key: cord-0748519-619rbm1o
authors: Li, Guanjie; Zhong, Yanxu; Htet, Hein; Luo, Yunyan; Xie, Xizhuo; Wichaidit, Wit
title: COVID-19 Vaccine Acceptance and Associated Factors among Unvaccinated Workers at a Tertiary Hospital in Southern Thailand
date: 2022-03-02
journal: Health Serv Res Manag Epidemiol
DOI: 10.1177/23333928221083057
sha: 76f0a6369aece4d787f22f5189d5b1187350f490
doc_id: 748519
cord_uid: 619rbm1o

BACKGROUND: Hospital workers are at high risk of COVID-19 infection which is now vaccine-preventable. However, vaccine refusals also occur among hospital workers, but the associated factors have not been described. OBJECTIVES: To describe: (1) the level of COVID-19 vaccine acceptance, and; (2) the extent that history of pre-pandemic vaccine hesitancy and health beliefs regarding COVID-19 were associated with COVID-19 vaccine acceptance among workers at a tertiary hospital in southern Thailand. METHODS: We conducted a cross-sectional study using a paper-based self-administered questionnaire at a tertiary hospital in south Thailand in April 2021 and used multivariable logistic regression to identify psychological-behavioral factors associated with vaccine acceptance. RESULTS: Of 359 workers invited to participate, 226 participants returned the questionnaires, 67% of whom reported willingness to accept the vaccine. Vaccine acceptance was associated with perceived severity of disease (Adjusted OR = 2.07, 95% CI = 1.04, 4.10), perceived harm from non-vaccination (Adjusted OR = 2.51, 95% CI = 1.27, 4.96), and lower expectation of vaccine efficacy (Adjusted OR = 3.80, 95% CI = 1.87, 7.71). CONCLUSION: Most workers in this study were willing to accept the COVID-19 vaccine, and such acceptance was associated with components of the health belief model. However, the cross-sectional study design did not allow causal inference, and study data were all self-reported with no probing of the responses. These limitations should be considered as caveats in the interpretation of the study findings.

As of December, 2021, there have been more than 267 million confirmed cases of COVID-19, and more than 5 million deaths. 1 Hospital workers have higher exposure to COVID-19 compared to the general population. 2 Vaccination is considered to be the most effective form of prevention against COVID-19. Although hospital workers are prioritized for receiving COVID-19 vaccine, they are also known to refuse vaccination. [3] [4] [5] Information regarding vaccine acceptance among hospital workers and its associated factors is thus relevant to stakeholders in COVID-19 prevention and control.

Most quantitative studies on vaccine acceptance have used structured questionnaires with binary "Yes/No" responses [6] [7] [8] or Likert Scale multiple-choices questions on level of agreement with a given statement. 4, 5, 9, 10 Such measurement methods may not completely reflect the construct of vaccine acceptance in realworld situations. Furthermore, some studies only measured vaccine hesitancy, 5, 9, 11, 12 which covers a narrower spectrum of belief and decision than acceptance. 13 Studies on factors associated with COVID-19 vaccine acceptance showed that acceptance was associated with occupation, age, sex, 8, 9 working environment, policy and protective equipment, 3 as well as psychological factors including the "5C" (complacency, constraints, calculation, confidence, and collective responsibility), 14 and vaccine-related attributes (efficacy, protection duration, adverse effects, national origin of vaccine, approval by regulating authorities, and endorsement entity). 15 However, we have found no study that explored all of these factors simultaneously in order for the factors to account for one another's confounding effects in the assessment of their associations with vaccine acceptance.

Songklanagarind Hospital, Hat Yai, Thailand is a tertiary hospital with its capacity to accommodate up to 855 beds that serves as the referral facility for severe COVID-19 cases. Hospital workers are at high risk of exposure to COVID-19 infection. Assessment of prevalence of vaccine acceptance and related determinants among hospital workers are important to provide empirical evidence for relevant stakeholders in infection control and control of communicable diseases. The objectives of this study were to describe: (1) the level of COVID-19 vaccine acceptance, and; (2) the extent that history of prepandemic vaccine hesitancy and health beliefs regarding COVID-19 were associated with COVID-19 vaccine acceptance among workers at a tertiary hospital in southern Thailand.

We conducted a cross-sectional study using paper-based selfadministered questionnaires from 31st March to 28th April in 2021 at Songklanagarind Hospital, Hat Yai, Songkla Province, Thailand.

