety. It is crucial to know people’s attitudes toward the newly developed COVID-19 vaccine and the characteristics of the hesitant population. Hesitancy to accept COVID-19 vaccine by a certain number of people, even a relatively small number, may cause a disproportionate effect on herd immunity and uncontrollable situation of the continuous COVID-19 pandemic.

Several studies have already shown that nearly 20%–30% of the population hesitated to accept the COVID-19 vaccine [9,15–17]. The level is higher in Japan and England at 34.5% [18] and 36% [19], respectively. With the emerging evidence of hesitancy of COVID-19 vaccine globally, the acceptance of COVID-19 vaccination should be a key element for increasing vaccination coverage [20]. A previous study found that age, sex [21], perceived severity of COVID-19 disease, perceived personal vulnerability [22,23], health beliefs, conspiracy theories, general worries for the safety and effectiveness [24,25] of the future vaccine, individual health engagement [26],and psychological characteristics, such as lower levels of trust in scientists and health care staff, lower levels of altruism, higher level of social dominance, authoritarianism, and higher levels of internal locus of control [9], were the factors associated with hesitancy to COVID-19 vaccine acceptance. However, limited studies have reported COVID-19 vaccine willingness in a large Chinese population or elucidated on the underlying reasons for hesitancy.

The objective of this study was to determine the level of vaccine willingness of Chinese residents, especially hesitancy to accept the vaccine, and to identify the key determinants of hesitancy. The results could provide suggestions for increasingCOVID-19 vaccination coverage and improvingSars-Cov-2 herd immunization.

Health behavior theories, which revealed multiple factors influencing human health-related behaviors, have provided references for our research. The KAP model predicted that people would adopt a certain behavior based on three stages: knowledge acquisition, belief generation, and behavior formation [27]. Appropriate knowledge is the primary foundation of people’s behavior. It can not only prompt people to take action, but also urge them to take a suitable one [28]. The protective action decision model (PADM) predicts people’s responses to environmental hazards and disasters [29]. It identifies the role of information processing and risk perception in human response to threat. Public knowledge acquisition, compliance, and refusal behaviors toward governmental advice are strongly influenced by information communication [28]. During the COVID-19 pandemic, especially during the lockdown period, social media has played an indispensable role in the spread of COVID-19 vaccination information, informing the public knowledge and awareness of COVID-19 vaccination [30–32]. In the context of new media, the transmission of information, if not accompanied by enough trust of official media from the large population, may lead to misunderstanding or mistrust toward vaccination. Risk perception is widely accepted as the cause of behavioral changes [33]. New and unknown COVID-19 is believed to be one of the central factors elevating the level of risk perception [34]. Some researchers have found that biased risk perception might prevent people from making rational behavioral decisions [35,36].

Non-medication preventive behaviors, such as avoiding crowds, wearing face masks, maintaining good indoor airflow, following official guidelines, and hand hygiene, are habitual and difficult to change. Given the deeper understanding of COVID-19 vaccine knowledge, broader information access and trust, non-medicine protective behaviors, and vaccine willingness may mutually influence each other. Therefore, in this study, the relationship between them has also been discussed.

2. Patients and methods

2.1. Study setting

The mixed method was used to analyze the influencing factors and reasons for hesitancy to accept the COVID-19 vaccine.

A cross-sectional questionnaire survey was conducted to analyze factors influencing hesitancy. Qualitative analysis was used to explore the underlying reasons for hesitancy to accept the COVID-19 vaccine, which was not suitable to be represented in the questionnaire.

2.2. Study population and data collection

2.2.1. Quantitative research

We conducted a pilot study and collected 70 questionnaires using a convenient sampling method to improve our design and questionnaire quality. A nationwide, cross-sectional quantitative study was conducted online from 4^th to 13^thAugust 2020 using a stratified cluster sampling method via the Wenjuanxing platform, a widely accepted online questionnaire survey platform in China. Residents aged 18 years and above residing in mainland China were eligible to participate in the survey anonymously. Written informed consent was provided in the introduction part of the questionnaire before their answers.

