A mixed-methods study to assess COVID-19 vaccination acceptability among university students in the United Arab Emirates

ABSTRACT

To effectively achieve herd immunity against COVID-19, a substantial proportion of a population would need to be vaccinated. However, vaccine hesitancy and refusal are significant issues globally. This mixed-methods study aimed to investigate university students’ attitudes in the United Arab Emirates (UAE) toward the COVID-19 vaccination, determine the factors associated with vaccine hesitancy, and understand the underlying reasons. We conducted an online survey between 16th-24th February 2021 in 669 students from the University of Sharjah (UAE) and semi-structured qualitative interviews with a subsample of 11 participants. Data on COVID-19 vaccine intention and uptake, risk perception, beliefs and attitudes toward the disease and the vaccine were collected. Multinomial logistic regression was applied and thematic content analysis was conducted with qualitative data. Overall, 31.8% of students demonstrated vaccine hesitancy; 24.4% of students reported a high intention to get the vaccine, and 43.8% were already vaccinated. Vaccine hesitancy was associated with less positive beliefs and attitudes toward the COVID-19 vaccine (AdjOR = 0.557;95%CI 0.468–0.662), high perceived adverse effects (AdjOR = 1.736;95%CI 1.501–2.007), and not perceiving easy access to a vaccination center (AdjOR = 0.820;95%CI 0.739–0.909). The main reasons underlying vaccine hesitancy were related to uncertainty about the effectiveness of the vaccine, knowledge about negative experiences from vaccination among family and/or community, overvaluing the risks of the vaccine in relation to the potential benefits, and not perceiving immunization as a social norm. To increase COVID-19 vaccination uptake, interventions to reduce hesitancy could focus on reducing fears about adverse effects and highlighting individual and societal benefits of the vaccination.

KEYWORDS:

• COVID-19 vaccine
• vaccination hesitancy
• students
• attitudes
• beliefs

Introduction

The novel coronavirus disease 2019 (COVID-19) presents an unprecedented health challenge and has led to a dramatic loss of human life globally. Confirmed cases of the disease are over 200 million worldwide.¹ The United Arab Emirates (UAE) registered over 600,000 positive cases and 1,900 deaths.¹ Less than one year after the World Health Organization’s (WHO) declaration of a global pandemic,² many countries started distributing vaccines. The UAE is currently one of the countries with the highest proportion of people who have received at least one vaccine dose against COVID-19.^3,4 Vaccines are available throughout the UAE, free of charge and on an optional basis for medically eligible citizens and residents, and according to the priorities announced by the UAE’s health authorities.⁵ Four vaccines are currently available in the UAE from Sinopharm, Pfizer-BioNTech, Sputnik V, and Oxford-AstraZeneca.⁵

Vaccination programs need to enroll high proportions of the population to achieve herd immunity. However, vaccine hesitancy and refusal are significant issues globally. The World Health Organization defines vaccine hesitancy as a “delay in acceptance or refusal of vaccines despite availability of vaccine services.”⁶ Previous studies have identified several factors related to vaccine hesitancy: communication and media environment, religion, culture, geographical accessibility, perception of the pharmaceutical industry; individual and group influences, such as the beliefs and attitudes about health and prevention, the knowledge and awareness about the disease, the perceived risk/benefit of receiving the vaccine; and, lastly, vaccine-specific issues, such as the scientific evidence available about the risk/benefit, the mode of administration, the costs, and the strength of recommendation of healthcare professionals.⁶

The Common-Sense Model of Self-Regulation (CSM) provides a conceptual framework for how these perceptual, behavioral and cognitive processes may explain individuals’ self-management of ongoing and future health threats, such as contracting COVID-19.^7,8 These processes are primarily influenced by the representation of the illness for the individuals, which, according to Leventhal and colleagues, has five components: identity, which is the label or name given to the condition and its associated symptoms; cause, the ideas about the perceived cause of the condition, which may be based on information obtained from personal experience as well as the opinions of significant others, health professionals and the media; timeline, the belief about how long the condition might last; consequences, the individual beliefs about the consequences of the condition and how this will impact on them at physical and social levels; and curability/controllability, the beliefs about whether the condition can be cured and the degree to which the individual plays a part in achieving this.⁷ Thus, the decision to get the vaccine would be influenced by how individuals think about the susceptibility to and severity of COVID-19, vaccine benefits, barriers to being vaccinated (access), subjective norms, locus of control, anticipated regret, and perceived knowledge. Previous studies about vaccination acceptability in the UAE showed that the uptake of the influenza vaccination among healthcare workers was low, with a vaccination rate of 24.7%,⁹ and the uptake of human papillomavirus vaccination among women was very low (6.3%), although 79.5% considered taking it in the future.¹⁰ However, when it comes to parents’ attitudes regarding childhood vaccines, acceptability seems high, with only 12% of parents indicating vaccination hesitancy.¹¹ Specifically, regarding the acceptability of COVID-19 vaccination, a recent survey across all seven emirates of the UAE study showed that 25% of the population were unwilling, and 21% were “a little” willing to take a COVID-19 vaccine.¹² Yet, no sufficient studies have been conducted to assess University students’ perceptions and attitudes toward COVID-19 vaccination.

