COVID-19 vaccine acceptance and its association with knowledge and attitude among patients with chronic diseases in Ethiopia

ABSTRACT

COVID-19 vaccine acceptance is crucial for patients with chronic diseases, but previous studies in Ethiopia have yielded inconsistent and inconclusive findings. To fill this gap, we conducted a systematic review and meta-analysis following established guidelines. Our search included relevant articles published between 2019 and 2023 from various sources. We assessed study heterogeneity and publication bias, and performed subgroup and sensitivity analyses. Our findings indicate that the COVID-19 vaccine acceptance rate among patients with chronic diseases in Ethiopia was 55.4%. We also found that good knowledge and a favorable attitude toward the vaccine were positively associated with the acceptance rate. Based on these results, we recommend that healthcare professionals, policymakers, and healthcare guide developers should work more to address the relatively low acceptance rate. Improving the knowledge and attitude further about the COVID-19 vaccines is crucial. Future research should include community-based and qualitative studies to enhance our understanding of vaccines acceptance.

KEYWORDS:

• COVID-19 vaccine acceptance
• patients with chronic diseases
• knowledge
• attitude
• Ethiopia

Background

A Coronavirus disease 2019 (COVID-19) vaccine is intended to provide acquired immunity against Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the virus is responsible for COVID-19, and to reduce the number of reported incidents and deaths.¹ On the other hand, vaccination uptake is highly dependent on vaccine acceptance and eventually plays a role in developing herd immunity, which is key for the control of COVID-19 related illness.²

According to the Centers for Disease Control and Prevention (CDC), it is strongly recommended that people who have chronic medical disorders, are immunosuppressed or, are at a high risk of developing serious COVID-19 should be vaccinated. Furthermore, because they are more likely to have complications from some vaccine-preventable diseases, people with chronic health conditions should continue to receive recommended vaccines.^3,4

Morbidity and mortality due to COVID-19 is higher among populations with chronic disease conditions such as HIV, diabetes mellitus, cardiovascular disease, mental illness and cancer.^5–8 As numerous studies have demonstrated, people with chronic health conditions such as diabetes, cardiovascular disease (CVD), and obesity with COVID-19, are more likely to be hospitalized and die.^9–13 Furthermore, the mortality and morbidity rate of COVID-19 of patients with chronic health conditions is directly related to their vaccination rate.¹⁴ The deaths from COVID-19 continue to increase, although morbidity is relatively decreasing.¹⁵ In particular, populations with chronic diseases who are not vaccinated or partially vaccinated had a significantly higher death rate than the fully vaccinated ones.¹⁶ However, this scenario is not well reported in the African WHO region including Ethiopia.¹⁵

Knowledge and attitudes of chronic disease patients about the COVID-19 vaccine, have an impact on whether or not they accept the vaccination.^17,18 To increase their acceptance rate, it is essential to inform patients with chronic diseases about the efficacy and safety of COVID-19 vaccination.¹⁹ The COVID-19 Vaccination can help protect people with chronic diseases from serious complications and hospital stays.²⁰ It is a crucial weapon in the fight against COVID-19.¹⁷ In long-term view, COVID-19 Vaccination is a great solution, particularly for populations with chronic health conditions.²¹ However, widespread hesitancy in the use of vaccines can hinder the adoption and efficacy of vaccines, as it encompasses a variety of attitudes and beliefs about immunization that lead to delay or downright avoiding vaccination.^21,22

It is fundamental to optimize the acceptability and confidence of the COVID-19 vaccines in order to achieve high rates of vaccination and herd immunity in populations with chronic medical conditions. However, concerns from patients with chronic health conditions about the safety and efficacy of COVID-19 vaccinations could develop given to the more rapid or sooner development and dissemination.^21,22

Reducing the global burden of COVID-19 among populations with chronic health conditions require the COVID-19 vaccination.⁸ The degree to which vaccinations have been accepted will determine how effectively this plan works. However, studies have shown that the acceptance and uptake rate of COVID-19 vaccination among chronic disease populations are relatively lower than the expected standards.^23–37

Previous studies with inconsistent findings revealed that populations with chronic diseases had a higher probability of accepting the COVID-19 vaccine in Ethiopia.^{23,25,31,38–40} However, there have been no pooled measures for the COVID-19 vaccine acceptance by patients with chronic diseases in Ethiopia. In addition, primary studies have shown that the association of knowledge and attitude toward the COVID-19 vaccine acceptance among patients with chronic diseases had inconsistent findings in Ethiopia.^23,24,40,41 Therefore, this systematic review and meta-analysis aim to show the national pooled magnitude of the COVID-19 vaccine acceptance and its association with knowledge and attitude among patients with chronic diseases in Ethiopia.