According to the IP address of the questionnaire records, each participant could only answer once; if a questionnaire was completed in more than 8 min, which is the shortest time to complete the questionnaire by our team testing, and two logic questions were answered logically, it would be judged as valid and included for analysis; otherwise, it was removed.

2.2.2. Qualitative research

A semi-structured outline was used to learn participants’ willingness to vaccinate and the underlying reasons for the impediments to their vaccine uptake. Key residents, selected by purposive sampling, were interviewed through telephone by our research team in September 2020 for more than 20 min. Persons aged 22 to 78 years, with an educational background from primary school to post-graduation, were included.

2.3. Measurements

2.3.1. Dependent variable

Participants were asked to answer the question, ‘If there is an available COVID-19 vaccine, will you vaccinate?’ Participants will select the answer, ‘yes,’ ‘no’ or ‘it depends.’ In multivariate logistic regression analysis, the dependent variable (hesitancy to accept vaccine) was divided into two categories, including 0 indicating acceptance of vaccine(selecting ‘yes’) and 1 indicating hesitancy(selecting ‘no’ or ‘it depends’).

2.3.2. Independent variable

Socio-demographic characteristics: The socio-demographic characteristics of the participants included sex, age, education, marital status, residency(urban or rural areas), monthly income, self-reported health, health and medical-related occupation or not, living with vulnerable family members(the elderly over 65 years old or children below 5 years old), and COVID-19 infection experience. The risk level of the sample provinces was dichotomized into high and low levels, according to the number of confirmed disease cases at that time.

Knowledge: Four parts of knowledge questions were designed according to the official guidelines, including general knowledge about COVID-19, preventive and control countermeasures, transmission routes, and effective inactivation of Sars-Cov-2. Answers to all questions can be found on official websites [37,38]. A correct answer was assigned a score of 1, otherwise, 0. The average score for the knowledge component was calculated. A score above 3 was considered a high-level group, otherwise a low-level group.

Information: In the survey, the participants were asked to rate their attention on COVID-19 information and trust in official media on a one-item 5-point Likert scale(ranging from 1, complete indifference, to 5, deep concern, and from 1, completely distrust, to 5, completely trust). A score above 3 was considered deemed high-level group, otherwise a low-level group.

Risk perception: We used the risk perception scale developed by our team published in 2020 [39], which has shown good reliability(Cronbach’s α = 0.824) and validity(GFI = 0.982, AGFI = 0.961, IFI = 0.972, RMSEA = 0.062).Participants were asked to make a judgment on the 9-item three component of risk severity, susceptibility, and controllability of the COVID-19 pandemic on a 6-point Likert scale (ranging from 1, complete non-conformity, to 6, complete conformity). A score above 3 was considered a high-level group, otherwise, a low-level group.

Non-medication protection behaviors: Study participants were asked to rate their behaviors, such as avoiding crowds, wearing face masks, keeping good indoor airflow, and practicing good hand hygiene, etc., on a five-point Likert scale ranging from 1‘never’ to 5 ‘always.’The average score was then calculated. A score above 3 was considered deemed high-level group, otherwise a low-level group.

2.4. Statistical analysis

2.4.1. Quantitative analysis

To estimate whether the sample could represent the whole population, we compared the sample with the latest available detailed age and sex proportion data from the China Statistical Yearbook 2019.Chi-square test results showed that the sample age percentage was different from that of the general population in China(χ² = 8608.995, P< 0.05). Weighting based on age parameters was then adopted to adjust the sample population [40].

Descriptive statistical analysis was used to describe the characteristics of the participants and the general situation of their vaccine willingness. The chi-square test was used to determine the significance of the association between COVID-19 vaccine willingness and independent variables of different categories and levels. Multivariate logistic regression was constructed to determine the factors influencing vaccine willingness, while adjusting for the effects of other independent variables. Variables in relation to socio-demographic characteristics, knowledge level, information attention, trust in official media, risk perception, and protective behaviors were entered into the multivariate logistic regression model using an enter approach. Meanwhile, the OR, P value, and 95% CI were calculated. All statistical analyses were conducted using SPSS25.0, and two-sided statistical significance was set at P< 0.05.

