Factors influencing COVID-19 vaccination intention among overseas and domestic Chinese university students: a cross-sectional survey

ABSTRACT

To vaccinate the Chinese on a nationwide scale timely and effectively, it is necessary to assess the vaccination uptake intention of the public. University students are opinion leaders, who have an important impact on the vaccination uptake intention of others around them. As a group with strong population mobility, overseas university students have an extra influence on the spread of COVID-19 and the prevention and control of the pandemic. Thus, it is necessary to investigate the vaccination uptake intention of overseas and domestic university students to promote vaccination and control the pandemic globally. However, little is known about the COVID-19 vaccination uptake intention among overseas and domestic university students. This study aimed to explore the difference between overseas and domestic Chinese university students’ COVID-19 vaccination uptake intentions and influencing factors using the Health Belief Model. A cross-sectional survey using an online questionnaire was conducted among 370 overseas university students and 463 domestic university students between January and February 2021. More than half of the respondents (536, 64.3%) reported vaccination uptake intentions, with overseas and domestic university students reporting similar vaccination uptake intentions (64.1% vs 64.6%, p > .05). Perceived benefits, perceived barriers, and cues to action were important factors that influenced the vaccination intention among overseas and domestic university students. It is worth trying to communicate the benefits of the vaccine, enhance the role of cues to action, and eliminate the potential barriers among overseas and domestic university students through creative propagation to further promote the COVID-19 vaccination.

KEYWORDS:

• COVID-19 vaccine
• overseas university students
• domestic university students
• vaccination intention
• Health Belief Model

Introduction

Since the first case was reported in December 2019, coronavirus disease 2019 (COVID-19) has spread rapidly around the world, causing an unprecedented global public and economic crisis.^1,2 On March 11, 2020, the World Health Organization (WHO) declared COVID-19 as a worldwide pandemic.³ Although great efforts have been paid to combat the pandemic by governments all over the world, grave losses continue. Considering the infectiousness and virulence of COVID-19, an effective countermeasure that is suitable for different countries and areas is of great urgency.

Recognized as one of the most powerful and cost-effective weapons against infectious diseases,^4,5 a safe and effective vaccine is expected to eventually prevent and control COVID-19. Previous studies have confirmed vaccination intentions to be the foremost issue affecting the success of a nationwide vaccination promotion.^6,7 The COVID-19 vaccine developed in China was officially approved on December 31, 2020, and the vaccination of high-risk groups was initiated since that time.⁸ Since March 2021, the supply of COVID-19 vaccines in China has been sufficient,⁹ and vaccination stations have been set up in primary health care institutions to ensure that all residents have access to the COVID-19 vaccine.¹⁰ China advocates universal vaccination of COVID-19 vaccines and has launched a massive vaccination rollout.¹¹ Thus, to vaccinate the Chinese on a nationwide scale timely and effectively, it is necessary to assess the vaccination intention of the public.

With the spread of the COVID-19, many cities worldwide have implemented preventive measures including quarantine and lockdown to protect the local residents and reduce the risk of infection. University students, as a part of these groups, due to their academic and living needs, have high mobility and active and frequent social activities,¹² which puts them at a higher risk of infection and may increase the risk of disease transmission. The vaccination intention of university students should be greatly valued because they are important opinion leaders with great influence on social development trends.¹³ On October 20, 2020, the Chinese government announced that high-risk groups, including those who must visit countries with high confirmed cases due to work or study reasons, should be given priority for vaccination.¹⁴ China is a major exporter of overseas university students,¹⁵ and in the long run, a high proportion of overseas Chinese university students will return to China after completing their studies abroad.¹⁶ The vaccination situation of overseas and domestic Chinese students will affect the prevention and control of the pandemic in China, and the vaccination situation of overseas university students who are still studying abroad may also have an impact on the control of the pandemic in the country where they are located. Furthermore, previous studies confirmed that European and American countries had long been hesitant about vaccines and many people held strong hesitant attitudes toward vaccines,^17,18 which may abet overseas Chinese university students’ vaccine hesitancy. This hesitancy will affect the vaccination of overseas university students and may also make them different from domestic university students in vaccination intentions. Although there have been several studies on the COVID-19 vaccination intention, most of them centered on healthcare workers,^19,20 working-age population,²¹ and university students,^22–24 with little attention to university students currently studying abroad. Hence, exploring the status and possible factors of the vaccination intention among university students, both overseas and domestic students, should help accelerate the promotion of COVID−19 vaccines among the general public in China.