Methods

Study designs and search strategy

A systematic review and meta-analysis of previously published and unpublished studies was carried out to determine the pooled magnitude of the COVID-19 vaccine acceptance among adult patients with chronic diseases in Ethiopia. The effect of the knowledge and attitudes of patients about COVID-19 vaccine acceptance was also determined. The present review was conducted strictly in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines⁴² (S₁). Before beginning the review, we searched different databases for existing systematic reviews and meta-analyses on our topic to avoid duplication. We also checked whether any ongoing projects related to the current systematic review and meta-analysis were available. There were no previous systematic reviews or meta-analyses on the topic in question. The following databases were used to look for articles: PUBMED, MEDLINE, Google Scholar, Cochrane Library, EMBASE, and PSYCINFO. And also University library repository websites were searched.

The key terms used for this review were: (((((((((((((((((((((((COVID 19 vaccine acceptance” [MeSH Terms]) OR (”COVID 19 vaccine prevalence” [MeSH Terms])) OR (“COVID 19 vaccine proportion” [MeSH Terms])) OR (”COVID 19 vaccine willingness” [MeSH Terms])) OR (”COVID 19 vaccine intention” [MeSH Terms])) AND (”chronic patient*”[MeSH Terms]) OR (”chronic diseased patient*”[MeSH Terms])) OR (”Diabetes patient*”[MeSH Terms])) OR (”hypertensive patient”[MeSH Terms]) OR (”heart Failure patients”[MeSH Terms]) OR (”chronic medical conditions”[MeSH Terms])) OR (”asthmatic patients”[MeSH Terms])

((((((((((((((((((Knowledge)) OR (awareness) AND “COVID 19 vaccine acceptance” [MeSH Terms]) OR (“COVID 19 vaccine prevalence” [MeSH Terms]) OR (“COVID 19 vaccine proportion”[MeSH Terms])) OR (“COVID 19 vaccine willingness” [MeSH Terms]) OR (“COVID 19 vaccine intention” [MeSH Terms])) AND (“chronic patient*”[MeSH Terms])) OR (“chronic diseased patient*”[MeSH Terms])) OR (“Diabetes patient *”[MeSH Terms])) OR (“Hypertensive patient”[MeSH Terms]) OR (“heart failure patients”[MeSH Terms])) OR (“chronic medical conditions”[MeSH Terms])) OR (“asthmatic patients”[MeSH Terms])

(((((((((((((((Attitude [MeSH Terms])) OR (intention[MeSH Terms])) OR (belief [MeSH Terms])) OR (hesitancy [MeSH Terms]) AND “COVID 19 vaccine acceptance” [MeSH Terms]) OR (“COVID 19 vaccine prevalence” [MeSH Terms]) OR (“COVID 19 vaccine proportion”[MeSH Terms])) OR (“COVID 19 vaccine willingness” [MeSH Terms]) OR (“COVID 19 vaccine intention” [MeSH Terms])) AND (“chronic patient*”[MeSH Terms])) OR (“chronic diseased patient*”[MeSH Terms])) OR (“Diabetes patient*”[MeSH Terms])) OR (“Hypertensive patient”[MeSH Terms]) OR (“heart failure patients”[MeSH Terms])) OR (“chronic medical conditions”[MeSH Terms])) OR (“asthmatic patients”[MeSH Terms])) OR (“chronic obstructive pulmonary disease*”[MeSH Terms])) AND (Ethiopia[MeSH Terms])))). These key terms were modified accordingly to the databases.

Study selection and eligibility criteria

All studies conducted between 2019 and 2023 were taken into account. However, studies were available only in 2021 and 2022. The references of the chosen articles were also screened for additional articles that could be included in this review. The review included all types of the studies published or unpublished in the form of articles, master’s theses, and dissertations written in English. Additionally, all studies that used observational study designs were considered. However, only cross-sectional analytical studies were available and included. We excluded studies with methodological problems and review articles. The retrieval of studies for inclusion in the final review was evaluated by assessing their title, abstract and full text to ensure compliance with our eligibility criteria.

