Future COVID-19 Booster Vaccine Refusal in Healthcare Workers after a Massive Breakthrough Infection Wave, a Nationwide Survey-Based Study

Background: An unprecedented coronavirus disease 2019 (COVID-19) wave occurred in China between December 2022 and January 2023, challenging the efficacy of the primary series of COVID-19 vaccines. The attitudes toward future COVID-19 booster vaccines (CBV) after the massive breakthrough infection among healthcare workers remain unknown. This study aimed to explore the prevalence and determinants of future CBV refusal after the unprecedented COVID-19 wave among healthcare workers. Methods: Between 9 and 19 February 2023, a cross-sectional nationwide online survey was conducted using a self-administered questionnaire vaccine among healthcare workers in China. Sociodemographics, profession, presence of chronic medical conditions, previous COVID-19 infection, attitudes towards future CBV, and reasons for future CBV refusal were collected. We estimated odds ratio [OR] with 95% confidence interval [CI] using a multivariable logistic regression model to explore the factors associated with future CBV refusal. Results: Among the 1618 participants who completed the survey, 1511 respondents with two or more doses of COVID-19 vaccines were analyzed. A total of 648 (41.8%) of respondents were unwilling to receive a future CBV. Multivariable logistic regression analysis revealed the association of CBV refusal with profession (vs. other staff, physician-adjusted OR 1.17, 95%CI 0.79–1.72, nurse-adjusted OR 1.88, 95%CI 1.24−2.85, p = 0.008), history of allergy (adjusted OR 1.72, 95%CI 1.05–2.83, p = 0.032), a lower self-perceived risk of future COVID-19 infection (p < 0.001), and a lower belief in CBV effectiveness (p = 0.014), safety (p < 0.001), and necessities for healthcare workers and the public (p < 0.001, respectively). Conclusions: Our findings highlight that a considerable proportion of healthcare workers were against a future booster dose after an unprecedented COVID-19 wave. Self-perception of future COVID-19 risk, and potential harm or doubtful efficacy of vaccines are the main determinants. Our findings may help public health authorities to establish future COVID-19 vaccination programs.


Introduction
Several types of vaccines have been introduced to prevent and control the coronavirus disease 2019 (COVID-19) pandemic caused by severe acute respiratory syndrome coronavirus (SARS-CoV-2) infection [1][2][3]. Generally, achieving adequate immunity requires a two-dose series of SARS-CoV-2 vaccines. However, accumulating evidence showed that the new Omicron virus variants could escape most therapeutic monoclonal antibodies and vaccine-elicited antibodies [4][5][6]. Substantial concerns have been raised regarding the duration of immunity after the primary two doses and potential breakthrough infection, suggesting a need for a COVID-19 booster vaccine (CBV) [7]. Some researchers argued that a fourth-dose booster might help maintain a high serum-neutralizing antibody level to protect against infection [8,9].
Since the introduction of the COVID-19 vaccine, nearly 3.5 billion doses of vaccines over different platforms have been administrated in China until 23 December 2022 [10]. Several previous observational studies showed that mild-to-moderate infections accounted for the majority of COVID-19 infection in China, 2022 [11-13]. Moreover, vaccination coverage has reached more than 85% in Mainland China, and each person has been inoculated with 2.2 doses on average [13]. Shortly after China moved away from its zero-COVID policy in late December 2022, an unprecedented COVID-19 wave spread in China. After reaching the peak (6.94 million) on 22 December 2022, it gradually decreased to 24,000 on 30 January 2023 [14]. Until 9 February 2023, there were 98,742 hospitalized COVID-19 patients, 7918 of whom experienced severe infection [15]. Healthcare workers are at a particularly high risk of SARS-CoV-2 infection, and their beliefs and attitudes toward CBV are extremely important for primary prevention strategies. Although some studies reported the attitudes toward CBV among healthcare workers [16,17], attitudes toward CBV may change over time because of the ever-changing public's perception of COVID-19 risk and vaccination safety and efficacy. Data from Singaporean public primary care clinics showed that vaccine hesitancy among healthcare workers had improved from the first dose to CBV [18]. A cross-sectional study at the beginning of the fourth COVID-19 vaccination dose campaign in Israel showed that 53.9% of physicians were unwilling to be vaccinated compared to 83.3% of nurses and 69% of other healthcare professions, with the most frequent concerns regarding the vaccine being its efficacy, benefit, and necessity [19]. In the present study, we aimed to explore the prevalence and determinants of future CBV refusal at this particular time point.