The Health Belief Model (HBM) is one of the most commonly used and ideal theoretical models for studying health behaviors.^25–27 It can be used to guide health promotion and disease prevention and to explain individual health behaviors.²⁸ The HBM holds that people will achieve optimal behavior change if they manage to target perceived susceptibility, severity, benefits, barriers, self-efficacy, and cues to action. The model predicts that health behaviors are affected by weighing the threats (susceptibility and severity), their ability to reduce the threats (benefits and barriers), and cues to action.²⁹ Recognized as an important predictor of vaccination, the HBM has been used in previous studies.³⁰ HBM provides a framework to show the relationships between these five variables and intentions to perform recommended health behaviors.³¹ In a study of the general population in Hong Kong, vaccination intention was associated with perceived susceptibility and cues to action.³² Walker and his fellows found in a sample of international college students studying in China that the promotion of benefits and the elimination of barriers were significantly associated with vaccination uptake intention.³³ Lin and colleagues also found perceived severity associated with vaccine acceptance.³⁴ Therefore, the present study employed the HBM to explore influencing factors of overseas and domestic Chinese university students’ COVID−19 vaccination intentions.

Based on the aforementioned review of the literature and theoretical framework, the present study aimed to describe the COVID-19 vaccination intention of overseas and domestic Chinese university students; and compare the COVID-19 vaccination intention of these two populations based on the HBM. For the logistic regression, we hypothesized that there would be differences in the vaccination intention of overseas and domestic university students. Besides, we hypothesized that perceived susceptibility, perceived severity, perceived benefit, and cues to action would positively influence the vaccination intention for both overseas and domestic university students. Whereas, perceived barriers would negatively influence their vaccination intention. Furthermore, for the path analysis, we hypothesized that different groups of populations might directly or indirectly affect the COVID-19 vaccination intention by the scores of each dimension of the HBM.

This study is expected to provide policy evidence and a basis for informing multi-pronged strategies to improve vaccine promotion in China. Meanwhile, it is hopeful that this study can shed light on factors hindering university students’ vaccination to improve public health and pandemic control globally.

Methods

Data collection

This cross-sectional study was conducted from January to February 2021 among overseas and domestic Chinese university students using snowball sampling. The questionnaire was presented via an electronic questionnaire tool named Wen Juan Xing. For overseas university students, based on the ranking of countries with COVID−19 newly confirmed cases in January 2021, and the 26 high-risk countries of the pandemic announced by the Chinese Foreign Ministry on April 2, 2020,³⁵ we selected six high-risk countries with a large number of overseas Chinese students as sample countries for the present study. They were the United States, the United Kingdom, Germany, Canada, Spain, and Australia. Later, to adapt to the changing situation of the pandemic, we added two Asian countries based on the population size of overseas Chinese students¹⁵ and the local epidemic status quo. These two countries were Japan and Singapore. The final sampled countries were the above eight countries. Secondly, the anonymous questionnaire on Wen Juan Xing was disseminated through Sina Weibo. Sina Weibo, similar to Facebook and Twitter, is one of the most common social media platforms used by Chinese university students. It provides an online community where overseas Chinese university students from various countries gathered and can facilitate us to distribute the electronic questionnaire. The first four to five overseas university students in each sample country to complete the questionnaire were invited as original deliverers. These original deliverers were encouraged to share the questionnaire as widely as possible with their Chinese friends studying in the same country.