Outcome measurements

The primary outcome of this review was COVID-19 vaccine acceptance rate among patients with chronic diseases in Ethiopia. The included studies measured it by at least one of the following questions¹ “Are you willing to be vaccinated against COVID-19?” (Yes/No) (2.) ‘Should you take or have you taken the COVID-19 vaccine?’ And (3.) ‘If a vaccine against COVID-19 is available, will you take it?.”^{24,26,28,29,41,42,44–49} As a result, this review included all studies that used either of the above-mentioned definitions.

The other outcome of this review was to identify the association between the knowledge and attitude of COVID-19 vaccine acceptance by patients with chronic diseases. Therefore, the knowledge of the participants about the COVID-19 vaccine was measured through either of the following means¹: Questions were asked and respondents who scored 70% or more were taken as having good knowledge.^{24,28,43,45,46,48} Questions were asked and respondents who scored higher than the mean were considered to have good knowledge.^26,40,44 Therefore, this review included all studies that used either of the above-mentioned definitions.

Similarly, the attitude of the participants about the COVID-19 vaccine was measured¹: Likert scale questionnaires were asked and the respondents scored 70% and above were taken as having favorable attitude.^{24,28,43,45,48} Based on the mean score: respondents who scored higher than the mean were considered to have favorable attitude.^26,40,44 This review included articles based on at least one of the above measurements used by the primary studies.

Quality assessments and data extraction

Before being included in the meta-analysis, articles were selected using their titles, abstracts, and full paper evaluations. To combine database search results and manually remove duplicate articles, we use reference management software (Endnote version X9). For the evaluation of the study, the Joanna Briggs Institute Meta-analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) was used⁴⁹ (S₂). The instrument includes details on participant recruitment, sample representativeness of the study population, sample size adequacy, thorough description of the study subject and study setting, adequate coverage of the data analysis, objective criteria in measuring the outcome variable and subpopulation identification, reliability, appropriate statistical analysis, and identification of confounding variables. The quality ratings of the included studies were evaluated and the mean scores were used to categorize the studies as high or low quality.

Two independent review teams (TDT and ZBA; and AMK and BA) extracted data using a predefined standardized data extraction format (S₃). The primary author’s name, year of study, year of publication, study area, region, study design, sample size, number of subject outcomes, response rate, number of participants with knowledge status, number of participants with attitude status, odds ratio for knowledge and attitude with 95% confidence interval, and quality score of the study were all included in the data extraction spreadsheet. Data on publication status was also collected for each included study.

Disagreements among review teams were discussed with the other review team until consensus was reached. Disputes between two independent review teams were settled by bringing in a third review team (TMD and AGY). When full-text articles were unavailable, the corresponding authors were contacted only once via e-mail. The articles were removed from the review if no response was received within one month.

Heterogeneity and publication bias

The presence of heterogeneity was evaluated using the Cochrane’s Q and I² statistics (p-value < .05 considered heterogeneity). I² test statistics results of 25, 50, and 75% were classified as low, moderate, or high heterogeneity, respectively.⁵⁰ To evaluate publication bias, funnel plot, Begg rank correlation, and Egger regression test were employed.^51–54 To account for publication bias, the non-parametric trim and fill analysis⁵⁵ using the random effect model was used for results that showed the presence of publication bias (Egger and Begg test = p < .05).

Statistical methods and analysis

STATA 17 software was used to perform the meta-analysis after the data was exported from Microsoft Excel (S₄). Forest plots⁵⁶ were used to illustrate the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia. There was a high level of heterogeneity among the included studies. Therefore, the random effect analysis model was used to estimate the effect of Der Simonian and Laird’s pooled magnitude of COVID-19 vaccine acceptance.⁵⁷

A variety of study characteristics were also used to perform subgroup analyzes, including sample size, region, response rate, publication status (published or unpublished), study year (2021 or 2022) and study quality score (low or high score).

Using a random-effects model, we performed a sensitivity analysis to assess the impact of a small independent study on the overall pooled magnitude.⁵⁸ Furthermore, a meta-analysis regression was carried out to determine the sources of the heterogeneity of the studies. The Sample size, study year, study quality score, region and publication status of the included studies were used as study-level variables for meta-regression. Odds ratios (OR) with a 95% confidence interval (CI) were used to show the association of attitudes and knowledge with the COVID-19 vaccine acceptance.

