COVID-19 vaccine hesitancy worldwide and its associated factors: a systematic review and meta-analysis

Introduction The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic has taken a toll on humans, and the development of effective vaccines has been a promising tool to end the pandemic. However, for a vaccination program to be successful, a considerable proportion of the community must be vaccinated. Hence, public acceptance of coronavirus disease 2019 (COVID-19) vaccines has become the key to controlling the pandemic. Recent studies have shown vaccine hesitancy increasing over time. This systematic review aims to evaluate the COVID-19 vaccine hesitancy rate and related factors in different communities. Method A comprehensive search was performed in MEDLINE (via PubMed), Scopus, and Web of Science from January 1, 2019 to January 31, 2022. All relevant descriptive and observational studies (cross-sectional and longitudinal) on vaccine hesitancy and acceptance were included in this systematic review. In the meta-analysis, odds ratio (OR) was used to assess the effects of population characteristics on vaccine hesitancy, and event rate (acceptance rate) was the effect measure for overall acceptance. Publication bias was assessed using the funnel plot, Egger's test, and trim-and-fill methods. Result A total of 135 out of 6,417 studies were included after screening. A meta-analysis of 114 studies, including 849,911 participants, showed an overall acceptance rate of 63.1%. In addition, men, married individuals, educated people, those with a history of flu vaccination, those with higher income levels, those with comorbidities, and people living in urban areas were less hesitant. Conclusion Increasing public awareness of the importance of COVID-19 vaccines in overcoming the pandemic is crucial. Being men, living in an urban region, being married or educated, having a history of influenza vaccination, having a higher level of income status, and having a history of comorbidities are associated with higher COVID-19 vaccine acceptance.


Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) first emerged in 2020 Pandemic in December 2019 and soon became a global concern owing to its high transmissibility; it was announced as a pandemic by the World Health Organization (WHO) in March 2020.Various strategies have been considered to prevent further transmission.Obligatory wearing of face masks, social distancing, travel restrictions, lockdowns, and quarantine were the first steps taken by most countries worldwide.Despite being partially successful in limiting further disease transmission, these strategies resulted in tremendous economic devastation [1,2].
The constant emergence of new variants of SARS-CoV-2, which are either more transmissible or cause greater morbidity and mortality, further raised concerns about a more cost-effective solution.Therefore, ever since vaccines became available and approved for use at the end of 2020, they have been considered the most effective and crucial strategy to fight coronavirus disease 2019 (COVID-19) [3][4][5].
Nevertheless, for a vaccination program to be successful, a considerable proportion of the community must be vaccinated.At least 70% of each community must be fully vaccinated to achieve herd immunity, and this number may be greater based on the vaccine type and transmissibility of the circulating variants [4,6,7].Hence, public acceptance of COVID-19 vaccines has become the key to controlling the pandemic.It is therefore important to ensure that maximum vaccine coverage is reached by making vaccines accessible and affordable and increasing public awareness to achieve maximum vaccine acceptance [8].
However, studies have shown that the rate of vaccine hesitancy has increased over time, making it the most important concern in the fight against COVID-19.The WHO declared vaccine hesitancy, defined as the refusal to get vaccinated despite the availability of vaccines, as one of the top 10 global health threats in 2019.This growing hesitancy may be because of an altered perception of the disease risk, uncertainty about available vaccines, fear of side effects, misinformation, and the spread of fake news [3,8,9].
Perception of health risk is strongly associated with vaccine hesitancy.Consequently, for public health to improve people's knowledge and attitudes toward vaccination, it is necessary to first understand this phenomenon's social, demographic, and psychological determinants.Furthermore, the language and communication strategies or media used to convey a health message influence how the vaccine is received.To accomplish this, all authorities involved in health communication must work together to produce clear and coherent messages [10].
With vaccines being the most important and effective weapon in the battle against COVID-19, it is essential to address the factors contributing to vaccine hesitancy and attempt to increase the rate of vaccine acceptance in the community.Herein, a systematic review was performed to detect the intention to receive COVID-19 vaccines among different communities and identify different population characteristics and factors associated with COVID-19 vaccine hesitancy.

Eligibility criteria
This systematic review included all relevant descriptive and observational studies (cross-sectional and longitudinal) on vaccine hesitancy and acceptance.No time constraints for studying or publishing articles nor restrictions on the population were imposed.Non-English studies, studies without full-text access, and those not relevant to vaccine hesitancy or acceptance were excluded.Narrative reviews, systematic reviews, meta-analyses, editorials, commentaries, letters to the editor, unpublished data, books, and conference papers were also excluded.