For domestic students, a similar methodology was employed. We firstly conveniently selected four comprehensive universities that cooperated closely with us as sample universities, namely, Zhejiang University, Hangzhou Normal University, Zhejiang University of Technology, Zhejiang Chinese Medical University. Secondly, the questionnaire was issued on the bulletin board system (BBS) and university students were invited to fill in. The first four to five university students who completed the questionnaire were selected as original deliverers. These original deliverers were encouraged to share the questionnaire as widely as possible with their classmates or roommates. Inclusion criteria were that participants were over 16 years of age.

The sample size of the present study was determined based on a pilot survey of 145 participants including 45 overseas university students and 100 domestic university students, with 72.5% of overseas students and 65.0% of domestic students reporting they were willing to be vaccinated. Given a power of 95% and a two-sided significance level of 0.05, this study required minimum sample size of 306 overseas students and 350 domestic students.

Measures

The questionnaire comprised of three parts: (1) sociodemographic characteristics of participants, including gender, age, degree currently pursuing, average monthly household income, and present location; (2) the COVID-19 vaccination intention which was measured via a five-point Likert scale from strongly agree to strongly disagree. The answers were dichotomized into “Yes” (strongly agree and agree) and “No” (strongly disagree, disagree, and neutral); (3) perceptions of COVID-19 and COVID-19 vaccination by the HBM using a five-point Likert scale from strongly agree to strongly disagree.

All the measurements related to HBM were adapted from previous studies to measure the COVID−19 vaccination,^34,36–38 and revised according to the COVID-19 pandemic situation in China. After the first draft of the questionnaire was completed, it was submitted to a group of healthcare experts for the validity test, and a pre-survey was conducted among 45 overseas university students and 100 domestic university students to further ensure the reliability and validity of the questionnaire. Perceived susceptibility was used to ask the likelihood of COVID-19 infection. Perceived severity was used to ask the severity of COVID-19 infection, including “fear of infection, gravely ill, die, and deterioration of the financial situation.” Perceived benefits were used to ask the participants how willing they would be to receive the vaccine if it protected them and their family or friends from COVID-19. Perceived barriers were used to ask whether the efficacy, safety, protection duration, side effects, and the mutated virus would lessen the participants’ chances of getting vaccinated. Cues to action were used to ask whether the recommendations from family members, friends, healthcare experts, government officials, and the media would influence the vaccination intention of university students.

Statistical analysis

Descriptive analyses were conducted to present frequencies and percentages. Germany and Spain were combined as European Union countries, and Japan and Singapore were combined as Asian countries. Chi-square tests and t-tests were performed to examine the differences in sociodemographic characteristics, items of the HBM, and the vaccination intention between overseas students and domestic students. Scores of perceived susceptibility, perceived severity, perceived benefits, perceived barriers, and cues to action of the HBM were generated by adding corresponding items in each dimension. The differences in these dimensions between overseas university students and domestic university students were addressed by t-tests. Binary logistic regressions were employed to identify possible factors associated with the COVID-19 vaccination intention among overseas and domestic students. To understand the relationships among groups, the HBM, and the COVID-19 vaccination intention, we adopted a path analysis. All statistical analyses were performed using the SPSS (IBM Corp. Released 2011. IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp.), and p < .05 was considered statistically significant.

Confirmatory factor analysis (CFA) was used to evaluate the reliability and construct validity of the questionnaire based on the HBM. Items with factor loading less than 0.5 were excluded. This model was tested through path analysis using Amos 21.0 (Arbuckle JL and SPSS Inc., Chicago, USA). Multiple indices were used to evaluate the model fit, including 1) the likelihood ratio test statistic (χ²); 2) Comparative fit Index (CFI); 3) Tucker-Lewis index (TLI); and 4) the Root Mean Square Error of Approximation (RMSEA).

CFA of the HBM model

To evaluate the internal accuracy of survey items, the Cronbach’s alpha was calculated. A value of 0.7 or above was widely considered desirable.³⁹

The CFA was used to ensure an adequate model fitness level and structural validity and reliability. A good model fit was defined as: χ²/df < 2, CFI > 0.90, and RMSEA < 0.06, and an acceptable model fit was defined as: χ²/df < 3, CFI = 0.80–0.89, and RMSEA< 0.10.⁴⁰ The fit indices for the model were χ2 /df = 2.917, CFI = 0.961, and RMSEA = 0.048, indicating an acceptable model fit for the data.