Results

Study selection

A total of 183 studies were found that were conducted between 2019 and 2023 (181 published and 2 unpublished). Fifty-six studies that were duplicated were eliminated. After examining the titles and abstracts, 84 studies were removed. The eligibility and report of interest of the remaining 43 studies were evaluated in whole. Thirty-one of these studies were deemed ineligible and excluded because they did not have a compelling outcome. Twelve of the remaining studies met the JBI quality assessment eligibility criteria with a score of six or higher, and were included in the final meta-analysis. Their quality status was classified into low and high quality according to the mean score, which was 6.5. The selection procedure and presentation of the systematic review were guided by the flow diagram of preferred reporting items for systematic reviews and meta-analysis (PRISMA 2020) (Figure 1).

Figure 1. PRISMA flow diagram of the included studies in the systematic review and meta-analysis of the COVID-19 vaccine acceptance and its association with knowledge and attitude among patients with chronic diseases in Ethiopia from 2019 to 2023.

Characteristics of the included studies

A total of 4758 patients with chronic diseases from 12 primary studies were included in this review. Ten were published in international reputable journals. However, two were unpublished papers found in research square registries as preprint⁴⁵ and as gray literature on Google scholar published from the website of the University.²⁶ The minimum and maximum sample sizes were from the Bahir Dar and Gondar studies, 280 and 423 respectively.^45,46 Of the 12 studies: seven from Amhara region,^{24,28,41,45,46,48} three from Central Ethiopia (formerly part of the Southern nations, nationalities and peoples of Ethiopia region),^26,29,44 one from Oromia,⁴⁰ and one from Addis Ababa administrative region.⁴⁷ Furthermore, 75% of the studies included adult patients with chronic diseases, who had follow up at hospital.^{24,26,41,44–47,49} The other three studies included adult cancer, diabetes mellitus and HIV patients separately^28,29,47 (Table 1).

Table 1. Characteristics of studies included in the systematic review and meta-analysis on the COVID-19 vaccine acceptance and its association with knowledge and attitude among patients with chronic diseases in Ethiopia.

Authors(study year)	Study area	Region	Sample size	Response rate	Vaccine acceptance	Good knowledge	Favorable attitude	Quality scores
Adella GA et al. (2021)^Citation44	Wollaita	Central Ethiopia	409	97%	238	324	290	8
Berihun et al. (2021)^Citation24	South Wollo	Amhara	416	98.6%	247	261	298	7
Daba et al. (2021)^Citation40	West Gujji	Oromia	408	100%	161	161	144	6
Dagnew B. et al. (2022)^Citation41	South Gondar	Amhara	422	100%	166	168	172	8
Kebede M et al. (2022)^Citation45	Bahirdar	Amhara	280	100%	170	136	176	7
Kassa M et al. (2021)^Citation46	Gondar	Amhara	423	100%	270	221	–	6
Taddele A. (2021)^Citation47	Addis Ababa	Addis Ababa	422	100%	152	–	–	6
Tegegne et al. (2021)^Citation48	Gondar	Amhara	401	94.8%	219	140	162	7
Eden A. et al. (2021)^Citation43	Northwest Ethiopia	Amhara	422	99.76%	213	215	–	6
Melak K. et al. (2022)^Citation26	Hadiya zone	Central Ethiopia	371	100%	319	160	212	7
Mesele et al. (2022)^Citation28	Woldiya	Amhara	386	100%	319	312	297	8
Mesfin Y et al. (2021)^Citation29	Guragie-Butajira	Central Ethiopia	398	93.8%	134	90	–	6

Despite 66.67% being conducted in the year 2021,^{24,29,41,42,45,47–49} around 50% of the articles were published in 2022.^{28,41,44,46,48} The other four studies were conducted in 2022.^26,28,41,45 All articles included in the review were hospital-based cross-sectional studies. The findings revealed from these articles were inconsistent and inconclusive, particularly in the magnitude of COVID-19 vaccine acceptance. In the same way, a varied strength of association of knowledge and attitude with the COVID-19 vaccine acceptance was reported, although almost all studies revealed a positive association (Table 1).