2.4.2. Qualitative analysis

All interviews were digitally recorded, transcribed verbatim, and thematically coded immediately after each interview. Data from the interviews were analyzed thematically. Codes were derived initially from the data, and then the theoretical code was used. A set of themes was obtained based on initial codes. Thirty key interviewees were determined by the saturation of information until no new sub-themes emerged.

3. Results

3.1. Participant characteristics

In general, 19, 647 persons completed the questionnaire, and 19, 132 valid questionnaires were included in the study. The effective rate was 97.38%.

Among the participants (Table 1), 52.4% were female, 50.0% were aged 26–45 years, 51.4% were married, and 56.9% received junior college and university education degrees. In terms of location, 56.5% lived in the city and 52.5% lived in high-risk areas. Regarding health conditions, 83.4% reported themselves in good health conditions, and 69.6% had vulnerable people in their families. In terms of economic situation, 59.2% of respondents had middle and high income.

Table 1. Characteristics of respondents and their hesitancy to accept vaccine (n = 19,132)

Characteristics	Respondents (n [%])	Hesitation (n [%])	χ^2weighted	Pweighted values
Sex			56.448	0.000
Female	10,021(52.4%)	3200(31.9%)
Male	9111(47.6%)	2576(28.3%)
Age			1171.039	0.000
46 and above	1562(8.3%)	558 (35.7%)
26–45	9383(50.0%)	2081(22.2%)
18–25	7808(41.6%)	2932(37.6%)
Education			448.721	0.000
Master and above	1255(6.6%)	515(41.0%)
Junior college or bachelor	10,892(56.9%)	2867(26.3%)
High school and below	6985(36.5%)	2394(34.3%)
Marital status			1156.789	0.000
Married	9838(51.4%)	2266(23.0%)
Others(single, divorced or widowed)	9294(48.6%)	3510(37.8%)
Residency			182.015	0.000
Urban	10,818(56.5%)	3018(27.9%)
Rural	8314(43.5%)	2758(33.2%)
Monthly income			1389.168	0.000
Higher	1253(6.5%)	307(24.5%)
High	6602(34.5%)	1422(21.5%)
Middle	4723(24.7%)	1383(29.3%)
Low	6554(34.3%)	2664(40.6%)
Risk level of sample provinces			69.292	0.000
High-risk	10,041(52.5%)	3251(32.4%)
Low-risk	9091(47.5%)	2525(27.8%)
Self-reported health			341.822	0.000
Good	15,962(83.4%)	4517(28.3%)
Poor	3170(16.6%)	1259(39.7%)
Living with vulnerable family members			317.374	0.000
No	5809(30.4%)	2130(36.7%)
Yes	13,323(69.6%)	3646(27.4%)
Health or medical-related staff			219.947	0.000
No	16,887(88.3%)	5293(31.3%)
Yes	2245(11.7%)	483(21.5%)
COVID-19 infection experience			202.331	0.000
No	17,356(90.7%)	5407(31.2%)
Yes	1776(9.3%)	369(20.8%)
Knowledge			343.003	0.000
High	15,312(80.0%)	4300(28.1%)
Low	3820(20.0%)	1476(38.6%)
Information attention			267.176	0.000
High	14,998(78.4%)	5291(29.5%)
Low	4134(26.1%)	1685(40.8%)
Trust in official media			277.589	0.000
High	15,230(79.6%)	4288(28.2%)
Low	3902(20.4%)	1488 (38.1%)
Perceived severity			0.094	0.760
High	14,973 (78.3%)	4503(30.1%)
Low	4159 (27.1%)	1273(30.6%)
Perceived susceptibility			82.735	0.000
High	3698(19.3%)	968(26.2%)
Low	15,434(80.7%)	4808(31.2%)
Perceived controllability			62.768	0.000
High	15,517(81.1%)	4091(27.3%)
Low	3615(18.9%)	954(26.4%)
Non-medication protection behaviors			488.317	0.000
Good	15,003(78.4%)	4158(27.7%)
Bad	4129(21.6%)	1618(39.2%)