Ethics approval

This study was approved by the Ethics Committee of the School of Public Health at Zhejiang University (Approval code: ZGL202101-6). Participants were well informed that completing the questionnaire signified their informed consent.

Results

Sociodemographic characteristics

A total of 470 overseas university students and 536 domestic university students were identified as potential participants. Data from 382 overseas university students and 463 domestic university students were available for analysis (response rate 81.3% and 86.4% respectively). Twelve overseas university students (2.6%) were excluded due to being under the age of 16.

As shown in Table 1, about two-thirds of university students in the present study (62.3%) were female, with an age of 23.6 (±3.3) years. More than half of them (53.9%) were pursuing a bachelor’s degree. The average monthly household income of most university students (39.3%) was less than 10,000 RMB (Renminbi). There were significant differences between overseas and domestic university students in age, degree currently pursuing, and average monthly household income. Compared with domestic university students, overseas students reported an older age, a higher proportion of currently pursuing master’s or doctorate degree, and a higher average monthly household income (p < .05).

Table 1. Sociodemographic characteristics of the study population

Sociodemographic characteristics	Total (N = 833) N (%)	Overseas students (N = 370) N (%)	Domestic students (N = 463) N (%)	^χCitation2/t	p
Gender				0.005	0.946
Male	314(37.7)	139(37.6)	175(37.8)
Female	519(62.3)	231(62.4)	288(62.2)
Age, years (Mean, SD)	23.6(3.3)	25.1(3.2)	22.4(2.9)	12.44	<0.001
Degree currently pursuing				197.4	<0.001
Bachelor’s degree	449(53.9)	99(26.8)	350(75.6)
Master’s or Doctorate degree	384(46.1)	271(73.2)	113(24.4)
Average monthly household income (RMB)				199.3	<0.001
≤10,000	327(39.3)	59(15.9)	268(57.9)
10,001–30,000	307(36.9)	151(40.8)	156(33.7)
≥30,001	199(23.9)	160(43.2)	39(8.4)
Present location				/	/
China	463(55.6)	/	463(100.0)
America	79(9.5)	79(21.4)	/
the United Kingdom	77(9.2)	77(20.8)	/
Canada	37(4.4)	37(10.0)	/
Australia	25(3.0)	25(6.8)	/
European Union countries	103(12.4)	103(27.8)	/
Asian countries	49(5.9)	49(13.2)	/

COVID-19 vaccination intention

Table 2 shows the COVID-19 vaccination intention of participants. A total of 536 (64.3%) participants reported the intention to be vaccinated against COVID-19, with overseas and domestic university students reporting similar vaccination uptake intentions (64.1% vs 64.6%, p > .05).

Table 2. The distribution of health belief items and vaccination intention by study groups (N = 833)