Understanding vaccine acceptability and uptake among university students is imperative in planning prevention efforts. A recent analysis of COVID-19 transmission dynamics in the United States revealed that increased COVID-19 positivity among young people preceded the identification of hotspots by over 30 days and was followed by several weeks of increased positivity among older groups, including those over 60 years.¹³ Several factors explain university students’ role in COVID-19 transmission, including domestic and international travel to pursue studies, decreased compliance with preventive measures, and low perception of COVID-19 risk. For these reasons and considering their potential role as important agents in disseminating evidence-based knowledge that can increase population awareness, adolescents and young adults have been suggested to be the key demographic for preventing COVID-19 transmission globally.^14–16 Moreover, the important role of universities as health promotion environments demands that information about these health dimensions is gathered so that appropriate intervention can be designed.¹⁷

Thus, this mixed-methods study aimed to examine university students’ attitudes in the UAE toward the COVID-19 vaccination. This study quantitatively determined the factors associated with vaccine hesitancy and, through qualitative methods, we sought to gain an in-depth understanding of students’ experiences, perceptions, and reasons underlying vaccine hesitancy.

Materials and methods

Study design

This study used a mixed-methods approach with a sequential explanatory design,¹⁸ in which qualitative data helps explain initial quantitative results.^18,19 An online cross-sectional survey was used to determine the acceptability of the COVID-19 vaccine, uptake, and associated factors. Subsequently, semi-structured individual interviews were conducted with a subsample of the students who completed the online survey to explore the reasons underlying vaccine acceptability.

Cross-sectional survey

Setting and data collection

The cross-sectional survey was conducted among undergraduate and postgraduate students at the University of Sharjah. The University of Sharjah is a private national university located in the Emirate of Sharjah in the UAE. It was established in 1997.²⁰ It offers 109 undergraduate and postgraduate programs in various disciplines, with around 17,000 students from 101 countries enrolled in 2020.²⁰ As a precautionary measure against the spread of COVID-19, the University implemented distance learning for all students since March 2020. Most students are still learning remotely, with limited exceptions for students to attend certain practical sessions and experiential learning at training sites.

Students were invited through their institutional e-mails to complete an online survey between 16th and 24th February 2021, a reminder e-mail was sent on 21th February. Students were provided with information about study objectives and methods, and provided informed consent before completing the questionnaire.

Survey development

The survey comprised three sections and was adapted from the COVID-19 vaccination acceptability study (CoVAccS) by Sherman et al.,²¹ conducted in the United Kingdom. In the first section, participants were asked about sociodemographic, health, and psychosocial characteristics such as gender, age, nationality, employment status, and household living arrangements. Students also reported whether they or someone else in their household had a chronic disease that could make them clinically vulnerable to serious illness from COVID-19. To assess students’ locus of control, an adapted version of the Short Scale for the Assessment of Locus of Control (KMKB scale) was used, which was developed according to Rotter’s social learning theory.²² Locus of control defines a personal belief about whether outcomes are determined by one’s actions or by forces outside an individual’s control. Internal locus of control describes an individual’s belief that their behavior can control life events, while external locus of control refers to the belief that external forces control one’s life.²² Students were given four statements and were asked to report the extent to which these statements applied to them. The response options ranged from ‘applies to me’ to ‘does not apply to me.’ An arithmetic mean of the total scale was calculated, with higher numbers representing a higher internal and external locus of control.²² The second section assessed the perception of risk posed by COVID-19 on people in the UAE in general and the students personally on a four-point scale. Moreover, students’ beliefs and attitudes toward the COVID-19 disease were assessed with six items using a 5-point Likert scale from “strongly disagree” to “strongly agree.” The arithmetic mean of the scale was calculated, with higher scores representing higher perceived threat and impact of COVID-19. In the third section, participants reported their beliefs and attitudes toward the COVID-19 vaccination, reporting their agreement to 22 items on an eleven-point scale from “strongly disagree” to “strongly agree.”²¹ From the principal component analysis, three theoretical meaningful components emerged. The designations for the subscales used by Sherman et al.²¹ were adopted. The first component measured “general COVID-19 vaccination beliefs and attitudes” (perceived vaccine effectiveness, social norms, likelihood of catching COVID-19 without a vaccine, anticipated regret of not being vaccinated) (α = 0.88); the second component, “COVID-19 vaccination adverse effects,” measured perception of adverse effects from getting the vaccine (α = 0.69); and the third component measured “perceived information sufficiency” to be able to make an informed decision about vaccination (α = 0.81). For each component, higher values corresponded to more positive beliefs and attitudes toward the vaccine, to attributing higher adverse effects to it, and to perceiving higher levels of knowledge about the vaccination, respectively. Items about ease of access to a coronavirus vaccination center, fear of needles and not having to follow social distancing and other restrictions if one were vaccinated did not load on to the previous components so these were analyzed as isolated items.