Of the participants,80.0% hada high level of knowledge regarding COVID-19. Of the respondents,78.4% paid attention to COVID-19 information, and 79.6% trusted in official media. A total of 78.3%, 19.3%, and 81.1% participants perceived higher severity, susceptibility, and controllability of COVID-19, respectively. Moreover, 78.4% performed higher level of protective behaviors.

3.2. Factors associated with hesitancy to accept vaccine

Research showed that 30.9% of the participants indicated that they would refuse or delay COVID-19 vaccination despite its availability.

The hesitancy to accept the COVID-19 vaccine and its associated factors are presented in (Table 2). Women with a master ’s or a postgraduate degree or high school and below education degree, with other marital status(single, divorced, or widowed), living in urban or in low-risk provinces, with poor self-reported health, with non-health-related occupation, not living with vulnerable family members, without COVID-19 infection experience were more likely to hesitate to accept COVID-19 vaccination. Low levels of knowledge, information attention, trust in official media, perceived susceptibility, and non-medication protection behavior were associated with a high level of hesitancy to accept the vaccine. High perceived controllability also showed a high level of hesitancy in the COVID-19 vaccine.

Table 2. Factors associated with hesitancy to accept COVID-19 vaccine in multivariate logistic regression

Variables	OR(95% CI)	ORweighted(95% CI)
Sex
Male(reference)
Female	1.189(1.111, 1.273)	1.297(1.213, 1.387)
Age
>45(reference)
26–45	0.599(0.530, 0.676)	0.614 (0.565, 0.667)
18–25	0.805(0.693, 0.936)	0.911(0.803,1.033)
Education
Master and above(reference)
Junior college or bachelor	0.578(0.508, 0.658)	0.578 (0.507, 0.659)
High school and below	0.641(0.559, 0.736)	0.562(0.488, 0.647)
Marriage
Others(single, divorced or widowed) (reference)
Married	0.764(0.689, 0.847)	0.800(0.718, 0.891)
Residency
Urban(reference)
Rural	0.979(0.911, 1.051)	0.912(0.850,0.978)
Monthly income
Higher (reference)
High	0.866(0.747,1.005)	0.931(0.805,1.077)
Middle	1.085(0.929,1.268)	1.130(0.971,1.314)
Low	1.283(1.090, 1.509)	1.194(1.017, 1.403)
Risk level of provinces
High-risk(reference)
Low-risk	0.796(0.745, 0.851)	0.760(0.711, 0.812)
Self-reported health
Good (reference)
Poor	1.382(1.268, 1.506)	1.312(1.211, 1.421)
Health or medical-related staff
Yes(reference)
No	1.129(1.007, 1.266)	1.129(1.010, 1.262)
Living with vulnerable family members
Yes(reference)
No	1.206(1.123, 1.296)	1.294(1.209, 1.384)
COVID-19 infection experience
Yes(reference)
No	1.499(1.320, 1.703)	1.523(1.336, 1.736)
Knowledge
High(reference)
Low	1.246(1.147, 1.354)	1.371(1.260, 1.492)
Information attention
High(reference)
Low	1.480(1.368,1.601)	1.430(1.322, 1.546)
Trust in official media
High(reference)
Low	1.264(1.164, 1.372)	1.336(1.234, 1.46)
Perceived susceptibility
High(reference)
Low	1.281(1.173, 1.399)	1.367(1.255, 1.489)
Perceived controllability
High(reference)
Low	0.845(0.774, 0.922)	0.805(0.741, 0.875)
Non-medication protection behavior
High(reference)
Low	1.293(1.192, 1.402)	1.195(1.100, 1.297)

3.3. Reasons for hesitancy to accept vaccine

Three main themes were derived from the qualitative analysis: self-conceit, confidence, and accessibility. Some interviewees agreed that the necessity and importance of COVID-19 vaccination would influence their attitudes toward COVID-19 vaccination. The confidence of the vaccine effectiveness and safety and trust in COVID-19 vaccination decision makers, to some extent, influenced their hesitancy to accept the COVID-19 vaccine. Economic and information accessibility were also the main reasons for their hesitancy. The interviewees’ main opinions are presented in (Table 3).