Items of Health Belief Model and vaccination intention	Overseas students (N = 370) N (%)	Domestic students (N = 463) N (%)	χ^Citation2/t	P	OR (95% CI)
Perceived Susceptibility^a
Perceived susceptibility score (Mean, SD)	3.25(0.84)	2.91(0.93)	5.549	<0.001
I may get COVID-19.	108(29.2)	103(22.2)	5.241	0.025	1.44(1.05–1.97)
Contacting with COVID-19 patients or close contacts of COVID-19 patients is possible for me.	171(46.2)	164(35.4)	9.968	0.002	1.57(1.19–2.07)
COVID-19 may break out in the area where I live.	182(49.2)	116(25.1)	52.14	<0.001	2.90(2.16–3.88)
Even with good hygiene practices (wearing a mask, etc.), I may still get COVID-19.	174(47.0)	221(47.7)	0.041	0.889	0.97(0.74–1.28)
Perceived Severity^a
Perceived severity score (Mean, SD)	3.01(0.82)	3.11(0.85)	1.719	0.086
I am afraid of getting COVID-19.	202(54.6)	257(55.5)	0.069	0.833	0.96(0.73–1.27)
I may probably die if I get COVID-19.	87(23.5)	167(36.1)	15.30	<0.001	0.55(0.40–0.74)
I will be gravely ill if I get COVID-19.	79(21.4)	123(26.6)	3.044	0.088	0.75(0.54–1.04)
My financial situation will get worse if I get COVID-19.	119(32.2)	161(34.8)	0.628	0.461	0.89(0.67–1.19)
Perceived Benefits^a
Perceived benefits score (Mean, SD)	3.77(0.71)	3.75(0.68)	0.411	0.681
I can reduce the risk of my family/friends getting COVID-19 after COVID-19 vaccination.	270(73.0)	324(70.0)	0.901	0.356	1.16(0.86–1.57)
I will feel more secure in my daily life after COVID-19 vaccination.	237(64.1)	294(63.5)	0.027	0.885	1.02(0.77–1.36)
In general, I will benefit from COVID-19 vaccination.	259(70.0)	327(70.6)	0.039	0.879	0.97(0.72–1.31)
Perceived Barriers^a
Perceived barriers score (Mean, SD)	3.51(0.71)	3.37(0.67)	2.884	0.004
I worry about the efficacy of the COVID-19 vaccine.	180(48.6)	183(39.5)	6.963	0.009	1.45(1.10–1.91)
I worry about the safety of the COVID-19 vaccine.	198(53.5)	207(44.7)	6.383	0.012	1.42(1.08–1.87)
I think the protection duration of the COVID-19 vaccine is too short.	173(46.8)	139(30.0)	24.59	<0.001	2.05(1.54–2.72)
I worry about the side effects of the COVID-19 vaccine.	223(60.3)	247(53.3)	4.008	0.049	1.33(1.01–1.75)
I worry that COVID-19 virus has mutated, and vaccines could not work.	216(58.4)	244(52.7)	2.682	0.107	1.26(0.96–1.66)
Cues to action^a
Cues to action score (Mean, SD)	3.70(0.69)	3.56(0.66)	3.099	0.002
My family members recommend me to get the COVID-19 vaccine.	196(53.0)	172(37.1)	20.88	<0.001	1.91(1.44–2.52)
My classmates/friends recommend me to get the COVID-19 vaccine.	185(50.0)	169(36.5)	15.34	<0.001	1.74(1.32–2.30)
Healthcare experts recommend me to get the COVID-19 vaccine.	272(73.5)	292(63.1)	10.27	0.001	1.63(1.21–2.19)
The local government advocates COVID-19 vaccination.	238(64.3)	312(67.4)	0.860	0.377	0.87(0.65–1.16)
Media advocate COVID-19 vaccination.	230(62.2)	263(56.8)	2.445	0.119	1.25(0.95–1.65)
Intention^b
I intend to get COVID-19 vaccination.	237(64.1)	299(64.6)	0.025	0.884	0.98(0.74–1.30)

^aAll health belief items used the five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree), the study population who choose strongly agree/agree were included in the form, and the mean scores from the HBM variables were calculated.

^bIntention item used the five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree), the study populations who have already been vaccinated and choose strongly agree/agree were included in the form.

Health belief model

Scores for the five dimensions of the HBM are presented in Table 2. Compared with domestic university students, overseas university students reported significantly higher scores in perceived susceptibility (3.25 ± 0.84 vs 2.91 ± 0.93, p < .05), perceived barriers (3.51 ± 0.71 vs 3.37 ± 0.67, p < .05), cues to action (3.70 ± 0.69 vs 3.56 ± 0.66, p < .05), slightly higher scores in perceived benefits (3.77 ± 0.71 vs 3.75 ± 0.68, p > .05), and slightly lower scores in perceived severity (3.01 ± 0.82 vs 3.11 ± 0.85, p > .05).