The composite outcome measure was the vaccination uptake and vaccine hesitancy (measured by intention to have the vaccine). Intention was measured with one item assessing students’ likelihood of getting the vaccine on an eleven-point scale from “extremely unlikely” to “extremely likely.” Previously proposed cutoffs were used to dichotomize intention to get the vaccine as “very unlikely and uncertain” (0–7 points) and “very likely” (8–10 points).²¹ Therefore, students were grouped into three groups: vaccinated, unvaccinated with high intention to get the vaccine (low hesitancy group), and unvaccinated with low intention to get the vaccine (high hesitancy group). The survey was available to students in Arabic and English, according to preference. The Arabic version of the survey was pilot-tested with nine students, and the English version was pilot-tested with four students. On average, the questionnaire took 8 minutes to complete.

Sample size

Before a COVID-19 vaccine became available, a study in the general population in the UAE estimated vaccine hesitancy to be about 45%. Thus, for the calculation of the sample size, we assumed 50% vaccine hesitancy. Using Cochran’s sample size formula and a 95% CI for the prevalence of vaccine hesitancy with a maximum precision error of 5%, 384 students were needed to complete the survey. The calculated sample size was increased by 20% to account for non-responses and incomplete surveys, making the final minimum sample size required 461 students.

Statistical analysis

The association between vaccination status and vaccine hesitancy and the categorical variables was quantified through the Chi-square test or the Fisher’s Exact Test. For the continuous variables, individuals’ scores were calculated using the arithmetic mean of the scale. Mean differences were compared using one-way ANOVA. Bonferroni and Tukey HSD Post Hoc tests were used when appropriate, and significant results are reported. Multinomial logistic regression was used to identify the predictors of vaccine hesitancy, and adjusted odds ratios (AdjOR) and 95% confidence intervals (95% CI) are presented. The main criteria for entering the variables in the regression model was a p-value ≤ 0.05 in the bivariate analysis, and the Enter method was used. The level of significance was set at 0.05. The IBM Statistical Package for the Social Sciences (SPSS) Statistics for Windows, version 27.0, Armonk, NY, USA, was used for all analyses.

Qualitative study

Participants and data collection

In the last section of the survey, students were asked to provide their e-mail if they were interested in participating in a subsequent semi-structured individual interview. A list of students who expressed interest to be interviewed (n = 86) was created. Heterogeneous sampling was used to select students from the list based on the following criteria: sex, vaccination status, and intention to get the vaccine. Recruitment stopped when data saturation was achieved (i.e. the point at which no new data was expressed).²³ In total, 11 interviews were conducted.

The selected students received an information sheet about the qualitative study by e-mail, describing the objectives and methods. Before conducting the interviews, each student provided oral consent, which was recorded. All interviews were conducted online, over MS Teams, between 7 and 18 March 2021 and lasted 27 minutes on average. Six interviews were conducted in English and five in Arabic.

The interview guide was developed based on the WHO determinants of vaccine hesitancy matrix.⁶ Both English and Arabic versions of the guide were pilot tested with seven university students to ensure understandability, and minor linguistic changes were made. Interviews were audio-recorded, and transcribed verbatim. Arabic interviews were translated into English and back translated to ensure consistency of the translations.

Content analysis

Data were analyzed by thematic content analysis,²⁴ using the software NVivo 11 (QSR International, USA, 2015). A triangulation strategy was used to ensure rigor and quality of the research – one author identified, sentence by sentence, the reasons underlying acceptability of COVID-19 vaccination among students and another author collaborated on developing the coding framework. Quotations with similar meanings were deductively synthesized into categories and then grouped into the following analytical themes, according to the determinants of vaccine hesitancy matrix by the World Health Organization: contextual influences, including factors related to historical, socio-cultural, environmental, health system/institutional, economic or political factors; individual and group influences, when referring to the personal perception of the vaccine or influences of the social/peer environment; and vaccine/vaccination-specific issues, including all the dimensions directly related to the vaccine, as the balance between risks/benefits, the mode of administration or the costs.⁶ Categories were added and included in these themes. The most illustrative verbatim quotes were selected.

Ethical approval

Ethical approval was granted by the Research Ethics Committee of the University of Sharjah (reference number REC-21-02-07-01).

