COVID-19 Vaccine Acceptance Level in Ethiopia: A Systematic Review and Meta-Analysis

Background The coronavirus disease 2019 pandemic has had a devastating impact on the everyday lives of the world's population and to this end, the development of curative vaccines was upheld as a welcome panacea. Despite the undeniable negative impact of the disease on human beings, lower than expected proportions of people have taken up the vaccines, particularly in the developing non-Western world. Ethiopia represents an interesting case example, of a nation where COVID-19 vaccine acceptance levels have not been well investigated and a need exists to assess the overall level of vaccine acceptance. Methods A systematic multidatabase search for relevant articles was carried out across Google Scholar, Web of Science, Science Direct, Hinari, EMBASE, Boolean operator, and PubMed. Two reviewers independently selected, reviewed, screened, and extracted data by using a Microsoft Excel spreadsheet. The Joanna Briggs Institute prevalence critical appraisal tools and the modified NewcastleOttawa Scale (NOS) were used to assess the quality of evidence. All studies conducted in Ethiopia, reporting vaccine acceptance rates were incorporated. The extracted data were imported into the comprehensive meta-analysis version 3.0 for further analysis. Heterogeneity was confirmed using Higgins's method, and publication bias was checked by using Beggs and Eggers tests. A random-effects meta-analysis model with a 95% confidence interval was computed to estimate the pooled prevalence. Furthermore, subgroup analysis based on the study area and sample size was done. Results and Conclusion. After reviewing 67 sources, 18 articles fulfilled the inclusion criteria and were included in the meta-analysis. The pooled prevalence of COVID-19 vaccine acceptance in Ethiopia was 57.8% (95% CI: 47.2%–67.8%). The level of COVID-19 vaccine acceptance in Ethiopia was at a lower rate than necessary to achieve herd immunity. The highest level of vaccine acceptance rate was reported via online or telephone surveys followed by the southern region of Ethiopia. The lowest vaccine acceptance patterns were reported in Addis Ababa.


Introduction
e coronavirus (COVID)-19 disease is caused by a highly contagious acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and for several months after its emergence, no proven vaccine was available. Globally, the fatality rate of COVID-19 infection was estimated at 0.5% to 1% [1]. Its negative impact reached the everyday lives of all human beings globally [2][3][4][5] to the extent of disrupting the normal economic and social activities of the world's population [6][7][8][9][10]. In response, global communities took different measures to contain its spread, and these included lockdowns and border closures [11][12][13].
ere has been an implementation of various public health measures that included hand hygiene, lockdowns, and social distancing in most parts of the world. Even so, the overall impact of these COVID-19 disease prevention measures has varied from one set to the other [14].
Medically, vaccines have been separated from other preventative measures on the basis of their superior prevention and disease control profiles [15,16]. Even so, their uptake remains an issue of contention.
Barriers to vaccine acceptance are complex; moreover, they are context-specific and fluctuate across the place, time, and vaccine type [17]. However, there has been a continuous distribution of COVID-19 vaccines across the world population including Ethiopia.
at said, hesitation against vaccines represents the single most notable obstacle to having adequate coverage across various populations. Vaccine hesitation was identified by the World Health Organization as one of the top global health threats as of the year 2019 [18]. e corpus of evidence related to COVID-19 vaccine acceptance suggests that a considerable portion of people are opposed to the vaccine. A global report on COVID-19 vaccine acceptance reported the acceptance rate to be under 67% [19].
Notably, most reports in Ethiopia show lower vaccine coverage. Additionally, the lack of specified investigation of the above strongly points to an acknowledgment that, the overall level of COVID-19 vaccine acceptance in Ethiopia represents a poorly investigated phenomenon. Guided by this, this meta-analysis offers an assessment of the overall level of vaccine acceptance in Ethiopia.

Question.
What is the overall level of COVID-19 vaccine acceptance in Ethiopia?

Reporting.
e preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [20] was used as a reporting framework within this meta-analysis (additional file 1 S1).

Searching Strategies.
e PRISMA systematic review protocol was followed as a reporting guideline and eligible studies for the analysis were selected in terms of abstracts, titles, and then for full articles on the basis of the inclusion criteria. EMBASE, PubMed, Hinari, Google Scholar, Web of Science, Science Direct, and African Journals Online were systematically searched to identify articles that included Medical Subject Headings and free-text languages. ese databases were searched by using both controlled and freetext languages. In terms of free-text searches, the keywords included the following combination of terms: (Willingness, OR COVID-19 Vaccine OR Acceptance) AND Ethiopia. e controlled searches included the following Medical Subject Headings (MeSH) terms: "COVID-19 Vaccine Acceptance" and "Ethiopia" as recommended for each database. Search terms were used individually and in combination using "AND" and "OR" Boolean operators. e search was guided by PICO, a population that was intended to take the vaccine.

Inclusion and Exclusion Criteria.
e following types of studies from 2019 to 2022 were included; study populations comprised any age group, study outcome was "willingness or intention to take vaccine or vaccine acceptance," study design is cross-sectional and studies written in English were included. However, in this systematic review and metaanalysis; qualitative studies and data on those who took the vaccine were excluded.