Study Design and Participants
We conducted an online survey-based cross-sectional study between 9 and 19 February 2023, aiming to measure current attitudes towards future CBV among Mainland China healthcare workers. We also explored the factors that were associated with the future CBV refusal. This study was performed shortly after an unprecedented nationwide COVID-19 wave in Mainland China between December 2022 and January 2023 [14]. We utilized a self-administered questionnaire disseminated online through an app (Wen Juan Xing, Changsha Ranxing Information Technology Co., Ltd., Hunan, China) for data collection. Responses from participants were collected voluntarily and anonymously, with no personal data. We reported this study according to the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) statement [20] and the American Association for Public Opinion Research (AAPOR) reporting guidelines for survey studies [21]. A minimum sample size of 1068 was calculated using an online sample size calculator (Rao soft, http://www.raosoft.com/samplesize.html (accessed on 31 January 2023)), with a 3% margin of error, 95% confidence interval, 50% response distribution, and targeted population size of 14 million [22], which was further inflated by 10% (n = 107), resulting in a final sample size of 1175. Invitations were sent to hospital-based and professional medical society-based networks, which then forwarded the invitations to their members, approximately 3000 individuals, to facilitate participation in the study. Agreeable participants may also invite other potential participants through their respective networks [16].

Questionnaire Design
Survey questions were developed by experts in primary care, disease control, infectious disease, and qualitative research, after referring to available systematic reviews [23][24][25].
The preliminary questionnaire focuses on previous COVID-19 infection and vaccination, self-perception of future COVID-19 risk, and attitudes towards a future CBV as well as its potential factors. The relevance, clarity, simplicity, and ambiguity of each question item were assessed using an ordinal four-point Likert scale (1 = not relevant, 2 = relevant but needs major revision, 3 = relevant but needs minor revision, 4 = very relevant) [24]. The content validity index was 0.88, calculated by the proportion of a score of 3 or 4 regarding each item content valid [26,27]. The improved questionnaire was administered in face-to-face interviews with 10 volunteer healthcare workers from different departments to determine whether the question items were well-understood. Five other healthcare workers completed the questionnaire twice at an interval of 10 days to test the reliability (the reliability coefficient was 0.898), with a median completion time of five minutes. The validity test of the questionnaire indicates good internal consistency (Cronbach's alpha 0.809, KMO 0.782, Bartlett's test of sphericity, p < 0.001). The final version of this questionnaire includes five domains (eMethods in Supplementary Materials). The first domain aims to collect sociodemographics (age, sex, marital status, profession, etc.) and health status variables. The second domain includes seven items regarding the previous COVID-19 infection, including onset symptoms (i.e., fever, cough) and disease severity (i.e., need for hospitalization or intensive care unit). The third section includes four items regarding COVID-19 vaccination history, including types of vaccines, number of doses, and adverse effects. The fourth domain aims to assess participants' attitudes that could be associated with a future CBV refusal, addressing the perceived risk of future COVID-19 and the perceived benefits (preventing infection, preventing severe infection, etc.) and barriers (safety) of a CBV [28]. Responses for these questions were assessed using a three-point scale, with 1 = disagree, 2 = uncertain, and 3 = agree.

Outcomes
The first outcome was the refusal to receive a future CBV (yes/no) [29]. We explored the factors associated with the future CBV refusal. Secondary outcomes were the preference of the future CBV types and the drivers for the CBV acceptance and refusal. Additional outcomes included the prevalence and severity of previous COVID-19 infection.