Study selection
After searching the databases, all retrieved records were screened for inclusion by reviewing the title/abstract and full text based on the eligibility criteria.Six authors (MB, FF, FG, HB, RR, and FS) performed both title/abstract and full-text screening, such that every article was reviewed by two independent reviewers.They resolved any disagreements by consulting a third reviewer (AL, NS, or MA).

Data extraction and analysis
One reviewer (FG, HB, RR, or FS) extracted the relevant data from the included papers, which were then rechecked and confirmed by another reviewer (AL, NS, or MA).The following data were extracted for each study: title, first author's name, date of study (year and month), study design, number of respondents/participants, age groups, gender, race/ ethnicity, religion, marital status, country, metropolitan classification (rural or urban), income, insurance status, education, occupation/ employment status, work setting (high-risk or non-high-risk), presence of any disease/chronic situation/history of comorbidities (physical/psychiatric), ongoing treatments, smoking status/alcohol consumption, mistrust in the government/healthcare system, received training on COVID-19 prevention, contact with confirmed/suspected COVID-19 patients, history of COVID-19 diagnosis, lost someone from COVID-19, health believes on COVID-19 (perceived susceptibility, severity, benefits, barriers, cues to action, etc.), being informed about COVID-19 vaccines, vaccination-related intentions, parents' willingness and hesitancy toward children's vaccination, COVID-19 vaccination status (not vaccinated, 1 dose, etc.), vaccine hesitancy, willingness to pay for vaccination,  Hesitancy, the primary outcome of the study, was considered as any reluctance, delay, or doubt in acceptance, and also refusal of the COVID-19 vaccines.Acceptance was defined as already vaccinated or willing to accept COVID-19 vaccines in the future without any doubt.For studies in Fig. 2. Forest plot displaying the comparison of COVID-19 vaccine hesitancy between women and men.which only hesitancy or acceptance was reported, the other was calculated by subtracting the total number of respondents from the reported outcome.If a study reported hesitancy and acceptance as two different variables (i.e., measured using two different questionnaires), acceptance was calculated by subtracting the number of hesitant respondents from the total respondents.To measure the effects of age and income on vaccine hesitancy, the highest category of each variable was reported in the included studies and compared with the lowest category.Odds ratio Fig. 3. Forest plot displaying the comparison of COVID-19 vaccine hesitancy between older and younger people.
(OR) was used to assess the effects of population characteristics on vaccine hesitancy, and event rate (acceptance rate) was the effect measure for overall acceptance.Additionally, 95% confidence intervals (CIs) were calculated for both measures.Crude data were extracted when available; otherwise, ORs were calculated.Owing to the heterogeneity in the variables included in the regression models of different studies, only univariate ORs were extracted for use in the meta-analysis.The randomeffects model was used when heterogeneity was more than 50% (I 2 > 50%).Publication bias was visually assessed using funnel plots and Egger's test.The trim-and-fill method was used to impute the missing studies and adjust for the effects of publication bias.
All meta-analyses were performed using Comprehensive Meta-Analysis, Version 2.2 (CMA; Biostat Inc., Englewood, NJ, USA).The acceptance rate was plotted on the Earth map using Python version 3. The studies were ordered alphabetically in all forest plots.

Study characteristics
A total of 6,417 articles (PubMed ¼ 1,548, Web of Science ¼ 1,077, and Scopus ¼ 3,792) were retrieved from the database search.After removing 2,369 duplicated records, 4,048 remained, which were assessed for eligibility using title/abstract and full-text screening.Finally, 135 studies were included in this analysis.Fig. 1 shows the identification process of the included studies according to the PRISMA 2009 flow diagram [12].Table 1 shows a summary of the included studies.
A meta-analysis of 114 studies encompassing 849,911 participants showed an overall acceptance rate of 63.1% (59.3-66.7%;Fig. 12).Moreover, Fig. 13 shows the acceptance rate by country, and Fig. 14 shows a map of the acceptance rate worldwide.
All analyses were performed using a random-effects model due to significant differences observed in the design, setting, and population of the included studies.In addition, the (I 2 ) for all analyses was greater than confirming heterogeneity among the included studies.Furthermore, applying Duval and Tweedie's trim-and-fill method altered the results for age (OR ¼ 0.90; 95% CI¼ 0.73-1.10)and healthcare workers (OR ¼ 0.95; 95% CI¼ 0.82-1.11).Funnel plots and other publication bias tests are shown in the Supplementary Material (Figs.S1-S11 and Table S1).