Test for heterogeneity

The I-squared (I² = 98.5%, P-value <.001) and Q statistics (Q¹¹ = 734.88, P-value <.001) show that the studies included in this review had considerable heterogeneity. Based on the Tau-squared result, 4% heterogeneity was detected between studies. However, 94.5% of the heterogeneity was within the studies. In addition, H² result (H² = 66.81%) also showed a high amount of variance was explained by random effects analysis rather than fixed effects analysis, which indicated that there was no homogeneity of the studies included in this review to be handled by fixed effects analysis.

Furthermore, the heterogeneity of the studies in this review was also evaluated using a Galbraith plot. However, none of the studies were out of the 95% CI, and there was no evidence of heterogeneity (Figure 2).

Figure 2. Galbraith plot for the studies included in the systematic review and meta-analysis on the COVID-19 vaccine acceptance and its association with knowledge and attitude among patients with chronic diseases in Ethiopia.

Magnitude of the COVID-19 vaccine acceptance

Significant heterogeneity (I² = 98.5, P-value <.001) was detected in included studies. And the pooled magnitude of the COVID-19 vaccine acceptance was determined using the DerSimonian and Laird (DL) random effects model. The pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases, who had follow-up at the hospital level, was 55.40% (95%CI: 44.70%, 66.10%) (Figure 3).

Figure 3. Pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia from 2019 to 2023.

Since there was statistically significant heterogeneity, a meta-regression analysis was performed to determine the source of the heterogeneity and to ensure accurate interpretation of the results. However, no significant variable that could account for the heterogeneity was discovered by meta-regression analysis by study-level variables: region, study year, quality scores, sample size, publication year, and response rate of included studies. Therefore, other variables not evaluated in this review could explain the heterogeneity (Table 2).

Table 2. Meta-regression analysis of the study-level variables to explain the sources of heterogeneity for meta-analysis of the COVID-19 vaccine acceptance and its association with knowledge and attitude among patients with chronic diseases in Ethiopia.

S. No	Variables	Coefficients	P-value
1	Study area	0.0320734	.324
2	Region	−0.1334455	.132
3	Study year	−0.4416045	.204
4	Publication year	0.0996112	.326
5	Publication status	0.2153286	.097
6	Sample size	−0.0069386	.210
7	Response rate	0.0014038	.397
8	Quality scores	0.1403256	.263

Subgroup analysis

Since, there is heterogeneity between the primary studies, the pooled magnitude was evaluated using subgroup analysis. The subgroup analysis was performed using study-level variables such as the region where the study was conducted, publication status, sample size, response rate and quality level of the study. The subgroup analysis showed that a slight effect size difference was detected between groups (Figures 4–10).

Figure 4. Subgroup analysis by region for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Figure 5. Subgroup analysis by study year for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Figure 6. Subgroup analysis by publication status for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Figure 7. Subgroup analysis by study’s sample size for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Figure 8. Subgroup analysis by level of response rate for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Figure 9. Subgroup analysis by level of study’s quality for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Figure 10. Sensitivity analysis for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

The pooled magnitude of the COVID-19 vaccine acceptance was less than by 1% in the Amhara region than in the central Ethiopian region. However, the pooled magnitude of the central Ethiopian region was statistically insignificant (Figure 4).

In the same way, the pooled magnitude of the studies’ result conducted in 2021 was less than by 17% from the pooled magnitude of the studies’ result performed in 2022 (pooled magnitude with 95% CI: 50%; 41%, 58%: 67%; 46%, 88%, respectively) (Figure 5). Furthermore, the pooled magnitude was higher from unpublished studies than from published studies (Figure 6).

In this review, most of the studies had a sample size between 400 and 423. However, some of the studies had a sample size of less than 400. The pooled magnitude obtained from the studies having a sample size less than 400 was higher than the pooled magnitude of the studies having a sample size 400 and above (Figure 7). More than 58% of the studies included in this review had a 100% response rate, and we classified them as having a high response rate for the subgroup analysis. However, five of the primary studies had a response rate between 90 and 99.9% and were classified as having a low response rate. And the pooled magnitude of studies with a high response rate was higher (58%) than the studies with a low response rate (51%) (Figure 8).

According to the JBI Quality Assessment Checklist of the analytical cross-sectional study, around 58% of the studies have high quality scores (having a higher mean score = 6.5). Of the total primary studies included in this review, five of them had a quality score of 6 for each, classified as low quality. High-quality studies had a higher pooled magnitude (63%) than low-quality studies (45%) (Figure 9).