We included two populations: healthcare workers (HCWs, including doctor, allied health professional, nurse, and nursing assistants) at Songklanagarind Hospital who treated COVID-19 cases and non-medical workers (office staff, maintenance staff, housekeepers, and others). 16 Our inclusion criteria were: (1) being on the roster of staff at Songklanagarind Hospital; (2) aged 18 years and older, and; (3) functional literacy in Thai language. We excluded those who had already been vaccinated against COVID-19.

The Royal Thai Government started vaccine deployment at the end of February 2021, and healthcare staff at the study hospitals were informed of the COVID-19 vaccine's potential availability in March 2021, but were not informed which vaccine they were to receive. The vaccine was later confirmed to be Sinovac. However, very few workers had received the vaccine at the time of data collection; most workers were in the process of considering the vaccination offer. Furthermore, there was no vaccine mandate in Thailand at the time.

We assumed that the level of vaccine acceptance in this study was the same as in a previous study at 70% or higher. 17 At 5% margin of error and 95% level of confidence, we calculated the sample size using R and epicalc package 18 and obtained a sample size of 323 workers. Assuming 10% refusal to participate, the final sample size was 359 workers (n = 359).

We distributed the questionnaires to all 19 departments at the study hospital. The number of questionnaires sent to each department varied by probability proportional to size, 19 and can be found in Supplementary Table 2 .

Vaccine acceptance (outcome) refers to "a spectrum of behaviors and beliefs from rejection of all vaccines to active support of immunization recommendations". 20 Similar to previous studies, 4,10,21-24 multiple-choice questions on willingness to take the vaccine when available were adapted and changed into multiple binary questions.

Vaccine hesitation history refers to respondents' history on whether they ever refused, postponed or accepted non-COVID-19 vaccination in the past.

Psychological factors refer to attributes of respondents' psyche and thought process that could influence their vaccine acceptance, which were measured using Likert scale and based on the "5C" psychological factors (namely, complacency, constraints, calculation, confidence, and collective responsibility). 14 We modified questions from a previous study 14 to suit the context of COVID-19 pandemic and the socio-cultural context of the study setting.

COVID-19 related health beliefs refer to constructs defined by the Health Belief Model (HBM), 10 with modifications to suit the context of COVID-19 and COVID-19 vaccination, including perceived severity and susceptibility to COVID-19 and perceived benefits of COVID-19 vaccination.

Organizational factors refer to characteristics of the respondents' department or unit, including policy related to the policy of availability of personal protective equipment, isolation rooms, social distancing measures, hand hygiene, respiratory hygiene, disinfection measures, protective training, screening tests and triage system. 25 Vaccine attributes refer to characteristics of the vaccine with regards to a vaccine's efficacy, interval between two doses, adverse effects, approval by public health authorities, country of production/origin, adverse effect, affordability.

Demographic characteristics refer to the characteristics and personal attributes of the respondents, including gender, age in years, religion, occupation, and health problems that might counter-indicate vaccination.

The study instrument was a self-administered paper-based questionnaire with 48 items in total. Each questionnaire took approximately 15 min to complete. The questionnaire was developed under the supervision of a content matter expert who validated the questions during an iteration process. The questionnaire was developed in the English language then translated into Thai language. The research team back-translated the questionnaire from Thai into English to validate the Thai translation, and resolve parts where discrepancies between the original and back-translated versions were found. The research team pilottested the study instruments with 30 participants from the study hospital's Nursing Department from March 18th to March 24th, 2021. The pilot study participants were not included in the main study. An English version of the final questionnaire can be found in Supplemental Material 1.

The research team contacted the head of each department at the Hospital to inform them about the study and request their permission to collect data. A cover letter explaining the study's purpose with copies of the questionnaire and a cardboard box to the heads of each department of the hospital were sent and requested them to distribute the questionnaire to their staffs to participate into the study at their own discretion. In each questionnaire, a separate participant information sheet and an informed consent form (2 copies) were attached to be signed by the respondent and then need to send back.

The informed consent form included the project title, a summary of the purpose, procedure, risks and discomfort, benefits, confidentiality issues pertaining to the study, and contact information of the principal investigator and the Human Research Ethics Committee. The form also contained a space where the participants would confirm that they had given consent freely by signing their name, the printed name, and the date of signing the form. There was also a box where participant can check to declare that "[ ] I am willing to participate but do not wish to sign or write my name". At the bottom of the form, the investigators included a reminder "Please DO NOT remove this form from the questionnaire".

The completed questionnaires were placed in the cardboard box and requested to send back to the research team within the 14 business days. After receiving the completed questionnaire and informed consent form, investigators separated the informed consent forms from the questionnaire and stored the forms in a secure location.