Discussion
Vaccine uptake rate plays a significant role in achieving herd immunity against COVID-19.The basic reproductive number of an infectious disease is used to calculate the level of population immunity required to limit the spread [141].According to the most recent COVID-19 estimates, a population of 60-75% immune individuals is necessary to prevent the virus from spreading further and infecting the community [142][143][144].
Three factors influence vaccination acceptance: complacency, confidence, and convenience [145].Complacency refers to the belief that the risk of developing a specific disease is low, making immunization unnecessary and avoidable [146,147].The level of faith and trust in the safety and effectiveness of vaccination is referred to as confidence.The comfort afforded by the population in terms of vaccine accessibility, price, and availability is referred to as convenience [146].
The findings showed that reasons for hesitation are more frequently associated with distrust of medical authorities and vaccine safety.Other factors related to the perception of health risks, such as fear of consequences and lack of information, are also important in vaccine hesitancy.Hence, future vaccination campaigns should emphasize the importance of the individual and include activities aimed at increasing their health knowledge.These actions should be performed at all levels of the healthcare system to increase awareness and trust.
The rapid development of effective and safe COVID-19 vaccines was unprecedented [148][149][150][151]. Nonetheless, COVID-19 vaccine apprehension could be a stumbling block in worldwide attempts to contain the pandemic's harmful health and socioeconomic consequences [152][153][154].The cost, effectiveness, and duration of protection provided by vaccines appear to be important factors in achieving this goal [150,155,156]; however, vaccine reluctance could be a major obstacle in successfully controlling the COVID-19 outbreak [35].
Consequently, estimates of vaccine acceptance rates can help plan actions and intervention measures to raise public awareness and reassure people about the safety and benefits of vaccines, which can help control the virus' spread and mitigate the negative effects of this unprecedented pandemic [157,158]  [153].These findings are partially in accordance with those of Danis et al. , who found that economic hardship was a driver of vaccination reluctance; however, no link was found between financial hardship and vaccine rejection [159].By contrast, parental education was a valid predictor of vaccination refusal in both mothers and fathers, whereas reluctance appeared to be unaffected by parental education.In addition, Black and African populations had lower acceptance rates in our study-a finding consistent with another study that found a higher level of skepticism and anxiety regarding the flu vaccine among African Americans [159].In contrast, our analysis indicates that income does have an impact on vaccination attitudes, with the high-income population showing lower COVID-19 vaccine hesitancy than the low-income population.
Vaccine acceptability among healthcare personnel yielded mixed findings.In general, healthcare workers had higher acceptance; however, Dror et al. found no significant differences in vaccine acceptance between healthcare and non-healthcare personnel in their study, and Barello et al. found no significant differences between healthcare and non-healthcare students [160,161].Our analysis found a statistically significant difference in COVID-19 vaccine hesitancy between healthcare and non-healthcare workers, with healthcare workers showing less hesitancy than non-healthcare workers.The impact of political ideology on vaccination acceptance or rejection has been one of the most intriguing aspects in some studies; for instance, Kennedy et al. conducted a study focusing on populist parties, finding that-at least in the Western European setting-populist party support might be used as a proxy for vaccination reluctance [162].
The constant advancement in technology suggests that the future of healthcare will be integrated with technology.Therefore, to combat vaccine hesitancy, it is critical to promote population-based communication and information strategies.These strategies include forging multidisciplinary alliances among healthcare providers, providing medical and scientific communications on vaccination, sharing recent data and shreds of evidence on virtual media or brochures, and increasing opportunities for dialogue and counseling regarding vaccination [10,163].
Finally, when the effects of gender and age on COVID-19 vaccination apprehension were examined, it was found that men were more likely to be immunized against COVID-19.This may be because of their stronger perception of COVID-19 hazards and weaker beliefs in diseaserelated conspiracies [164][165][166].As sampling bias-particularly in gender distribution-might alter the reported rates, these variables should be addressed for the appropriate interpretation of COVID-19 acceptance rates.According to our review, men have a higher acceptance rate than women.This finding is consistent with previous research, which indicated that a substantial percentage of women are concerned about vaccination safety and have little faith in the quality and impartiality of the information supplied by healthcare experts [167].Furthermore, according to our analysis, age was not associated with vaccine acceptance.The results of our study are inconsistent with those of prior research, which demonstrated that the COVID-19 vaccine acceptance rate increased with age [168].Similar to our results, another study conducted by Salibi et al. among Syrian refugees showed that vaccine rejection did not differ with age [169].We also analyzed two other sociodemographic factors in our review: marital status and place of residence.According to our results, married people had a lower level of vaccine hesitancy than single people.Moreover, rural people showed a higher rate of vaccine hesitancy than those living in urban areas.
We also analyzed the distribution of vaccine hesitancy and acceptance rates among different countries.The COVID-19 vaccination uptake rates in the Middle East were among the lowest worldwide, with Kuwait (23.6%),Jordan (28.4%), and Saudi Arabia (64.7%) having the lowest acceptance rates [164,170].Such low rates could be attributed to the region's broad adoption of conspiracy views as well as its subsequent anti-vaccination attitude [165,166,171,172].Nevertheless, a few nations in the area (such as Israel and the United Arab Emirates) were able to attain vaccination coverage rates that were among the highest in the world, which was ascribed to major efforts to increase vaccine trust [173,174].The vaccine acceptance rates were relatively high in Latin America, with results from Brazil and Ecuador reporting acceptance rates >70% [175,176].This was also observed in a survey in Mexico with a vaccine acceptance rate of 76.3% [175].Urrunaga-Pastor et al. attributed this to the fact that the region was one of the most affected by the pandemic internationally, with high Fig. 7. Forest plot displaying the comparison of COVID-19 vaccine hesitancy between high and low-income people.mortality rates per person, which might have contributed to lower levels of complacency [177,178].High rates of COVID-19 vaccination hesitancy have been reported in Western and Central Europe, with some European countries (Ireland, Italy, Norway, and the United Kingdom), Canada, and the United States having a better outlook.According to a recent study on vaccine hesitancy in the United States for COVID-19, geographic disparity in vaccine hesitancy is closely linked to socioeconomic variables such as race and income.The authors argue that policymakers, community groups, and religious leaders play important roles in building public trust and reducing vaccine-related hesitancy [179].
Furthermore, data from African nations revealed significant rates of COVID-19 vaccination apprehension, particularly in Cameroon (15%) and Senegal (21%); this was mostly because of a lack of confidence in  This study has some limitations.First, the studies included in this analysis varied in population, making comparison of the results challenging.Second, most studies have relied on self-reported surveys, which increased the risk of response bias.Third, it is crucial to consider that people's views on vaccines may change as real-world data become available.The studies included in our review captured public opinions during the peak of the COVID-19 pandemic, a time when information related to vaccines was still emerging and often scarce.
Further research is required to confirm this hypothesis.Longitudinal studies that follow individuals over time could be valuable for  understanding how attitudes evolve with new developments.Directly conducting interviews and focus groups with individuals can provide insights into beliefs and concerns that surveys may overlook.