Factors associated with the COVID-19 vaccination uptake intention

After controlling for sociodemographic characteristics, logistic regression (Table 3) indicated that perceived benefits, perceived barriers, and cues to action were significantly associated with the vaccination intention of overseas and domestic university students. There is no significant difference between the vaccination intention of overseas and domestic university students. This study found that perceived susceptibility and perceived severity were not significant influencing factors of vaccination intention of university students. Overseas students with high perceived benefits (aOR = 6.74, 95% CI [3.72–12.23], p < .05), low perceived barriers (aOR = 0.29, 95% CI [0.16–0.50], p < .05), and high cues to action (aOR = 3.94, 95% CI [2.18–7.13], p < .05) showed greater COVID-19 vaccination intentions. Domestic students with high perceived benefits (aOR = 2.63, 95% CI [1.72–4.02], p < .05), low perceived barriers (aOR = 0.48, 95% CI [0.31–0.74], p < .05), and high cues to action (aOR = 5.84, 95% CI [3.43–9.96], p < .05) showed greater COVID-19 vaccination intentions. Only two of the HBM constructs (perceived benefits and cues to action) were found to positively affect the COVID-19 vaccination intentions among overseas and domestic university students. Specifically, for overseas students, perceived benefits showed the strongest positive relationship with the vaccination intention; for domestic students, cues to action showed the strongest positive relationship with the vaccination intention. Perceived barriers played a negative role in both overseas students (aOR = 0.29, 95% CI [0.16–0.50], p < .05) and domestic students (aOR = 0.48, 95% CI [0.31–0.74], p < .05), which is consistent with our hypothesis.

Table 3. Factors associated with the COVID-19 vaccination uptake intention of study population

	Total participants (N = 833)		Overseas students (N = 370)		Domestic students (N = 463)
	aOR (95% CI)^a	P	aOR^a (95% CI)	P	aOR^a (95% CI)	P
Sociodemographic characteristics
Group
Overseas students	Reference		/	/	/	/
Domestic students	0.89(0.56–1.43)	0.632	/	/	/	/
Gender
Male	Reference		Reference		Reference
Female	1.06(0.73–1.54)	0.775	0.87(0.47–1.60)	0.656	1.19(0.73–1.96)	0.485
Age (Mean, SD)	1.05(0.98–1.11)	0.156	1.05(0.94–1.17)	0.411	1.05(0.96–1.14)	0.320
Degree currently pursuing
Bachelor’s degree	Reference		Reference		Reference
Master’s or Doctorate degree	0.70(0.44–1.12)	0.137	1.62(0.75–3.53)	0.222	0.44(0.24–0.83)	0.011
Average monthly household income (RMB)
≤10,000	Reference		Reference		Reference
10,001–30,000	0.92(0.60–1.42)	0.715	0.65(0.27–1.54)	0.323	0.99(0.60–1.67)	0.994
≥30,001	0.90(0.53–1.55)	0.711	0.66(0.27–1.58)	0.355	1.03(0.41–2.59)	0.947
Present location
China	/	/	/	/	/	/
America	/	/	Reference	/	/	/
the United Kingdom	/	/	0.57(0.23–1.43)	0.234	/	/
Canada	/	/	1.42(0.44–4.63)	0.558	/	/
Australia	/	/	0.61(0.18–2.14)	0.443	/	/
European Union countries	/	/	0.57(0.23–1.44)	0.230	/	/
Asian countries	/	/	0.59(0.21–1.60)	0.296	/	/
HBM variables^b
Perceived susceptibility	1.16(0.91–1.46)	0.228	1.34(0.90–1.99)	0.154	1.09(0.80–1.50)	0.572
Perceived severity	1.12(0.87–1.45)	0.371	1.16(0.76–1.76)	0.503	1.09(0.77–1.54)	0.626
Perceived benefits	3.56(2.54–4.98)	<0.001	6.74(3.72–12.23)	<0.001	2.63(1.72–4.02)	<0.001
Perceived barriers	0.39(0.28–0.55)	<0.001	0.29(0.16–0.50)	<0.001	0.48(0.31–0.74)	0.001
Cues to action	4.55(3.11–6.68)	<0.001	3.94(2.18–7.13)	<0.001	5.84(3.43–9.96)	<0.001

^aaOR, Adjusted odds ratio; CI, confidence interval

^bAll health belief items used the five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree) and the mean scores from the HBM variables were calculated.