Sensitivity analysis

To determine the influence of a single study on the pooled magnitude, a sensitivity analysis was performed using a random effects model. The analysis did not find statistically significant evidence for the influence of a single study (Figure 10).

Publication bias

Using funnel plots and Egger and Begg statistical tests at the 5% significant level, the existence of publication bias was evaluated. The Begg test did not produce statistically significant results, with p-values of 0.7317. However, the Egger test showed a statistical proof of publication bias with p-value = .014; and the funnel plot was also asymmetric (Figure 11).

Figure 11. Funnel plot showing publication bias for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Trim and fill analysis

Finally, a trim and fill analysis was performed to minimize the effects of publication bias on the pooled effect. After the model was adjusted, the pooled estimate was unchanged (Figure 12).

Figure 12. Funnel plot after trim and fill analysis for the pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Association of knowledge with the COVID-19 vaccine acceptance

Primary studies that reported on the effect of knowledge on COVID-19 vaccine acceptance showed variable levels of association strength with COVID-19 vaccine acceptance. For instance one study reported that knowledge of COVID-19 vaccines had no association with the COVID-19 vaccine acceptance among patients with chronic diseases.⁴⁵ However, others reported that it has a weak, moderate, and strong level of association for COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.^{24,26,28,29,41,42,44,45,47–49} Therefore, using the DerSimonian and Laird method, the pooled effect of good knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia was 2.64 (OR = 2.64, 95% CI 1.74, 4.07) times higher than among patients with chronic diseases having poor knowledge (Figure 13).

Figure 13. Forest plot of the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Subgroup analysis

The primary studies were heterogeneous, and subgroup analysis was used to assess the pooled effect of knowledge on the COVID-19 vaccine acceptance.

Subgroup analysis by study year

The pooled odds of the COVID-19 vaccine acceptance among articles conducted in 2021 were three (OR = 3.00, 95% CI: 1.86, 4.85) times higher compared to the pooled odds of the COVID-19 vaccine acceptance among articles conducted in 2022 (OR = 1.93, 95% CI: 0.93, 4.75), although the pooled effect of the studies conducted in 2022 was statistically insignificant (Figure 14).

Figure 14. Subgroup analysis by study year for the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Subgroup analysis by publication status

In this review, published studies revealed a statistically significant pooled effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia compared to unpublished studies (Figure 15).

Figure 15. Subgroup analysis by publication status for the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Subgroup analysis by sample size

Studies with a sample size ≥400 revealed a slightly higher statistically significant pooled effect of knowledge toward COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia than studies having <400 sample sizes (Figure 16).

Figure 16. Subgroup analysis by sample size for the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Subgroup analysis by study quality status

In this review, studies with lower quality revealed a slightly higher statistically significant pooled effect of knowledge toward COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia than studies with a better quality status (Figure 17).

Figure 17. Subgroup analysis by quality status for the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Publication bias

The studies included in this review to determine the pooled effect of knowledge on COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia, separately assessed the publication bias using a graphical and statistical method. Although subjective, the funnel plot (the graphical method) shows that there was publication bias (Figure 18). However, using statistical methods: the Egger and Begg test (Prob > |t| = 0.2421 and Prob > |z| = 0.2831 respectively) show that there is no statistically significant publication bias to the pooled effect of knowledge toward COVID-19 vaccine acceptance.

Figure 18. Funnel plot for the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Sensitivity analysis

In the same way, the studies included in this review to determine the pooled effect of knowledge on COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia separately evaluated the effect of small studies on the pooled effect using sensitivity analysis. Although subjective, the graph showed that there was no evidence for the presence of small study effect (Figure 19).

Figure 19. Sensitivity analysis for the effect of knowledge on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Association of attitude with COVID-19 vaccine acceptance

The effect of attitude toward COVID-19 vaccine acceptance was reported in primary studies with different levels of strength of association. The primary studies report has shown that it has a weak, moderate, and strong level of association for COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.^{24,26,28,40,45,48} And strong statistical evidence of heterogeneity was also detected from included studies to determine the pooled effect of attitude toward COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia. Therefore, the pooled magnitude was determined by a random-effects model using the DerSimonian and Laird method. The pooled effect of favorable attitude on COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia was 3.42 (OR = 3.42, 95%CI: 1.93, 6.09) times greater than that of patients with chronic diseases who had an unfavorable attitude (Figure 20).