Statistical Analysis
Continuous data were summarized as mean and standard variations [SD] or median with interquartile range (IQR) and categorical variables with frequencies with percentage, where appropriate. Missing data of weight (n = 16) and height (n = 22) were imputed using a conditional specification method while taking into account the joint distribution of age and sex. There were no missing data for other variables. We compared between-groups differences in continuous variables using the t-test or the Mann-Whitney test, and categorical variables using the Chi-square test or Fisher's exact test as appropriate. A multivariable logistic regression model was used to evaluate the adjusted odds ratio (OR) of the CBV refusal for each variable with a p value < 0.2 in the univariable analysis. We conducted two sensitivity analyses, by limiting the sample to those who had previous COVID-19 or excluding those who had received the fourth dose. All analyses were conducted using SPSS Statistics Version 25.0 (IBM, Chicago, IL, USA), and significance was set at α = 0.05.

Acceptance and Refusal of Future CBV
A total of 648 (41.8%) participants were unwilling to receive a future booster dose. Table 2 Table 3). Regarding the attitudes towards the future CBV, participants in the refusal group were less likely to agree "CBV could terminate the COVID-19 pandemic" (21 [ Table 3).

Discussion
This is the first study, to our knowledge, to explore the prevalence and associated factors of future CBV refusal after an unprecedented COVID-19 wave in fully vaccinated healthcare workers in China. Our study has several important findings. First, nearly half of the respondents were against a future CBV. Second, CBV refusal was significantly associated with profession (nurse), history of allergy, a lower self-perceived risk of future infection, and a lower belief in CBV effectiveness, safety, and necessities for healthcare workers and the public.
Compared to findings of available literatures [30,31], the prevalence of previous COVID-19 infection was higher in our participants. Thus, such a high infection prevalence in our participants may indicate massive breakthrough infections among fully vaccinated healthcare workers in Mainland China. One possible explanation is that the Omicron variants can escape antibody neutralization from the serum of both previously infected and vaccinated individuals [32,33]. Previous studies showed that the protective effect of the third dose only lasted for a short time against Omicron variants [34,35]. Moreover, data from Israel showed that the fourth-dose mRNA vaccine had short-term benefits in reducing the odds of severe COVID-19, hospitalization, and mortality in old individuals who had received the third dose more than 4 months previously [36,37]. Additionally, among triple-vaccinated healthcare workers who had a previous COVID-19 infection, the boosting effect of Omicron is minimal [38]. The aforementioned findings may suggest the potential need for a future booster dose in high-risk populations.
Nearly half of our respondents indicated a refusal to receive a future CBV. In contrast to our data, several previous literatures reported high vaccination acceptance in healthcare workers [39,40]. A previous study in China showed that 87% of respondents were willing to receive a CBV [40]. A review paper showed that the prevalence of hesitancy for COVID-19 vaccination varied from 4.3% to 72% among healthcare workers worldwide [41]. Possible explanations for the disparities might include policy and socio-cultural differences and different time points [39,42,43]. Due to the issues of methodology (for example, imbalanced response options) and self-selection of respondents, the prevalence of refusal is likely to be overestimated.
Our multivariable regression analyses showed that respondents who refused to receive a future CBV more likely disagreed with its necessity and safety. Such a high proportion (91%) of our vaccinated participants reported previous COVID-19 infection; thus, the experience of potential poor protective effect of vaccination could be a factor influencing the willingness to receive a future CBV. Concerns about effectiveness (32%) and possible long-term adverse effects (31%) accounted for the top reasons of the CBV refusal among Indian and Saudi Arabian healthcare workers [29]. Moreover, a cross-sectional study in China showed that post-vaccination adverse effects were associated with a lower willingness to receive a booster dose (OR = 0.37, 95%CI: 0.23-0.59) [40]. More than half (54.7%) of our respondents preferred inactivated vaccines as a future CBV shot. This might be explained by the low prevalence (0.2%) of severe adverse effects reported following inactivated vaccination in our participants. All of the aforementioned findings suggest that addressing concerns about vaccination efficacy and safety should be further incorporated into communication campaigns and healthcare systems to improve CBV acceptance and uptake intent.
Our regression analyses showed that nurses had higher odds of a future CBV refusal compared to other staff. Our results are comparable to findings of other studies that showed vaccination intention varies with profession [44][45][46][47]. Data from Singapore showed that hesitancy in nurses (15 [7.4%]) was higher than for staff in medical (0%) and applied health (5 [3.5%]) [45] professions. A cross-sectional study among Israel healthcare workers showed that 83.3% of nurses were against the fourth dose [19]. Possible explanations might include more concerns about vaccine effectiveness and safety, and less knowledge about the vaccine among nurses [48], since nurses were less likely to have a higher education background among the different professions of our participants (Table S6). Our findings suggest that immediate attention is needed on the strategies to promote the uptake of CBV for nurses.
Our data showed an association of CBV refusal with a lower perceived risk of future COVID-19 and a lower belief in CBV effectiveness. Our findings were comparable to those of previous studies. A previous survey showed that participants who agreed that the CBV could control severe COVID-19 (OR 3.18, 95%CI 2.00-5.05) and symptomatic SARS-CoV-2 infection (OR 2.77, 95%CI 1.81-4.23) were more likely to accept a booster dose [44]. Moreover, the top three reasons for COVID-19 vaccine acceptance among healthcare workers in Singapore were "To protect themselves from COVID-19", "To protect their family and friends", and "Due to high risk of acquiring COVID-19 because of their jobs" [45]. A scoping review showed that participants who had direct contact with COVID-19 patients or a higher self-perceived infection risk were predictors for a higher vaccine acceptance among healthcare workers [41].

