Implications for COVID-19 vaccine uptake: A systematic review

Background Globally, increasing coronavirus disease (COVID-19) vaccination coverage remains a major public health concern in the face of high rates of COVID-19 hesitancy among the general population. We must understand the impact of the determinants of COVID-19 vaccine uptake when designing national vaccination programmes. We aimed to synthesise nationwide evidence regarding COVID-19 infodemics and the demographic, psychological, and social predictors of COVID-19 vaccination uptake. Methods We systematically searched seven databases between July 2021 and March 2022 to retrieve relevant articles published since COVID-19 was first reported on 31 December 2019 in Wuhan, China. Of the 12,502 peer-reviewed articles retrieved from the databases, 57 met the selection criteria and were included in this systematic review. We explored COVID-19 vaccine uptake determinants before and after the first COVID-19 vaccine roll-out by the Food and Drug Authority (FDA). Results Increased COVID-19 vaccine uptake rates were associated with decreased hesitancy. Concerns about COVID-19 vaccine safety, negative side effects, rapid development of the COVID-19 vaccine, and uncertainty about vaccine effectiveness were associated with reluctance to be vaccinated. After the US FDA approval of COVID-19 vaccines, phobia of medical procedures such as vaccine injection and inadequate information about vaccines were the main determinants of COVID-19 vaccine hesitancy. Conclusion Addressing effectiveness and safety concerns regarding COVID-19 vaccines, as well as providing adequate information about vaccines and the impacts of pandemics, should be considered before implementation of any vaccination programme. Reassuring people about the safety of medical vaccination and using alternative procedures such as needle-free vaccination may help further increase vaccination uptake.


Introduction
The COVID-19 pandemic has negatively affected communities worldwide, triggering public health interventions aimed at eradicating or reducing the transmission of COVID-19 [1]. The societal impacts of COVID-19 have been economic, social, and psychological [2,3]. By the end of August 2022, the World Health Organization (WHO) had recorded approximately 600 million confirmed cases and 6.5 million deaths due to COVID-19 worldwide [4].
Scientists have developed vaccines to prevent the spread of COVID-19 and reduce serious adverse events such as hospitalisation and death. As of November 2022, 50 COVID-19 vaccines had been approved for global use. In addition, approximately 850 COVID-19 vaccine candidates were undergoing clinical trials [5]. COVID-19 vaccines have been effective in reducing the spread of infection, severity of symptoms, and death [6,7]. A high population uptake of vaccines can result in the achievement of a herd immunity threshold. A high uptake of effective vaccines, such as that for COVID-19, can lead to substantial reductions in infections [8,9]. It is estimated that a COVID-19 vaccine with 95% and 80% efficacy will require 63% and 75% of the population, respectively, to be immune to achieve herd immunity against the infection [9,10]. However, COVID-19 vaccine hesitancy has been reported among various populations [11][12][13], including low-and middle-income countries where COVID-19 vaccine hesitancy tends to be higher [14]. Globally, by 17 October 2022, 4.98 billion people have received at least one dose of a COVID-19 vaccine, accounting for 64% of the eligible vaccination population. Among them, 28.3% were from low-income countries [4]. Thus, given the benefits of vaccines and the COVID-19 vaccination prevalence rate, there is a need to investigate COVID-19 vaccine uptake and its associated determinants to increase the success rates of vaccinations globally.
Previous systematic reviews conducted before the start of the COVID-19 vaccination program found that factors related to COVID-19 vaccine hesitancy included distrust in institutions, lower educational levels, age, female sex, being a healthcare worker, African-American ethnicity in the US, low-income levels, and the use of social media for sourcing COVID-19 information [15][16][17][18][19][20][21][22][23]. There is also some evidence that there have been gradual attitudinal changes towards vaccine hesitancy in the general population [13,24]. Furthermore, no systematic reviews of before-and-after studies (e.g. from the first roll-out of a COVID-19 vaccination program) have been conducted. Therefore, a review of the empirical literature is needed to shed light on the relevant patterns of COVID-19 vaccine-uptake intentions. We carried out a systematic review to explore health behaviours (e.g. vaccine uptake determinants and attitudes) and to determine whether these change over time [25][26][27][28].

Methods
We conducted a systematic review to investigate attitudes towards COVID-19 vaccine acceptance and its determinants during the roll-out of global COVID-19 vaccination programmes. A previously registered protocol on PROSPERO (#CRD42021281769) guided this review. The review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [29]. We wished to understand temporal changes and therefore, for the purpose of this systematic review, Time 1 represents 'before the first roll-out of the COVID-19 vaccine' (prior to FDA approval: 11 December 2020), while Time 2 represents 'after the first roll-out of the COVID-19 vaccine'.