Outcome of Interest: PICO.
e population of the study was any age group and the outcome of interest was vaccine acceptance which thus was reported as "Are you willing to take COVID-19 Vaccine if available to you?" and measured as yes and no or willing versus unwilling. e level of vaccine acceptance was presented as a frequency and percentage.

Screening and Data Extraction.
Screening for titles and abstracts against the inclusion was conducted by the two reviewers (SAT and GGD). Furthermore, an independent assessment was made for the full-text articles based on the predetermined inclusion criteria. Inconsistencies across the reviewers were dealt with via a discussion and consensusseeking engagements involving all the investigators. Data extraction was made by three authors (TM, GZM, and BGE) independently from a random sample of 20% of the papers to check consistency and cross variation.

Study Quality Assessment.
A structured data abstraction form was constructed in Microsoft Excel. Attention was given to clarity of data, objective, study design, population, sample size, and proportion of vaccine acceptance (Table 1). e modified version of the NewcastleOttawa Scale for the cross-sectional study [39] was used for the methodological qualities of each article. Additionally, studies were critically appraised with the Joanna Briggs Institute prevalence critical appraisal tool [40].

Data Synthesis and Statistical
Analysis. Data were extracted using a Microsoft Excel spreadsheet and imported to comprehensive meta-analysis version3.0 software for further analysis. e pooled effect size with a 95% confidence interval of national COVID-19 vaccine acceptance; a rate that was determined using a weighted inverse variance random-effects model. e I 2 statistic; 25, 50, and 75% representing a low, moderate, and high heterogeneity consecutively assessed the heterogeneity across the studies [41], whereas the publication bias was evaluated by funnel plot and Eggers and Beggs test [42].

Selection of the Studies.
A comprehensive literature search of the databases yielded a total of 67 published articles, of which 20 articles were retrieved from Google Scholar, 13 articles from PubMed, 12 articles from African Journals online, 7 articles from Hinari, and 15 articles from EMBASE, Web of Science, and Science Direct. irty-one articles were excluded for duplication and scope. e other 18 articles were excluded for failing to offer reports on the outcome. A total of 18 full-text articles that fulfilled the eligibility criteria with a total sample size of 10873 were included in the final analysis for the systematic review and meta-analysis (Figure 1).

Subgroup Analysis.
According to the subgroup analysis report, the highest level of vaccine acceptance (68.7%; 95% CI: 34.1%-90.3%) was reported in online or telephone surveys whilst the lowest level of vaccine acceptance (51.8%; 95% CI: 33.3%-69.8%) was reported in Addis Ababa. Regarding the sample size, the highest level of vaccine acceptance (74.0%; 95% CI: 23.5%-96.4%) was reported in studies with a sample size of larger than 800 (Table 2).

COVID-19 Vaccine Acceptance.
In this systematic review and meta-analysis, the pooled estimate of the COVID-19 vaccine acceptance rate was illustrated via a forest plot. e pooled prevalence of vaccine acceptance in Ethiopia was 57.8% (95% CI: 47.2%-67.8%) (Figure 2).

Assessment of Publication Bias.
A symmetrical funnel plot was observed. Begg's and Egger's tests showed the absence of significant publication bias at a p value of >0.05 (Figure 3).

3.6.
Heterogeneity. For the identification of the possible causes of variation across different studies, meta-regression analysis was conducted using sample size and study area. e result showed that there was no significant heterogeneity across the studies (p > 0.05) ( Table 3).

Discussion
e current systematic review and meta-analysis provided critical evidence on the level of COVID-19 vaccine acceptance in Ethiopia. is study found the overall level of vaccine acceptance in Ethiopia to be at 58.7%.
is variation might be due to variation in the availability of vaccine type and population characteristics.
By contrast, the level of vaccine acceptance in the current study was higher than the findings of 34% in Liberia [66], 21.4% in Lebanon [67], and 27.7% in the Democratic Republic of Congo [68]. Such variation might be due to variations in sample size and level of awareness among the study participants. e vaccine acceptance level was higher in different countries including 78% in Scotland [69], 78.5% in Greece [70], 75% in Portugal [71], 77.65% in France [72], 68.5% in the United States [73], 80% in Canada [74], 90.1% in South   Africa [75], and 80.9% in Uganda [76]. is difference might be due to population characteristics and the availability of vaccine options.

Conclusion
e level of COVID-19 vaccine acceptance in Ethiopia was at a lower rate than necessary to achieve herd immunity. e highest level of vaccine acceptance was reported in online or telephone surveys followed by the Southern region of Ethiopia whereas a lower level of vaccine acceptance was reported in Addis Ababa. With regards to the sample size, the highest level of vaccine acceptance was reported in studies with a sample size larger than 800. Concerned bodies in Ethiopia including the government should work on scaling up the vaccine coverage for the Ethiopian people.

Data Availability
All the data are contained within the article. Disclosure e preprinted form of this manuscript was posted at research square in the form preprint accessed from https://www. researchsquare.com/article/rs-1332473/v1 DOI: 10.21203/ rs.3.rs-1332473/v1.

Conflicts of Interest
e authors declare no conflicts of interest.

Authors' Contributions
SAT and TM were involved in the development of the protocol, design, selection of the study, data extraction, statistical analysis, and development of the initial draft of the manuscript. GGD, GZM, and BGE got involved in the preparation and editing of the final draft of the manuscript. All authors read and approved the final draft of the manuscript.