Strengths and Limitations
Our study was included a high proportion of vaccinated participants who also reported previous COVID-19 infection. Moreover, the harvested sample from the anonymous online survey has an optimal size and decent representativeness to reflect several essential characteristics of the target population (i.e., sex, marital status, and professional groups). We acknowledge limitations. First, this cross-sectional survey with a non-random sample strategy inevitably leads to selection bias and self-reporting bias. For example, our participants were relatively young, partly because younger healthcare workers may be more willing to participate in the online survey. Additionally, those with strong views towards COVID-19 vaccination are more likely to respond to a survey. Thus, the sample may not be fully representative. Moreover, selection bias might also be introduced since respondents from the same network may share similar opinions and viewpoints. However, we did not have information on potential clusters and were thus unable to adjust for this in our analysis. Second, data regarding the proportion of respondents agreeing or refusing to the future CBV among the different society networks were unavailable. Third, this study used a self-administered questionnaire, and the results were not externally validated. Fourth, our findings may not be generalized to other populations since healthcare workers are more aware of the risk-benefit profile of COVID-19 vaccines. The generalization and external validity should be investigated in further studies. Lastly, the response rate could not be calculated in this online survey [49].

Conclusions
This study identified several determinants for the future boost dose refusal among healthcare workers in Mainland China, including profession, a lower self-perceived risk of future COVID-19 infection, and a lower belief in CBV effectiveness, safety, and necessity. Healthcare workers might have a role model effect on the public; thus, a comprehensive understanding their motivations for CBV refusal might be informative to future vaccination campaign strategies. The present study may help public health authorities to establish future COVID-19 vaccination programs.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/vaccines11050987/s1. eMethods: Questionnaire. Figure S1: Selfreported COVID-19 symptoms. Figure S2: Administered COVID-19 vaccines. Figure S3: Post-COVID-19 vaccination adverse effects. Table S1: Demographics among participants with different doses. Table S2. Adverse (or side) effects after previous COVID-19 vaccine types. Table S3: Sensitivity analysis limited to those with previous COVID-19. Table S4: Sensitivity analysis excluding those who had received the fourth dose. Table S5: Reasons for future COVID-19 booster vaccine refusal. Table S6: Demographics and characteristics among participants with different profession. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors have read and agreed to the published version of the manuscript.
Funding: This study was supported by the Fujian Provincial Key Clinical Specialty of Neurology (05029001). The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Institutional Review Board Statement:
The study protocol was reviewed and approved by the Fujian Medical University Union Hospital Ethics Committees (2023KY012). All clinical investigations were conducted in accordance with the principles expressed in the declaration of Helsinki. Participants were free to quit the survey at any time without justifying their decision, and no data was saved before the questionnaire was completed and submitted with confirmation.
Informed Consent Statement: Electronic informed consent was obtained when the respondents completed and submitted their responses of questionnaires.

Data Availability Statement:
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.