Results

Quantitative results

Of 966 students who began the survey, 740 completed it, and the completion rate was 76.6%. Seventy-one questionnaires were excluded because they did not meet attention checks (i.e., items designed to detect acquiescence and straight-lining). Thus, data from 669 students were analyzed. Most respondents were female (81.2%) and were enrolled in a Bachelor’s degree program, with half of the students belonging to the medical and health fields (50.4%). The mean age of students was 21 years (SD = 3.87), and most students were living with family members. More than half of students were non-Emirati Arab (53.7%), and almost all were currently living in the UAE (90.4%). Demographic characteristics are displayed in Table 1.

Table 1. Participants’ sociodemographic, clinical and psychosocial characteristics, according to vaccination uptake and intention (N = 669)

				Unvaccinated
			Vaccinated	Low intention	High intention
		Total	n= 293	n = 213	n = 163
Characteristics		n (%)	n (%)	n (%)	n (%)	p value
Gender	Male	126 (18.8)	62 (21.2)	34 (16.0)	30 (18.4)	0.338
	Female	543 (81.2)	231 (78.8)	179 (84.0)	133 (81.6)
Age (years), mean (SD)		20.98 (3.87)	21.12 (3.89)	21.39 (4.40)	20.20 (2.88)	0.009
Degree program	Bachelor	605 (90.4)	255 (87.0)	195 (91.5)	155 (95.1)	0.016
	Master	49 (7.3)	26 (8.9)	15 (7.0)	8 (4.9)
	Ph. D.	15 (2.2)	12 (4.1)	3 (1.4)	0
College	Arts, Humanities, Social Sciences …	161 (24.1)	70 (23.9)	60 (28.2)	31 (19.0)	0.203
	Informatics, Engineering, Sciences	171 (25.6)	79 (27.0)	45 (21.1)	47 (28.8)
	Medical Cluster	337 (50.4)	144 (49.1)	108 (50.7)	85 (52.1)
Nationality	Emirati Arab	202 (30.2)	126 (43.0)	45 (21.1)	31 (19.0)	<0.001
	Non-Emirati Arab	359 (53.7)	130 (44.4)	127 (59.6)	102 (62.6)
	Other	108 (16.1)	37 (12.6)	41 (19.2)	30 (18.4)
Currently resident in UAE	Yes	605 (90.4)	273 (93.2)	191 (89.7)	141 (86.5)	0.064
	No	64 (9.6)	20 (6.8)	22 (10.3)	22 (13.5)
Marital status	Not married	641 (95.8)	282 (96.2)	199 (93.4)	160 (98.2)	0.071
	Married	28 (4.2)	11 (3.8)	14 (6.6)	3 (1.8)
Employment status	Student	504 (80.0)	235 (83.3)	151 (77.4)	118 (77.1)	0.173
	Full- or part-time employed	126 (20.0)	47 (16.7)	44 (22.6)	35 (22.9)
Children	Yes	20 (3.0)	9 (3.1)	8 (3.8)	3 (1.8)	0.593
	No	649 (97.0)	284 (96.9)	205 (96.2)	160 (98.2)
Household living arrangements	Lives alone/non-family members	51 (7.6)	14 (4.8)	22 (10.3)	15 (9.2)	0.046
	Lives with family members	618 (92.4)	279 (95.2)	191 (89.7)	148 (90.8)
Number of people in household	1–4	130 (20.6)	44 (15.6)	44 (22.6)	42 (27.5)	0.010
	≥5	500 (79.4)	238 (84.4)	151 (77.4)	111 (72.5)
Clinically vulnerable – respondent	Yes	99 (14.8)	29 (9.9)	45 (21.1)	25 (15.3)	0.002
	No	570 (85.2)	264 (90.1)	168 (78.9)	138 (84.7)
Clinically vulnerable – other(s) in household	Yes	417 (62.3)	174 (59.4)	139 (65.3)	104 (63.8)	0.361
	No	252 (37.7)	119 (40.6)	74 (34.7)	59 (36.2)
Locus of control, mean (SD)	Internal	2.62 (0.41)	2.61 (0.41)	2.62 (0.43)	2.64 (0.40)	0.718
	External	2.26 (0.49)	2.24 (0.51)	2.29 (0.48)	2.26 (0.48)	0.575

Some percentages do not add up to 100% due to rounding.

Vaccination uptake and intention

Table 1 displays the clinical and psychosocial characteristics of participants by vaccination status and intention of being vaccinated. Overall, 31.8% (n = 213) of the students showed vaccine hesitancy, reporting low levels of vaccination intention; 24.4% (n = 163) reported high levels of vaccination intention, and 43.8% (n = 293) were vaccinated. A higher proportion of male students were vaccinated than female students, although these differences were not significant. Younger students were less likely to be vaccinated but had significantly higher vaccination intentions than older students (p = .009). Low vaccination intention was more frequent among non-Emirati Arab students (p < .001).