Table 3. Key concepts emerging from qualitative analysis of the interviews

Category	Concept	Representative quote(s)
Self-conceit	Necessity Importance	I am in good physical health situation, so I neednot vaccinate I do not think I am susceptible to COVID-19, maintaining precaution is enough. China has gained great achievements in pandemic prevention and control, so it is unnecessary to vaccinate
Confidence	Effectiveness	Because it is a new product, I would like to know if this is effective(5). I am worried if I will still be infected with COVID-19 after vaccination(4). I am concerned about the duration time of preventive function after vaccination.
	Safety	I will consider vaccination. But I will see if there is an adverse reaction and side effects(8). I will have influenza vaccination, because it has been developed for a long time. So I believe it is safe. I am prone to allergic reactions, so I will not vaccinate. I am worried that the adverse reaction will decrease my immunity.
	Trust in COVID-19 vaccination decision makers	If the government declares that it is necessary to vaccinate, I will vaccinate. I will vaccinate when government says a large scale of vaccination is necessary.
Accessibility	Economic accessibility	I hope that the price will be rational, and it willnot be too expensive(4). I am concerned about the price and supply amount, because I am afraid that I could not afford it. I want to vaccinate immediately, but I am afraid that I could not secure a vaccination appointment.
	Information accessibility	I have not heard of influenza vaccine. Is COVID-19 vaccine well developed, and has it appeared on the market?

4. Discussion

In the hope of mass vaccination, it is urgent to know the current situation of vaccine willingness among different populations and their characteristics, so as to provide more specific public health messages to increase the coverage of COVID-19 vaccination. In general, about one-third of Chinese residents hesitated about COVID-19 vaccination. The willingness was mainly associated with knowledge, information attention and trust in official media, perceived susceptibility and controllability, and preventive behaviors. The underlying reasons for hesitancy to accept the COVID-19 vaccine were self-conceit, confidence, and accessibility.

4.1. Socio-demographic characteristics

The high hesitancy level of females toward vaccination among the Chinese population reflected that, compared with males, females were more likely to demonstrateCOVID-19 vaccine hesitancy. This was similar to other studies conducted in the UK and Turkey [22,41,42]. According to social role theory, females tend to play a more caring role than males [43], which makes them more sensitive to consideration of the impact of vaccination on the elderly and children in the family. This result is similar to the results of Wang [44], Gabriella Di Giuseppe [45], and Malik Sallam [8].

As reported in the study, a U-type characteristic of hesitancy was found in educational attainment. People with lower or higher levels of education (high school and below, or master and above) showed a stronger hesitant attitude toward vaccination compared with people with middle-level education(junior college or bachelor’s degree). One of the interesting findings of the study was that people with higher education tended to be more hesitant, which is inconsistent with other studies [8,46]. It appears that people with higher education commonly have a better understanding of the complex biological mechanism for vaccine development, which might view the newly developed vaccine dialectically. Moreover, due to the well-controlled situation of the COVID-19 pandemic in China, the necessity of vaccination is not urgent. Well-educated people tend to adopt non-medication protection behaviors instead of bearing the risk of vaccination. A less educated population might face more difficulties in understanding complex vaccine knowledge, which might cause more hesitancy. This is consistent with other studies [8,46].

This study found that people with poor health status had a higher level of hesitancy to accept the vaccine. These people would worry more about adverse reactions. These concerns were quite understandable, since some vaccines were reported to beunsuitable for some populations due to relatively poor immunity function or allergic predisposition [47].