Search terms and strategies
Seven databases were searched: PubMed, Medline and Embase (via Ovid), Scopus, PsycINFO, Web of Science, and CINAHL. Search strategies for the review were aligned to each database according to indexing terms, in addition to Medical Subject Headings (MeSH), truncations, and Boolean operators. Terms describing the concept of COVID-19 were used in each database, and phrases denoting vaccine uptake and intentions were also used in each database. A preliminary literature search began on 12 July 2021 and all searches were completed on 18 March 2022. Searches were not limited to any specific geographical location but were limited to human studies only. The search included studies published since the emergence of COVID-19 in 2019. See Appendix 1 for the specific search terms used for each database.

Inclusion and exclusion criteria
Studies that explored the determinants of COVID-19 vaccine uptake based on quantitative nationwide surveys were included in this review. Thus, the included studies surveyed national populations (i.e. either representative or non-representative sample sizes) of at least 100 participants aged 18 years and over. The included studies were required to report the reliability of the non-binary scales used in the study. Only English-language publications were included in this analysis.
Exclusion criteria were applied to cross-national comparison studies (i.e., between-country studies), as we intended to provide country-specific evidence. Studies that used only a defined population characteristic (e.g. health workers or students only) and studies that provided only descriptive findings of COVID-19 vaccine uptake rates were also excluded. No gray literature (such as reports, speeches, and newsletters) was added to the selected papers.

Screening
Papers containing search words were extracted from seven databases and imported into Endnote version 20, and duplicates were removed (Fig. 1). The initial data screening was performed by the first author (PA). The titles of the articles were first screened to identify those relevant to the determinants of COVID-19 vaccine uptake (Fig. 1). This was followed by screening of abstracts and sample/method sections based on the study design and nationwide studies. The full texts of the articles identified at this stage were screened independently by two authors (PA and JM) to determine their eligibility based on the inclusion criteria of the review (Fig. 1). Disagreements between the two authors (PA and JM) on a paper [30] were resolved by a third investigator (CS).

Data extraction and quality assessment
Data from the studies used in the current review were extracted independently by two authors (PA and JM). Data were extracted under  the following headings: author, year, aim, country, study period, sample size, study design, scales and reliabilities, number of participants, recruitment method, and results (Table 1). Data quality was assessed by two authors using an adapted version of the Newcastle-Ottawa scale for cohort studies ( Table 1). The quality of the studies ranged from unsatisfactory to good, with 17 (30%) of the 57 studies appraised as unsatisfactory, 31 (54%) as satisfactory, and 9 (16%) as good (Table 1). Major quality issues included inadequate information on sample justification and statistical power and study results that did not adjust for relevant predictors, risk factors, or confounders.

Data analysis
The features of all studies are summarised, including the determinants of COVID-19 vaccine uptake (Table 1). Researchers were unable to conduct a meta-analysis due to the heterogeneity of the measurement of COVID-19 vaccine determinants (i.e., some studies used dichotomous measures and others used scales); hence, data was described narratively. Predictive factors of COVID-19 vaccine uptake were grouped under four broad headings: demographic, social, psychological, and infodemic (false or misleading information in digital and physical environments during the breakout of a disease) [31].

Results
Overall, 12,502 articles were identified from seven databases. After removing duplicates, 10,881 articles were screened by title and abstract, and 357 articles were identified. Finally, 57 articles met the inclusion criteria after screening the full text of the selected articles (see flow chart for details: Fig. 1). To reiterate, the use of the COVID-19 vaccine uptake only reflects intentions or willingness to accept the COVID-19 vaccine.
Additionally, the following factors were common positive predictors of COVID-19 vaccine uptake for both Times 1 and 2: selfefficacy, confidence in receiving the COVID-19 vaccine without any side effects [58,74,78,85]; life satisfaction and a positive view of the world [39]; health engagement; belief in the importance of herd immunity [47,49]; and concerns about the safety of relatives and friends; and society [69,73,74,92]. However, less self-efficacy in preventing the infection negatively predicted COVID-19 vaccine uptake in a study conducted in Bangladesh during Time 2 [94]. During both times, desire for natural immunity, confidence in having a strong immune system, and belief in traditional remedies as a cure for COVID-19 were found to be associated with COVID-19 vaccine hesitancy [32,33], including the desire for others to be vaccinated first [38,42].