Path analysis

Based on the results of the logistic regression model, we established a path analysis model to explore the relationships among the HBM constructs, groups and vaccination intention. Path analysis showed that perceived barriers, perceived benefits, and cues to action were directly associated with the COVID-19 vaccination intention. Moreover, perceived barriers and cues to action can mediate the association between groups and the vaccination intention (ps <0.05). As shown in Figure.1 (Standardized solutions for the structural model among university students based on the health belief model.), the results demonstrated a desirable model fit (CFI = 0.999; TLI = 0.983; RMSEA = 0.022; χ² = 2.816; p = .245). Standardized coefficients were presented in Figure. 1.

Figure 1. Standardized solutions for the structural model among university students based on health belief model.

Discussion

To the best of our knowledge, this is the first study to describe and compare the COVID-19 vaccination intention between overseas and domestic Chinese university students. Our results demonstrated that both overseas and domestic university students had a low COVID-19 vaccination intention with no significant difference identified between groups. As hypothesized, perceived benefits and cues to action were both positively associated with university students’ COVID-19 vaccination uptake intention; perceived barriers were negatively associated with their vaccination uptake intention. However, perceived susceptibility and severity did not show significant impacts as we hypothesized. This insignificant association echoed previous studies which found that perceived severity was not typically strongly associated with preventive health behavior.⁴¹

In the present study, only less than two-thirds of overseas (64.1%) and domestic (64.6%) university students expressed a clear intention to vaccinate, which is lower than that of other Chinese populations.^4,5,34 It should be noted that most relevant studies were conducted during the outbreak of COVID-19 and the vaccination intentions have probably changed because of a well-controlled situation of this pandemic in China.⁴² Existing studies also showed a substantial reduction of public vaccination uptake intention rates as the pandemic progressed over different phases.^43,44 Overseas and domestic university students reported similar vaccination intentions. Previous studies noted that highly educated groups like university students were more likely to hold a negative attitude toward vaccination, regardless of location.⁴⁵ Previous studies showed that there were significant differences in vaccination intention among populations in different countries.^46,47 However, the study groups in these studies, which focused on healthcare workers and adult parents, were different from the university students surveyed in this study. University students have similar age, similar educational level, and low possibility of occupational exposure, which were different from the characteristics of respondents in previous studies. Thus, it was reasonable that overseas and domestic university students reported similar vaccination uptake intention because of their similar social economic characteristics.

In the path analysis, we found that perceived barriers, benefits and cues to action had a direct impact on vaccination intention, and perceived barriers and cues to action were mediating factors between groups of university students and vaccination intention. Moreover, our results showed that groups of university students were indirectly related to vaccination intention through perceived barriers, perceived benefits and cues to action. Cues to action were directly and indirectly associated with vaccination intention through perceived barriers and perceived benefits.

The present study showed that perceived benefits of COVID-19 vaccination were positively associated with COVID-19 vaccination intentions among university students, which is consistent with previous studies.^48,49 Perceived benefits of COVID-19 vaccination can be communicated through creative and impressive slogans and posters in school campuses emphasizing the importance of the COVID-19 vaccine to protect our health and the role of the vaccine in controlling the pandemic as some community subdistrict offices did. Compared with western societies where individualism is prevalent, Asian countries are more inclined to collectivism⁵⁰. People with higher collectivism may show higher vaccination rates⁵¹. Moreover, cultural factors related to collectivism may also help individuals make better decisions⁵². For example, Asian culture may drive individuals to make vaccination decisions from the perspective of family benefits. Therefore, we suggest that when publicizing the benefits of vaccines, relevant departments can establish a health concept based on collectivism. For example, instead of emphasizing the benefits of university students themselves, policymakers should emphasize that vaccination can protect their valued family and friends.