In contrast to other studies, suggesting that doctors and nurses had vaccine hesitancy [22,44,48,49], our study found that medical and health-related staff were more likely to accept the COVID-19 vaccine. This might be related to population differences. Medical and health-related personnel in this study included public health-related professional staff, who are more exposed to Sars-Cov-2 and are more likely to submit to preventive health measures such as vaccination.

4.2. Factor of knowledge

Consistent with a previous study [50,51], we found that people who had a lower level of COVID-19 knowledge had a higher level of hesitancy to accept the vaccine. People who had a lower level of COVID-19 knowledge might know less about the severity [51], transmission path, preventive methods, or the effect of the COVID-19 vaccine on disease prevention. However, compared with people with lower levels of knowledge, adverse events would not mislead people with higher levels of knowledge, and the hesitancy to accept the vaccine was less [50].

4.3. Factors of information

People who had a lower level of information attention and lower trust in official media had a higher level of hesitancy to accept the vaccine. Reluctance induced by new media and the Internet may result in false information and doubt the scientific effectiveness and safety of the COVID-19 vaccine. Bestch [52] found that surfing on the anti-vaccine website for 5–10 min would negatively affect vaccination decision-making. With the rapid information dissemination thru social media, people can grasp information faster. However, the identification of legitimate information is challenging. If the public loses their trust in the official media, they would be misled by unverified information and generate improper risk perception, even developing improper protective behaviors, which might cause social and economic sequelae.

In addition, healthcare workers’ recommendations as a source of information were associated with patients’ vaccine willingness [18,53–57]. Healthcare workers’ functions can be fully displayed to influence residents’ willingness to vaccinate.

4.4. Factors of risk perception and non-medication protection behaviors

In previous studies of risk perception, researchers found a special phenomenon of regional difference, that is, the public in the risk event center had a lower risk perception than the public in the surrounding areas. For example, some researchers found that people in the high-risk provinces had a lower anxiety level than those in the periphery or low-risk ones, which was named as ‘psychological typhoon eye’ phenomenon [58,59]. In fact, the people in ‘psychological typhoon eye’ regions face an increased risk of infection and need more protection, such as vaccination. Therefore, policies and countermeasures should be implemented to promote vaccination in these regions.

Risk perception includes the perceived possibility, severity, and controllability of risks. However, we found that people with lower perceived susceptibility and higher controllability would have a higher level of hesitancy in vaccine acceptance. Perceived severity had no impact on vaccination willingness, which is different from Bacon’s research [60]. This might be because, in China, the COVID-19 pandemic in August 2020 showed a decreased spread due to effective control measures from the government; thus, residents experienced a lower level of perceived severity. After assessing and balancing the cost and benefit of the behaviors toward the hazard, people would then decide on vaccination [61]. A previous study found that a higher risk perception level would stimulate people to adopt protective countermeasures due to the higher pressure of risk [62]. To some extent, vaccination is a protective behavior of the public. People who had a lower level of non-medication protection behaviors would be more hesitant.

4.5. Reasons for hesitancy to accept vaccine

People’s self-conceit, confidence, and accessibility to COVID-19 vaccination were major barriers to accepting the vaccine.

Interviewees denoted hesitancy due to their good physical health status. Leask found that parents whose children were deemed healthy believed that it was unnecessary to vaccinate [63]. Under the protection of vaccination, people may falsely assume that the infection and severity of the disease are not too serious. Paradoxically, the success of immunization programs may result in complacency [52]. For example, immunization has successfully reduced the morbidity of VPD, which might decrease the risk perception of the vaccine function on VPD, and increase their hesitation about vaccination [64].