Additionally, the proportion of students living with family members was higher among vaccinated and those who had high vaccination intention compared to students with low vaccination intention (p = .046). A significantly higher proportion of students living with more than five family members were vaccinated than those living with a smaller number of household members (p = .010). The proportion of students with at least one condition that made them clinically vulnerable to serious illness from COVID-19 was higher among those with low vaccination intention (p = .002).

Attitudes and beliefs about COVID-19 and vaccination status

Overall, students believed that COVID-19 posed a moderate risk to people in the UAE (46.5%), with most also reporting they believed COVID-19 posed a moderate risk to them personally (42.2%). Neither perception of risk to the overall UAE population nor themselves varied by vaccination uptake status, as shown in Table 2. However, students who had a higher perceived threat of COVID-19 had higher vaccination intention (p < .001). The proportion who were diagnosed with COVID-19 was lower among vaccinated students (p = .001).

Table 2. Attitudes and beliefs about the COVID-19 illness and vaccination, according to vaccination uptake and intention (N = 669)

				Unvaccinated
Characteristics		Total	Vaccinated n= 293	Low intention n = 213	High intention n = 163
Attitudes and beliefs about the COVID-19 illness		n (%)	n (%)	n (%)	n (%)	p value
To what extent do you think coronavirus poses a risk to people in the UAE?	Major risk	307 (45.9)	132 (45.1)	97 (45.5)	78 (47.9)	0.756
	Moderate risk	311 (46.5)	136 (46.4)	103 (48.4)	72 (44.2)
	Minor risk	46 (6.9)	22 (7.5)	11 (5.2)	13 (8.0)
	No risk at all	5 (0.7)	3 (1.0)	2 (0.9)	0
To what extent do you think coronavirus poses a risk to you personally?	Major risk	141 (21.1)	52 (17.7)	47 (22.1)	42 (25.8)	0.251
	Moderate risk	282 (42.2)	123 (42.0)	94 (44.1)	65 (39.9)
	Minor risk	193 (28.8)	89 (30.4)	56 (26.3)	48 (29.4)
	No risk at all	53 (7.9)	29 (9.9)	16 (7.5)	8 (4.9)
COVID-19 diagnosis – respondent	No	579 (86.5)	270 (92.2)	178 (83.6)	131 (80.4)	0.001
	Yes	90 (13.5)	23 (7.8)	35 (16.4)	32 (19.6)
COVID-19 diagnosis – a known person	No	43 (6.4)	24 (8.2)	11 (5.2)	8 (4.9)	0.285
	Yes	626 (93.6)	269 (91.8)	202 (94.8)	155 (95.1)
		mean (SD)	mean (SD)	mean (SD)	mean (SD)
Perceived threat and impact of COVID-19		3.69 (0.59)	3.62 (0.58)	3.62 (0.59)	3.89 (0.55)	<0.001
Attitudes and beliefs about the COVID-19 vaccination		mean (SD)	mean (SD)	mean (SD)	mean (SD)
General COVID-19 vaccination beliefs and attitudes		6.44 (2.07)	7.19 (1.57)	4.64 (1.99)	7.46 (1.34)	<0.001
COVID-19 vaccination adverse effects		3.99 (2.21)	2.99 (1.78)	5.59 (1.90)	3.71 (2.10)	<0.001
Perceived knowledge sufficiency		6.68 (2.54)	7.26 (2.39)	5.86 (2.56)	6.72 (2.49)	<0.001
It would be/was easy for me to have access to a coronavirus vaccination center		7.41 (3.05)	8.78 (2.09)	6.17 (3.14)	6.57 (3.37)	<0.001
I am afraid of needles		2.50 (3.46)	2.30 (3.37)	2.49 (3.38)	2.85 (3.71)	0.268
With the vaccine, I do not need to follow social distancing and other restrictions for coronavirus		1.47 (2.88)	0.97 (2.51)	2.12 (3.35)	1.52 (2.66)	<0.001

Some percentages do not add up to 100% due to rounding.

Vaccinated students reported a higher positive attitude to the COVID-19 vaccine (M = 7.19, SD = 1.57) than unvaccinated students who had low vaccination intention (M = 4.64, SD = 1.99) (p < .001). Additionally, students who believed the COVID-19 vaccination had higher adverse effects had lower vaccination intention (M = 5.59, SD = 1.90, p < .001). Lower perceived knowledge of the vaccine was also associated with lower vaccination intention (M = 5.86, SD = 2.56, p < .001).