Social predictors of COVID-19 vaccine uptake
Information from the mass media, official national websites, government institutions, health professionals, newspapers, national television, YouTube, and significant others (e.g. family and friends) positively predicted COVID-19 vaccine uptake [49,85,103] for both time periods. However, in China, the frequency of social media use, reliance on information from WhatsApp, and using different social media were negatively correlated with COVID-19 vaccine uptake at Time 1 [93,103]. In the US, information from the White House during 2019-2021 and higher approval of President Trump were associated with COVID-19 vaccine hesitancy during Time 1 [84,87] (Table 1).

COVID-19 Infodemic predictors of COVID-19 vaccine uptake
Common infodemics, specifically the belief that COVID-19 is a biological weapon or a myth, correlated negatively with COVID-19 vaccine uptake at both Times 1 and 2 [32,37,45,46,50,73,82]. In one study, the reverse of this relationship was reported at Time 2 [37]. Belief that the COVID-19 pandemic is a strategy for big pharma to make money, caused by 5 G mobile networks, and that the COVID-19 vaccine is harmful predicted less willingness to be vaccinated for COVID-19. These factors were notable at Time 1 (46, 47, 54, 55, and 88). Nonetheless, during Time 2, not believing in the existence of COVID-19 and the belief that the COVID-19 vaccine contained substances derived from animals such as pigs was related to COVID-19 vaccine hesitancy [32]. Paradoxically, religious faith factors, such as 'the pandemic is humanity's destiny, were related to positive intentions to accept the COVID-19 vaccine [37]. (see Fig. 2 for a summary of all the identified predictors in this review).

Discussion
We found that there tended to be more COVID-19 vaccine hesitancy prior to the first FDA approval of a COVID-19 vaccine (Time 1) than after (Time 2). Attitudes aligning with acceptance of the COVID-19 vaccine also increased over time, representing a positive move towards vaccination. We found that people were concerned about the rapid development of the COVID-19 vaccine, its safety, side effects, and its effectiveness. These factors were reported consistently across both time periods by 27 studies conducted across five continents (Africa, Asia, North and South America, and Europe) and were found to be negatively related to COVID-19 vaccine uptake during both time periods.
Our findings were similar to those of previous studies on influenza vaccine uptake [89,104,105]. Previous studies found higher levels of anxiety, fear, and worry to be positive predictors of influenza vaccine uptake [106]. In agreement with other studies (e.g. [104,107,108], perception of the risk of COVID-19 infection and perceived benefit of COVID-19 vaccine were found to correlate positively with COVID-19 vaccine uptake. Likewise, COVID-19 information from health professionals, government institutions, and other social media (e.g. national websites) was related to COVID vaccine uptake for both time periods. These findings were similar to those of studies on influenza vaccines and other pandemic vaccine uptake studies conducted in the US [105,109]. Our review found evidence that previous experience with vaccination predicts the willingness to accept a vaccine [107]. Specifically, previous experiences of both COVID-19 infection and influenza vaccination were positively related to COVID-19 vaccine uptake. Cues to action (e.g. recommendations from professionals), being informed about COVID-19 preventive measures, and adherence to these measures were positively associated with COVID-19 vaccine uptake. That is, respondents who followed such health behaviours might have positive attitudes towards health behaviours in general, including vaccination [110]. Studies of influenza vaccine uptake intentions have reported selfefficacy as one of the determinants of influenza vaccine uptake [110], and our review of COVID-19 vaccine uptake provided additional evidence to support such a relationship.
Infodemics have been reported to impede vaccine uptake in different populations globally [111]. The COVID-19 infodemics identified in this review were also negatively associated with the COVID-19 vaccine uptake. Paradoxically, religious faith in COVID-19 infodemics was positively correlated with COVID-19 vaccine uptake. A possible reason for this could be that religious bodies sensitised individuals to the need to be vaccinated against COVID-19 to facilitate their ritual activities as the pandemic halted many religious gatherings worldwide.
Another unique factor found to predict COVID-19 vaccine hesitancy after FDA approval was fear of medical procedures (e.g. injection), which is in line with findings from general vaccination programs in India [112]. An underestimated perception of COVID-19 incidence, that is, participants might have lost focus on the pandemic, perhaps due to lack of or ignoring information available from different media, could have lessened the desire or urgency to vaccinate against the pandemic. Studies also indicated that inadequate information regarding both COVID-19 infection and the vaccine related to COVID-19 vaccine hesitancy. This finding will be of interest to relevant stakeholders, especially as it occurred after the first rollout of the COVID-19 vaccine. In addition, nationalism (e.g. national narcissism) and certain types of motivation have been found to predict COVID-19 vaccine hesitancy.
Common demographic factors linked to COVID-19 vaccine uptake included sex, marital status, age, education, area of residence, and religious affiliation. Religious affiliation was found to show specific relationships in terms of predicting COVID-19 vaccine uptake because religious affiliation negatively predicted COVID-19 vaccine uptake in Jordan, Somalia, Malaysia, and Ethiopia, which is consistent with previous literature [113]. In terms of sex, the majority of the studies reviewed supported previous studies suggesting that males were more likely to accept a vaccine than females [114].
To the best of our knowledge, this systematic review is the first to explore the determinants of COVID-19 vaccine uptake across two time periods. This timely exploration of differences in the trends of COVID-19 vaccine uptake determinants provides an overview to stakeholders about attitudinal changes occurring over time since the emergence of COVID-19. Again, the selection of studies that were assessed for quality ensured an adequate level of accuracy and confidence in our findings. However, our review has the following limitations. Most of the studies were cross-sectional surveys. Caution concerning the interpretation of our results should be taken, as we were unable to determine causation between the variables. Qualitative studies and studies published in languages other than English were excluded from the review because of the time and cost involved.
The global aim of achieving high uptake of a COVID-19 vaccine could be achieved if specific concerns associated with vaccine hesitancy, such as safety, effectiveness, potential side effects, and benefits related to COVID-19 vaccines, including disbeliefs and adequate information about the pandemic, are clearly communicated and understood. Addressing the difference in pre-and post-first FDA approval of COVID-19 vaccine determinants is important for policymakers to understand the factors that emphasise current COVID-19 vaccination programs. Infodemics were additional factors in this regard which were associated with hesitancy attitudes. A strategy found to help address misinformation is psychological inoculation (i.e. exposing individuals to a version of already known information, which they can refute) [115]. Again, since phobia of medical procedures was found to contribute to COVID-19 vaccine hesitancy after FDA approval, clinicians may consider dealing with medical procedure phobias by considering different administration routes of COVID-19 vaccines, for example, needleless injection procedures to increase COVID-19 vaccine coverage. This seems to be an important predictor given that 69% of participants (participating in an influenza survey) opted for a needleless route of administration [116]. Finally, a standardised method of measuring COVID-19 vaccine uptake will help ensure precision in the future, as most of the studies measured uptake dichotomously, which limits accuracy and makes it difficult to compare studies on COVID-19 uptake; hence, measuring COVID-19 vaccine uptake using a wellvalidated scale may help increase the measurement precision of COVID-19 uptake.