Research team members performed double data entry of questionnaire responses and assessed validation using EpiData software version 3.1. Research team members then cleaned the data and exported the data to R. Background characteristics of the respondents and the primary outcomes of the study were summarized using mean, frequencies and proportions. We assessed the association between vaccine acceptance and history of past non-COVID-19 vaccine hesitancy, components of the health belief model, perceived organizational support, and opinions regarding vaccine attributes using multivariate logistic regression analyses, with the exposure variables adjusting for one another. We did not adjust for general characteristics of the study participants as they were not significantly associated with vaccine acceptance and thus did not fit the criteria for confounding factors (Supplementary Table 1 ).

This study received ethical approval from the Human Research Ethics Committee of the Faculty of Medicine, Prince of Songkla University (Approval Number: REC.64-090-18-2).

A total of 226 out of 359 hospital workers filled and returned the questionnaires (response rate = 63.0%). The mean age of the participants was less than 40 years (Table 1) . Most participants were female, Buddhists, and had no health problems that made them worried about COVID-19 vaccination. The distribution of the participants by department within the study hospital can be found in Supplementary Table 2 . Two-fifths of participants reported history of vaccine hesitation ( Table 2) . Nearly all participants considered the vaccine's country of origin, approval by authorities, and adverse effects after vaccination as important factors. With regard to vaccine-related opinions, although less than half of all participants agreed that the vaccines were completely safe, they also reported that they would vaccinate to protect people with a weaker immune system. With regard to health belief, just slightly more than half of the participants reported perceived severity of COVID-19 infected and perceived benefit of COVID-19 vaccination, although four-fifths agreed that the benefits of the vaccine outweighed the potential harms. Most participants had a favorable opinion of their hospital's effort to prevent nosocomial infection. Most participants were willing to receive COVID-19 vaccine and recommend it to others when the vaccine is available at the hospital for free (Table 3) . We considered approximately 67% of the participants to be vaccine acceptors, 18% to be refusers, and 15% to be inconclusive (ie, reported self-contradicting answers). Half of the participants indicated intention to delay vaccination, with the common concerns being adverse effects, preference to wait for more effective vaccines, preference to wait for complete trials results, and preference for those who were more vulnerable to receive the vaccine first. When presented with facts about Sinovac and AstraZeneca vaccines, most participants answered that they would prefer Sinovac, although one-fourth were unsure.

Multivariate logistic regression analyses (Table 4) showed that after adjusting covariables, participants who indicated perceived severity were 2.07 times more likely to accept the vaccines than participants who did not. Participants who perceived harm from non-vaccination (ie, disagreed that adverse effects outweigh the benefits) were 2.51 times more likely to accept the vaccine than those without perceived harm. Participants who did not expect 100% vaccine efficacy were 3.8 times more likely to accept the vaccine than participants who did.

Among workers at the study hospital, we found that two-thirds of the participants would accept the COVID-19 vaccine and factors associated with vaccination included perceived severity of COVID-19, perceived harms of non-vaccination, and not expecting vaccine to be 100% effective. The findings of our study can potentially provide empirical evidence to support stakeholders in the decision-making process.

The prevalence of vaccine acceptance among healthcare workers (doctors, nurses, nurse assistants) was higher than nonmedical workers, which was similar with the studies conducted in Israel and France, but was much higher than studies in Hong Kong, Congo and Malta. 26 HCWs are at a higher risk of infection than non-medical workers, 2 which could have accounted for the higher vaccine acceptance. Another reason might be that HCWs might have gained more knowledge about COVID-19 and its effects on human health, so they might have been more willing to accept the vaccine when it became available. Although we excluded those who had already received the COVID-19 vaccine from our study, such vaccination was very uncommon during the study period. Private vaccination was not available in early 2021, and organizations in the Royal Thai Government had to follow strict rules and procedures to distribute the vaccine. In this context, it was unlikely that the exclusion of those who had received vaccination introduced selection bias to our study findings. However, there was an unexpected outbreak of COVID-19 in the local area at the time of our data collection, which delayed the completion of our data collection period by 2 additional weeks. These events could have influenced the level of vaccine acceptance among our participants, and the findings of this study might have limited generalizability to other contexts. Two parts of the Health Belief Model were associated with COVID-19 vaccine acceptance: perceived severity of COVID-19 and perceived harm from non-vaccination. The association with perceived severity was consistent with a populationbased survey in Hong Kong ,27 but not in China . 28 The association between vaccine acceptance and perceived harm from nonvaccination was consistent with the finding of other studies. 10, 15 Similarly, the association between expected vaccine efficacy and vaccine acceptance in our study was similar to that of a previous study ,29 although our method of categorization might have differed from others 15 and could have limited the comparability of our study findings.