Multinomial logistic regression was performed to model the relationship between the predictors for vaccination hesitancy among unvaccinated students with high intention and unvaccinated students with low intention as outcome variables, when compared to vaccinated ones. Addition of the predictors to a model that contained only the intercept significantly improved the fit between model and data, X² (df = 28, N = 630) = 464.149, Nagelkerke R² = 0.52, p < .001. As shown in Table 3, students who reported they had been previously diagnosed with COVID-19 had 3.412 higher odds of intention to receive the vaccine than those who did not report a COVID-19 infection. Similarly, reporting a perceived threat from COVID-19 (AdjOR = 2.187; 95% CI 1–411-3.389) and positive beliefs and attitudes toward the COVID-19 vaccination (AdjOR = 1.336; 95% CI 1.122–1.590) predicted higher vaccination intention. Less positive beliefs and attitudes toward the COVID-19 vaccine (AdjOR = 0.557; 95% CI 0.468–0.662), perceived adverse effects from the COVID-19 vaccine (AdjOR = 1.736; 95% CI 1.501–2.007), not perceiving easy access to a vaccination center (AdjOR = 0.820; 95% CI 0.739–0.909) and believing that social distancing would not need to be followed if vaccinated (AdjOR = 1.123; 95% CI 1.024–1.231) predicted lower vaccination intention.

Table 3. Multinomial regression model for the predictors of vaccination hesitancy (N = 630)

	Vaccination intention
	Unvaccinated – high intention			Unvaccinated – low intention
Variable	AdjOR	95% CI	p-value	AdjOR	95% CI	p-value
Age	0.920	0.838–1.010	0.079	1.038	0.956–1.127	0.374
Degree program
Master/PhD	1			1
Bachelor	1.471	0.482–4.488	0.482	1.576	0.478–5.196	0.455
Nationality
Emirati Arab	1			1
Non-Emirati Arab	1.635	0.936–2.853	0.084	1.601	0.850–3.016	0.145
Other	1.494	0.717–3.115	0.284	1.461	0.654–3.261	0.355
Household living arrangements
Lives alone/non-family members	1
Lives with family members	0.825	0.121–5.611	0.844	2.294	0.262–20.062	0.453
Number of people household
1–4	1			1
≥5	0.640	0.361–1.134	0.126	0.916	0.465–1.806	0.800
Clinically vulnerable – respondent
No	1			1
Yes	1.326	0.657–2.676	0.431	1.953	0.901–4.232	0.090
COVID-19 diagnosis – respondent
No	1			1
Yes	3.412	1.757–6.627	<0.001	1.837	0.837–4.036	0.130
Perceived threat and impact of COVID-19	2.187	1.411–3.389	<0.001	1.344	0.821–2.201	0.240
General COVID-19 vaccination beliefs and attitudes	1.336	1.122–1.590	0.001	0.557	0.468–0.662	<0.001
COVID-19 vaccination adverse effects	1.266	1.117–1.435	<0.001	1.736	1.501–2.007	<0.001
Perceived knowledge sufficiency	0.956	0.861–1.061	0.396	0.915	0.816–1.025	0.915
It would be/was easy for me to have access to a coronavirus vaccination center	0.771	0.701–0.846	<0.001	0.820	0.739–0.909	<0.001
With the vaccine, I do not need to follow social distancing and otherrestrictions for coronavirus	1.040	0.949–1.141	0.403	1.123	1.024–1.231	0.014
					Model summary F-value: 464.149 p-value: <0.001 R^2: 0.521

Reference class: Vaccinated; 1 = Reference category; AdjOR: Adjusted Odds Ratio for all the variables in the table; CI: Confidence Interval.

Qualitative results: reasons underlying vaccine hesitancy

The characteristics of the students who participated in interviews are described in Table 4. In general, the reasons underlying students’ decisions to choose to take the vaccine or not were related to vaccine-specific issues, including the perception about the risks versus benefits of taking the vaccine and its effectiveness and safety; individual and group influences such as the beliefs and attitudes about the disease and its consequences, knowledge and awareness about the vaccine and personal, family and community experience with vaccination and feelings of solidarity; and contextual influences, such as communication and media environment, the perception of the pharmaceutical industry and trust in health government policies.

Table 4. Characteristics of the interviewees

Interview*	Gender	Age	Vaccination status	Degree	Children
I1	Female	18	Unvaccinated – high intention	Bachelor in Pharmacy	No
I2	Female	42	Unvaccinated – low intention	Master in Arts, Humanities, & Social Sciences	Yes
I3	Male	21	Unvaccinated – low intention	Bachelor in Law	No
I4	Male	40	Unvaccinated – high intention	Bachelor in Computing & Informatics	Yes
I5	Female	21	Vaccinated	Bachelor in Pharmacy	No
I6	Female	19	Unvaccinated – low intention	Bachelor in Health Sciences	No
I7	Female	19	Unvaccinated – high intention	Bachelor in Computing & Informatics	No
I8	Male	23	Vaccinated	Bachelor in Sharia and Islamic Studies	No
I9	Female	18	Unvaccinated – low intention	Bachelor in Pharmacy	No
I10	Female	18	Vaccinated	Bachelor in Health Sciences	No
I11	Female	19	Unvaccinated – high intention	Bachelor in Health Sciences	No

*Participants are presented by the order of the interviews.