Conclusion
Being men, living in an urban region, married, educated, having a history of influenza vaccination, having a higher income level, and having a history of comorbidities were associated with higher COVID-19 vaccine acceptance.In contrast, older age, history of prior COVID-19 infection, and being a healthcare worker did not significantly change the COVID-19 vaccine acceptance rate.
people/participants' attitudes and beliefs toward COVID-19 vaccine, fear of vaccination's adverse effects, and acceptance of other vaccines.Web-PlotDigitizer version 4.5 (Pacifica, California, USA) was used to extract data from the figures[11].
. Evaluation of the attitudes toward and acceptance rates of COVID-19 vaccines can aid in launching much-needed communication initiatives to boost public trust in health authorities.Using the results of several COVID-19 vaccine surveys conducted worldwide, this systematic review aimed to analyze the prevalence and factors influencing COVID-19 vaccine acceptance, intention, and hesitancy.Pogue et al. found that income did not affect vaccination attitudes.Participants with a low educational level also had a lower acceptance rate

Fig. 5 .
Fig. 5. Forest plot displaying the comparison of COVID-19 vaccine hesitancy between married and single people.

Fig. 6 .
Fig. 6.Forest plot displaying the comparison of COVID-19 vaccine hesitancy between educated people and non-educated people.

Fig. 8 .
Fig. 8. Forest plot displaying the comparison of COVID-19 vaccine hesitancy between people with and without a history of COVID-19 infection.

Fig. 9 .
Fig. 9. Forest plot displaying the comparison of COVID-19 vaccine hesitancy between people with and without a history of influenza vaccine.

Fig. 10 .
Fig. 10.Forest plot displaying the comparison of COVID-19 vaccine hesitancy between healthcare and non-healthcare workers.

Fig. 11 .
Fig. 11.Forest plot displaying the comparison of COVID-19 vaccine hesitancy between people with and without comorbidities.

Fig. 12 .
Fig. 12. Forest plot displaying the overall acceptance rate of included studies.

Fig. 13 .
Fig. 13.Forest plot displaying acceptance rate of included studies categorized by countries.

Table 1
Summary of the included articles.