One point of consideration was that we measured the exposure (ie, expected efficacy) and outcome (ie, vaccine acceptance) in our study using a relatively indirect methods, ie, a multiple choice questions with 5 possible responses for the expected efficacy of COVID-19 vaccine and multiple yes/no questions to measure vaccine acceptance, while a previous study 15 measured vaccine acceptance using scenarios with randomized vaccine attributes. Future studies should consider incorporating techniques from other studies that might yield more detailed data.

Our study had several strengths. Firstly, we used proportionate sampling, which improved the representativeness of the responses to the population that gave rise to our participants. Secondly, we used paper-and-pencil questionnaires, which did not preclude participation by hospital workers who had limited internet access or were concerned about their privacy like in previous studies.

A number of limitations should be considered in the interpretation of the study findings. Firstly, the study was a crosssectional, thus causality could not be inferred. Secondly, we used self-administered questionnaire, which did not allow for probing of the participants' responses, limiting the amount of details available. Thirdly, fewer than 90% of the delivered questionnaires were returned (ie, response rate was less than 90%), thus the potential selection bias from non-response was non-negligible. Lastly, our study data were collected during a new wave of COVID-19 outbreak and at a time when hospital workers were informed that they were to receive a vaccine. The findings of this study might have limited generalizability to other contexts.

In this cross-sectional study, we assessed the level of COVID-19 vaccine acceptance and its association with components of health belief model and other attributes among workers at a tertiary hospital in southern Thailand. We found that most hospital workers would like to accept the COVID-19 vaccine, and that perceived severity of COVID-19 infection, perceived harm of non-vaccination, and more lenient expectation of vaccine efficacy were significantly associated with vaccine acceptance. However, the cross-sectional study design and issues pertaining to measurements should be considered as caveats in the interpretation of the study findings.

World Health Organization. WHO Coronavirus (COVID-19) dashboard

Risk of COVID-19 among front-line health-care workers and the general community: a prospective cohort study

Influenza vaccine uptake, COVID-19 vaccination intention and vaccine hesitancy among nurses: a survey

Attitudes toward a potential SARS-CoV-2 vaccine: a survey of U.S. Adults

Cross-sectional assessment of COVID-19 vaccine acceptance among health care workers in Los Angeles

Intention to get vaccinations against COVID-19 in French healthcare workers during the first pandemic wave: a cross sectional survey

Caregiver and service provider vaccine confidence following the Changchun Changsheng vaccine incident in China: a cross-sectional mixed methods study

Attitudes towards influenza vaccine and a potential COVID-19 vaccine in Italy and differences across occupational groups

Intention to participate in a COVID-19 vaccine clinical trial and to get vaccinated against COVID-19 in France during the pandemic

The use of the health belief model to assess predictors of intent to receive the COVID-19 vaccine and willingness to pay

Vaccine hesitancy" among university students in Italy during the COVID-19 pandemic

COVID-19 vaccine hesitancy among medical students

Vaccine hesitancy: definition, scope and determinants

Beyond confidence: development of a measure assessing the 5C psychological antecedents of vaccination

Factors associated with US adults' likelihood of accepting COVID-19 vaccination

Emerging respiratory viruses, including COVID-19: methods for detection, prevention, response and control

COVID-19 vaccine hesitancy worldwide: A systematic review of vaccine acceptance rates

Feemster KA. Special focus vaccine acceptance

Acceptability of vaccination against COVID-19 among healthcare workers in the Democratic Republic of the Congo

A global survey of potential acceptance of a COVID-19 vaccine

COVID-19 vaccine acceptance among health care workers in the United States. Vaccines (Basel)

Acceptance of COVID-19 vaccination during the COVID-19 pandemic in China. Vaccines (Basel)

COVID-19 Vaccine hesitancy worldwide: a concise systematic review of vaccine acceptance rates. Vaccines (Basel)

Acceptance of the COVID-19 vaccine based on the health belief model: a populationbased survey in Hong Kong

Determinants of influenza vaccination among young Taiwanese children

Acceptance and preference for COVID-19 vaccination in health-care workers (HCWs)

We would like to thank all participants for their time. Our thanks also go to the staff of the Department of Epidemiology for coordinating research activities. We would like to thank Prof. Virasakdi Chongsuvivatwong for his guidance throughout the research process.

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Guanjie Li is a second-year student in the Master of Science