Figure 20. Forest plot for the pooled effect of attitude toward the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Subgroup analysis

Therefore, the primary studies included to determine the pooled effect of attitude toward COVID-19 vaccine acceptance had heterogeneity, and the pooled effect of attitude toward COVID-19 vaccine acceptance also interpreted using subgroup analysis.

Subgroup analysis by publication status

Around 67% of the studies included to determine the pooled effect of the attitude toward the COVID-19 vaccine acceptance were published. And the pooled effect of favorable attitude toward COVID-19 vaccine acceptance was greater among published studies than among unpublished studies (Figure 21).

Figure 21. Subgroup analysis by publication status for the effect of attitude on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Subgroup analysis by sample size

Although an equal proportion of the studies included to determine the pooled effect of attitude toward COVID-19 vaccine acceptance had a sample size of <400 and ≥400, the pooled effect of favorable attitude toward COVID-19 vaccine acceptance was higher among studies that had a sample size ≥400 than among studies that had a sample size of <400 (Figure 22).

Figure 22. Subgroup analysis by sample size for the effect of attitude on the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Sensitivity analysis

In the same way, the studies included in this review to determine the pooled effect of attitude toward COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia separately evaluated the effect of small studies on the pooled effect using sensitivity analysis. Although it is subjective, the graph shows that there was no statistically significant evidence for the presence of small studies effect (Figure 23).

Figure 23. Sensitivity analysis of the effect of attitude toward the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia.

Publication bias

The publication bias was evaluated using statistical methods, particularly for the included studies, to determine the effect of attitude toward COVID-19 vaccine acceptance model. The Eggers and Begg test (Prob > |t| = 0.9932 and Prob > |z| = 0.9999 respectively) showed that there is no statistically significant publication bias to the effect of attitude toward COVID-19 vaccine acceptance.

Discussion

The death rate among populations with chronic health conditions who are unvaccinated or partially vaccinated for the COVID-19 vaccines is significantly higher compared to those who have received the vaccines.¹⁶ Increasing the COVID-19 vaccines immunization rate is necessary to reduce the global burden of COVID-19 related illnesses and deaths in these populations.⁸ However, studies have shown that the acceptance and uptake rate of COVID-19 vaccination among populations with chronic diseases is relatively lower than expected standards and the magnitude also varies between regions.^23–30–37 The degree to which vaccinations have been accepted will determine the uptake rates of the COVID-19 vaccine.³⁷ Therefore, this review identified the magnitude of COVID-19 vaccine acceptance among patients and/or individuals with chronic diseases in Ethiopia.

Our meta-analysis revealed that the pooled magnitude of COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia was 55.4% (95%CI: 44.70%, 66.10%). Even though it is unknown which portion of the populations with chronic diseases need to receive a COVID-19 vaccination in order to establish herd immunity, the magnitude of COVID-19 vaccine acceptance was relatively low. This could be associated with the majority of the Ethiopian populations exhibiting comparatively low levels of COVID-19 vaccine acceptance compared to the globe.³⁷ This variation may have been caused by lower mortality rates from the COVID-19, restricted access to multiple COVID-19 vaccines, and a delayed supply of COVID-19 vaccines. It may also have been due to a general negative belief and attitude about the safety and efficacy of the COVID-19 vaccines¹⁹; belief on traditional medicines to cure COVID-19 disease, and the general belief of the country’s population that they are less susceptible to the disease. The previous study’s finding revealed that with an increase in comorbidity, there is a greater likelihood of experiencing morbidity and mortality as a result of COVID-19.¹¹ However, in Ethiopia, not only the uptake of the COVID-19 vaccine, the acceptance level by the patients having chronic diseases was also low.^23,41

This finding is consistent with the COVID-19 vaccine acceptance level around the globe, reported 65%.¹⁹ Increasing the COVID-19 vaccine acceptance to the level more than this magnitude is of paramount importance to increase the vaccine uptake.^21,32 This is important to decrease the concomitant morbidity, mortality and hospital admission rate of patients with chronic diseases and COVID-19.^10,23 Therefore, communication-based interventions that take into account sociocultural and historical factors are crucial for populations with chronic health conditions. These interventions also require the integration of culture-affirming approaches, which have the potential to reduce vaccine hesitancy and increase the uptake of the COVID-19 vaccines.³³ Additionally, culturally sensitive interventional techniques and community involvement are needed to increase the acceptance of the COVID-19 vaccination. This could have a major positive impact on the uptake of COVID-19 vaccines by patients with chronic diseases.