The main reasons mentioned by students who were hesitant to take the COVID-19 vaccine were related to contextual influences and vaccine-specific issues. Within contextual influences, students referred to a communication and media environment where uncertainty about the effectiveness of the vaccine was common and the sources of information were not fully trusted:

Q1: We don’t know whom to believe. Even the World Health Organization comes out with a different decision every day. (I3)

Also mentioned was a perception that the pharmaceutical industry was more focused on financial gain than benefiting human health:

Q2: I think that when vaccines first came out in the world, their aim was to actually benefit humanity, but afterwards, this changed with drug companies, financial gain, and these things. (I2)

Among vaccine-specific issues, students with low vaccination intention considered that the vaccine currently presented more risks than benefits. They were fearful due to reports of adverse events, albeit in a small number of cases:

Q3: There hasn’t been enough time for studies to have been conducted for the coronavirus vaccine (…). There are people that I personally know who got the vaccine, both doses, and were infected. So I have some doubts. (I2)

Q4:(…) It’s not really a hundred percent successful, and one person makes millions of people scared of taking it just because it can cause some dangers, so this can be a disadvantage. (I9)

Although less mentioned, students with low vaccination intentions also referenced individual and group influences, including knowledge about negative experiences with vaccination from family and/or community (Q5); believing that their personal risk of contracting COVID-19 was low and that the consequences would not be serious (Q6); a personal experience that reinforces a sense of “self-sufficiency” (Q7):

Q5: People’s experiences also draw your attention – why do some people get it very mild and some die? (I2)

Q6: I feel that it [the personal risk of getting COVID] is low (…) because I’m still young (…), I haven’t got any diseases. (I9)

Q7: … I won’t take the vaccine, honestly, I’m not considering that at all because, honestly, since I was little, I’m used to natural remedies, with herbs and these things. So, I think that I can treat myself on my own from any disease. (I2)

Vaccinated students mostly referenced individual and group influences underlying their positive attitude toward the vaccine. While also acknowledging uncertainty regarding, for example, the communication and media environment about the vaccine, other factors weighed highly in the vaccinated students’ decision. These included beliefs that the benefits of the vaccine outweighed the risks (Q8); positive experiences of family and the community with the vaccine (Q9); trust in the health care system and government health policies (Q10):

Q8: So there, of course, more advantages than the disadvantages. Advantages are like it provides with the immunity, it would help us like bring the pandemic to an end (…). And the disadvantages I don’t really think there are disadvantages (…). (I1)

Q9: [Feeling safe to take the vaccine] (…) especially when people started to take the vaccine (…), and I felt like okay to take it! (I5)

Q10: The Ministry of Health, it’s their role, they are responsible for it. They advised us to get the vaccine for our protection and the protection of our families and community. We should listen to their advice because they are knowledgeable and they of course do not wish any harm upon the people. (I8)

Lastly, data from the interviews revealed a different perspective between vaccinated and students with high intention to get the vaccine compared to those with vaccine hesitancy regarding viewing immunization as a social norm and a public health issue. While students with low intention to get the vaccine adopted an individualistic approach to the protection of health (Q11), vaccinated students (Q12) and those with a high intention to get the vaccine (Q13) felt that they had a responsibility for the benefit of the whole society:

Q11: (…) Of course there is a responsibility on other individuals in society to get the vaccine and so on, but I don’t know, I think it should also be based on individual choice. (I6)

Q12: I think that it’s important that everyone gets the vaccine as I told you, to protect himself and his family first and his community, and second to reduce the burden on health facilities and the country. (I8)

Q13: That depends on the individuals, it’s not really a side effect if we all cooperate, if we all take part in stopping the pandemic. We should be responsible for it. (I1)

Discussion

COVID-19 vaccination is critical for curtailing the pandemic and represents an important tool for a “return to normality.” To achieve herd immunity against COVID-19, vaccines need to be widely accepted by the public, including university students.²⁵ In this mixed-methods study, we explored COVID-19 vaccine acceptability in university students at one of the largest education institutions in the UAE, and found that one-third (31.8%) of students in our sample were hesitant toward COVID-19 vaccination. Previous research showed that COVID-19 vaccine hesitancy among university students ranged from 14% to 49% globally.^26–30 However, direct comparisons with these proportions should be made with caution, because these studies occurred before the vaccine was available.