Research in context
By the end of August 2022, the World Health Organization (WHO) had recorded approximately 600 million cases of COVID-19 infections and 6.5 million consequent deaths worldwide. Vaccines are an effective means of reducing the spread of infection and preventing diseases, for example, by achieving the herd immunity threshold. However, COVID-19 vaccine hesitancy has been reported among various populations globally, particularly in resource-poor countries.
To understand this phenomenon, we reviewed materials published between July 2021 and March 2022. We searched PubMed, MEDLINE(Ovid), Web of Science, Embase (Ovid), Scopus, PsycINFO, CINAHL, and dimensions for systematic reviews and meta-analyses of studies relevant to COVID-19 vaccine uptake intentions published in English since COVID-19 was reported on 31 December 2019 in Wuhan, China.
This novel review explores the determinants of COVID-19 vaccine uptake since the emergence of COVID-19. We distilled the evidence with regard to demographic, psychological, social, and infodemic determinants of COVID-19 vaccine uptake, focusing on studies involving national populations in different countries. We identified common COVID-19 vaccine hesitancy determinants, such as concerns regarding COVID-19 vaccine safety, negative side effects, fast development of COVID-19 vaccine, and uncertainty about vaccine effectiveness, as well as country-specific predictors. We also assessed real-world evidence of factors associated with COVID-19 vaccine hesitancy both before and after FDA approval, such as phobia of medical procedures and inadequate information about vaccines and the pandemic.

Implications
This review informs clinicians and stakeholders about the most relevant predictors of COVID-19 vaccine uptake that should be considered to enhance vaccination success. Specifically, campaigns should consider concerns surrounding COVID-19 vaccine such as information about vaccines and pandemics and safety of vaccination procedures to increase COVID-19 vaccination coverage.

Declaration of interests
We declare no competing interests.