Our meta-analysis’s finding is lower than the findings of the study done in Hungary (88.62%).³⁶ This disparity could be associated with a slight differences in the study participants. In our review, all patients with chronic diseases who underwent follow-up in the hospital setting were included. However, Hungary’s study included only patients with those suffering from long term obstructive lung diseases.³⁶

Those patients with chronic diseases who had good knowledge about the COVID-19 vaccines had a higher probability of accepting the vaccine than their counterparts. This finding is in agreement with the previous study conducted in the general populations.^23,59 Having good knowledge of the vaccines could lead to a positive association with the COVID vaccine acceptance. Furthermore, most of the primary and review literatures prior to this date revealed that knowledge had a positive association with the COVID-19 vaccine acceptance.^{23,37,40,41,43,44,59} Until the end of 2023, Ethiopia has been struggling with poor vaccination acceptance rates among populations with chronic diseases, so the favorable correlation between good knowledge and the COVID-19 vaccine acceptance is critical. Increasing community participation, education, and awareness campaigns on the COVID-19 vaccines can help improve vaccination acceptance rates among patients with chronic diseases in Ethiopia. Together, government and healthcare professionals can eliminate myths and provide correct information about the vaccines while encouraging people to get vaccinated.

Furthermore, patients with chronic diseases who had a favorable attitude toward COVID-19 vaccines had a higher chance of accepting the vaccines than those who had an unfavorable attitude. This finding is in agreement with the previous study done in the general populations.⁵⁹ This stated that having a positive attitude toward the vaccines had a positive association with the COVID-19 vaccine acceptance. In the same way, previous studies have shown that favorable attitude had a positive association with COVID-19 vaccine acceptance.^{24,25,40,41,43} It is critical for Ethiopia that there is a favorable association between the COVID-19 vaccine acceptance and a hopeful attitude. This will also motivate the healthcare professionals working with these patients by improving their sense of the safety and effectiveness of the vaccines.

This review has certain strengths and limitations. It included a number of published and unpublished studies conducted in Ethiopia. The PRISMA guideline was strictly followed in all steps of systematic review and meta-analysis. However, the data collected to estimate the pooled magnitude was from hospital-based studies, so no primary studies were found during this review. This may have introduced bias since the population in these studies may not represent the general population. In Ethiopia, studies are published in the English language and this review included only these studies. Although a trim-and-fill analysis of random effects model of estimation was used, an unidentified source of heterogeneity and publication bias was detected in the primary studies included in this review. The quality assessment also showed evidence of poor quality in some of the primary studies, which can affect the findings. However, we conducted a proper subgroup analysis based on the quality of the included studies. Furthermore, most of the articles included in this review evaluated the association of knowledge and attitude with the COVID-19 vaccine acceptance, however they did not fully report the results. These were not included for the estimation of the association of knowledge and attitude. This review did not also include qualitative studies. This might be a possible future research area.

Conclusions

The overall pooled magnitude of the COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia was 55.4%. Good knowledge about the COVID-19 vaccines and favorable attitude toward the vaccines had a positive association with COVID-19 vaccine acceptance among patients with chronic diseases in Ethiopia. Therefore, policy makers should prioritize policies and programs that facilitate COVID-19 vaccine acceptance among patients with chronic diseases to increase the vaccines acceptance and uptake. In addition, healthcare professionals and healthcare guide developers should work more to improve the knowledge and attitude of patients with chronic diseases through providing tailored information’s and by improving health care professionals’ engagement in vaccine promotion education at patients with chronic diseases’ follow-up clinic. Finally, systematic review and meta-analysis is required by considering community-based and qualitative studies.

Authors’ contribution

TDT &AMK-conceptualization, first write-up and quality assessment; TMD, ZBA, BA&MY- software, supervision and validation; GMB, MAY and AGY- literature searching, and data extraction AFA, MGM &TA-data validation and crosschecking

Data availability

All data used are within the manuscript and it’s supporting information files.

Acknowledgments

We thank all authors of the studies included in this systematic review and meta-analysis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental data

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2350815

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.