Perceived barriers were significant determinants of the COVID-19 vaccination intention among university students. The identified barriers in this study, namely worries about efficacy, safety, protection duration, and side effects of the COVID-19 vaccine, were likewise reported in other studies about vaccination intentions and vaccine hesitancy.^5,21,37,49 The results also showed that overseas university students expressed more perceived barriers to the COVID-19 vaccine than domestic university students. Overseas university students were significantly more worried about the efficacy, safety, protection period, and side effects of COVID-19 vaccines than domestic university students. As proposed by previous studies, robust evidence from empirical clinical studies about the safety and efficacy of vaccines in plain language disseminated by famous healthcare experts can help relieve the vaccination hesitancy among well-educated groups.^5,53–55 Thus, healthcare workers should be regularly invited to communicate with overseas university students online to share the latest progress about vaccines. To alleviate public concerns about vaccine safety and respond to future possible pandemics, policymakers should also provide information and health education about vaccine safety to the public regularly. What’s more, channels for the public to communicate with authoritative experts will also be necessary.^5,56–59

External cues to action had a strong and positive effect on vaccination intentions among university students, which was consistent with previous studies.³⁷ Suffering from fear of infection, economic instability, stress, anxiety, and even depression,⁶⁰ overseas university students often turn to familiar and close people for help,⁶¹ who may lack certain medical knowledge to provide appropriate advice. Thus, our government should offer targeted health education to the families and friends of overseas university students, which may influence overseas students and improve their vaccine literacy indirectly. Moreover, overseas university students should be well informed that they can always seek help from the embassy. The embassy is equipped with professional healthcare workers who can offer health advice and recommendations to overseas university students.

The effects of perceived benefits, perceived barriers, and cues to action on the COVID-19 vaccination intention differed between overseas and domestic university students. Although overseas and domestic university students reported similar vaccination intention rates, key elements to the vaccination program of these two groups should better be different. As the present study showed, perceived benefits should be the priority of the COVID-19 vaccination promotion among overseas university students, while cues to action should be greatly valued among domestic university students in promoting the COVID-19 vaccination. For overseas university students, governments can cooperate with universities to produce creative and simple short videos which can be disseminated easily and without confusion targeting perceived benefits of COVID-19 vaccines. These videos can be publicized to overseas university students with the help of official organizations or local overseas students’ organizations. For domestic university students, the star effect of talent shows can be used to enhance their vaccination intentions. For example, audiences of talent shows are mostly university students. These students used to buy specified idol-related products to obtain voting rights and vote for their favorite idols. The National Health Commission and relevant departments can cooperate with talent shows’ organizers, stipulating that audiences who receive COVID-19 vaccines can have extra opportunities to vote for their preferred idols.

Several limitations should be taken into consideration when interpreting the present results. Firstly, we used the snowball sampling method to enroll university students which may result in certain sampling bias. To minimize possible bias, we selected different countries for overseas university students and comprehensive universities for domestic university students. Secondly, this study reported a relatively high proportion of female respondents. However, the gender of participants didn’t show effects on the vaccination intention in the present study. Thirdly, the characteristics of vaccines, such as the type and company of vaccines and the country in which vaccines were manufactured, were not considered, which may affect the vaccination intention.²¹ Further studies are needed to investigate the impact of different characteristics of vaccines on vaccination intentions.

Conclusion

This study reflected a relatively low COVID-19 vaccination intention among overseas and domestic university students. Perceived benefits, perceived barriers, and cues to action were important factors that influenced the vaccination intention among overseas and domestic university students. The key elements of the vaccination program for these two groups should be different. It is worth trying to promote the vaccination intention of overseas and domestic university students through the dissemination of creative short videos and star voting.

Author contributions

HW, JL, and JL made substantial contributions to the study design and supervised the data collection. HW, JL, XZ, TJ, and XW contributed to the data collection and interpretation. HW wrote the substantial parts of manuscript. XZ and JL commented on manuscript. All authors critically revised, reviewed, and approved the final version the manuscript.

Acknowledgments

We would like to thank for Jiayao Xu, Yulian Yang, Shixin Dong, Menmen Wang, Hailati Akezhuoli for their great assistance to conduct this survey. We are grateful to data collectors for their work and all respondents for their participation in the study.

Disclosure statement

The authors declare there is no conflict of interest.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.