The efficiency and safety of vaccination are associated with vaccination confidence [45].These two barriers for vaccine willingness have been verified in many previous studies [40,65,66]. Of the participants, 27.3% in Bosnia and Herzegovina, 20.1% in Russia, and 18.7% in Italy expressed their doubt on the efficiency of the vaccination; 45.2% in France and 38.3% in Bosnia and Herzegovina suspected the safety of the vaccination [40]. The investigation conducted by WHO and United Nations Children’s Fund in 194 countries and areas from 2014 to 2016 found that 22%–23% of people are worried about the safety and adverse reactions of vaccination [65]. It is note worthythat the available COVID-19 vaccine was still in development during the investigation. Therefore, the results on the safety and efficiency of the COVID-19 vaccine in China need to be proven and updated in the future, which might impact the acceptance of the COVID-19 vaccine. Therefore, based on the hypothesized safety and efficiency of the COVID-19 vaccine, rapid, transparent, and effective publicity of the research or clinical results or targeted educational interventions [18,44,45,67,68] should be updated along with the development of science and technology. Moreover, providers’ recommendations and social environmental factors, including public health policies [8,69], were the key determinants of vaccination behaviors. Strong government recommendations would promote COVID-19 vaccination, which is supported by the majority of the population, including healthcare workers, in some settings [40,70]. Thus, in order to bring the pandemic to the end, if publicity cannot achieve the goal of adequate vaccination, mandatory vaccination may take effect [71].

The high price, demand exceeding supply, and appointment accessibility were the interviewees’ focal points. Price and affordability of the vaccine are the key points of affordability [72]. History has witnessed unaffordable prices set by the big pharmaceutical companies, such as the high price of hepatitis drugs [73]. Limited by the development and production capacity, price and affordability will become a concern, especially in low-income countries [74]. Therefore, the COVID-19 vaccination price might influence vaccination willingness in some countries and areas, even in the pandemic prevention worldwide. The Chinese government has promised to provide COVID-19 vaccination free for residents [74], and the economic burden will not be an obstructive factor. Although COVID-19 Vaccines Global Access has promised to help low-and middle-income countries to promote vaccination projects [75], great challenges still exist [76,77].

4.6. Strengths and limitations

We adopted a mixed method to deeply explore the potential influencing factors of hesitancy to accept COVID-19 vaccine among Chinese residents from a more comprehensive perspective.

There are several limitations to this study. First, the cross-sectional study could not elucidate on causal relationships. Second, COVID-19 vaccine willingness may change with the progress of vaccine research. Nonetheless, people who are hesitant to accept COVID-19 vaccination now may be more likely to become the vaccine hesitancy population when it is available. Third, the research was conducted only in the Chinese population. However, there is still reference for other countries since the phenomenon of hesitancy to accept the COVID-19 vaccine exists globally. Fourth, the online response bias cannot be completely avoided, although we enlarged the sample and conducted analysis by weighting the age parameter. Fifth, there might be recall and social desirability biases in this research.

5. Conclusion

About one-third of people were hesitant to accept COVID-19 vaccination in China. Hesitancy was associated with perceived COVID-19 risk, information, and non-medication protective behaviors. People’s hesitancy were attributed to self-conceit, confidence, and accessibility. Our findings highlighted that health communication and publicity should be performed for the targeted population, and immunization programs should be designed to remove underlying barriers to vaccine uptake.

Ethical approval

Ethics approval for the study protocol was obtained from the Ethics Committee of Harbin Medical University(HMUIRB20200004). Informed consent was obtained from all participants through online responses before the start of the survey.

Author contributions

YHao and QW took total responsibility for the study design, conceptualization, methodology, coordination of the survey, development of the analysis framework and the original draft. PZ, KF and HG took part in the design of the survey, conducted the research, data analysis and the original draft. KZ, LL, MJ participated in the design of the survey, and proposed suggestions for the original draft. QZ and XB participated in the literature review and data collection. YHuang participated in analyzing the results and revising the original draft. PZ, QZ and HG contributed equally.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Data availability

The summary datasets generated and analyzed during the current study are not publicly available because the datasets are currently used for another project, but are available from the corresponding author on reasonable request under the data transfer agreement.

Additional information

Funding

This study was funded by the National Natural Scientific Fund of China (71673072, 72042001). Funding body had no influence on the study design, data collection, data analysis, data interpretation, or writing of the manuscript.