The main factors that explained vaccine hesitancy in this study were having a lower perceived threat of COVID-19; less positive beliefs and attitudes toward the COVID-19 vaccine; considering the high adverse effects resulting from it; and perceiving difficulties in access to a vaccination center. These factors are in accordance with the WHO matrix model for vaccine hesitancy, where contextual and individual and group influences, along with vaccine-specific issues, overlap to explain an individual’s intention to get a vaccine. Previous research has shown an association between general positive attitudes and beliefs about the COVID-19 vaccine and the likelihood to take it.^21,31 Consistent with previous research, this study also highlighted the importance of perceiving the vaccine as effective, with students with low intention reporting uncertainty about the effectiveness of the vaccine in protecting against COVID-19.^32–34 Greater beliefs about adverse events associated with the vaccines, increased students’ reluctance to have the vaccine. The thematic content analysis suggested that media coverage increased uncertainty and that students did not know which sources to trust. Relying on social media for information has been shown to be significantly associated with lower intention to get COVID-19 vaccines compared to relying on information from medical doctors, scientists, and scientific journals.²⁷ Moreover, risk perception about the disease is another factor affecting vaccine acceptability: those who were highly concerned about being infected and considered that the risk of catching COVID-19 was high were less likely to refuse the vaccine, which is in line with previous studies.^33,35,36 These beliefs may explain how individuals deal with health threats because they affect, according to the CSM theory, the representation of the illness for the individuals, its attributed identity and the beliefs about the consequences of the disease, which are based on information obtained from personal experience as well as the opinions of significant others, health professionals and the media.⁷

The perception about immunization being a social norm was one of the factors that appeared to influence why some students decided to get the vaccine or not. Those who are concerned with being vaccinated to protect others are more likely to get the vaccine,³⁶ since they understand it as an act of solidarity for the common good. Solidarity emerges when individuals feel they have something in common with the others in a relevant aspect, which in the case of a pandemic, would be sharing a common threat.³⁷ This may also explain why participants in our study were more likely to be vaccinated or have a high vaccination intention if they lived with other family members. Furthermore, solidarity also exists at the level of the institution. When the individuals feel some level of (indirect) reciprocity, portrayed in our study as trust in government health policies, and that the duty of the state to protect the population is being upheld, they will be more involved in solidaristic arrangements.³⁷ However, for those individuals who are strongly hesitant about COVID-19 vaccines, solidarity may not be an effective motivator, and the provision of information on personal benefit would diminish hesitancy instead.³⁸

Previous studies have shown that students willing to have the vaccine had fewer concerns about potential side effects than the ones who refused it.^21,39 In our survey, vaccine side effects were mentioned by all groups of participants whether they had already had the vaccine, were intending to have it, or were hesitant about it. However, how individuals balance the benefits and risks of the vaccine is what determines their final decision:⁶ those with low intention consider the side effects of the vaccine to be more serious than those who have the intention to have it, and this outweighs the potential benefits of vaccination.

Lastly, hesitancy was also associated with a perception of difficulty accessing a vaccination center. However, if we look at “convenience” of taking the vaccine, defined by the WHO as ‘the extent to which physical availability, affordability, geographical accessibility, the degree to which vaccination services are delivered at a time and place and in a cultural context that is convenient and comfortable,’⁶ one might consider that the UAE is placing efforts on enhancing convenience of COVID-19 vaccination. These measures include making vaccines free-of-charge and distributing them in easily accessible locations in each city.⁵ The perception of low access to the vaccine by these individuals calls for the need to advertise better and provide public information about the vaccination program and how each individual may access the vaccine.

Strengths and limitations

This is the first study after a COVID-19 vaccine was available to the general population in the UAE, which makes our findings more likely to reflect students’ ‘real’ perceptions about the vaccine rather than their intentions about a hypothetical point in the future at which the vaccine is available. Furthermore, the use of mixed methods enabled an in-depth understanding of the factors associated with vaccine hesitancy. Finally, the use of attention checks in the survey increased the trustworthiness of the collected data. However, some limitations should also be acknowledged. This study was carried out in a single university, which may limit the generalizability of the findings, however, the University of Sharjah receives students from diverse regions of the country as well as international students from various countries. Thus, the sample in this study is perceived to resemble the national student population. Lastly, this is a cross-sectional study that prevents assessing causality, and thus, only associations between the predictor variables and vaccine uptake and intention were established.

Conclusion

This study identified several factors associated with COVID-19 vaccine hesitancy among university students, which can inform the development of strategies to promote vaccine uptake. Public health interventions are needed to tackle specific causes of vaccine hesitancy and involve healthcare professionals to reiterate the importance of vaccination in stopping the spread of COVID-19, the safety and efficacy of currently available vaccines, and the ease and availability of vaccination in the UAE.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Author contributions

The study was designed by HA and CS. HA, CS, BS and WS designed the analysis strategy and interpreted the data. CS and WS collected the data; CS performed statistical analysis; and CS and WS conducted qualitative analysis. HA, CS, BS, WS and AA wrote and revised the manuscript. SS contributed to the conception of the data collection instrument and to the interpretation of the data; revised the manuscript critically and made substantial contributions to it. EA contributed with critical revision of the manuscript. All authors revised and edited the manuscript critically, approved the version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Additional information

Funding

This study was supported by an internal grant from the University of Sharjah [operational grant no. 150316].