COVID-19 Knowledge, Attitudes, and Vaccine Hesitancy in Ethiopia: A Community-Based Cross-Sectional Study

The current healthcare system’s efforts to reduce the spread of COVID-19 in Ethiopia and limit its effects on human lives are being hampered by hesitancy toward the COVID-19 vaccine. The aim of this study was to assess the knowledge levels, attitudes, and prevention practices of COVID-19, in the context of the level of vaccine hesitancy with other associated factors in Ethiopia. A community-based cross-sectional design with mixed-method data sources was employed. It comprised 1361 study participants for the quantitative survey, with randomly selected study participants from the studied community. This was triangulated by a purposively selected sample of 47 key informant interviews and 12 focus group discussions. The study showed that 53.9%, 55.3%, and 44.5% of participants had comprehensive knowledge, attitudes, and practices regarding COVID-19 prevention and control, respectively. Similarly, 53.9% and 47.1% of study participants had adequate knowledge and favorable attitudes toward the COVID-19 vaccine. Only 29.0% of the total survey participants had been vaccinated with at least one dose of vaccine. Of the total study participants, 64.4% were hesitant about receiving the COVID-19 vaccination. The most frequently reported reasons were a lack of trust in the vaccine (21%), doubts regarding the long-term side effects (18.1%), and refusal on religious grounds (13.6%). After adjusting for other confounding factors, geographical living arrangements, the practices of COVID-19 prevention methods, attitudes about the vaccine, vaccination status, perceived community benefit, perceived barriers toward vaccination, and self-efficacy about receiving the vaccine were significantly associated with vaccine hesitancy. Therefore, to improve vaccine coverage and reduce this high level of hesitancy, there should be specifically designed, culturally tailored health education materials and a high level of engagement from politicians, religious leaders, and other community members.


Introduction
The COVID-19 pandemic is a worldwide public health, social, and economic threat, against which efforts have been made to prevent and control its spread around the globe. This global disease infected more than 673 million population; of them, 6.7 million have already died up until January 2023. Globally, there are many therapeutic and non-therapeutic mitigation measures that have been implemented to reduce the transmission and the effect of COVID-19 on various systems [1,2].
COVID-19 has caused a global public health crisis affecting most countries, including Ethiopia, in various ways, including a devastating effect on the world economy and healthcare systems [1]. The COVID-19 pandemic is at a paradoxical stage, with vaccine roll-out initiated but with a still significantly elevated level of infections and deaths, while there are still delays in the acceptance or refusal of vaccines, which results in vaccine 2.2. Sample Size Determination 2.2. 1

. Quantitative Sample Size Determination
The sample size of the study was determined using a single population proportion formula, considering the objectives of the research. Since there was no previous study conducted in similar study sites, we used a 50% proportion for knowledge, attitude, and hesitancy regarding vaccines and the preventive and control methods practiced to gain the maximum sample size. In addition, the sample size calculation considered 95% CI, a 5% margin of error, a 20% non-response rate, and three design effects for its multistage sampling procedure. Finally, the calculated sample size was 1382 vaccinated and unvaccinated community members. The calculated sample size was distributed based on the age and sex of the respondent (see Table 1).
A multistage sampling procedure was applied to recruit study participants within the community. In stage 1, 16 cities were selected randomly, and the sample was allocated based on the population proportion. In stage 2, in each city, a sample of health centers and their population catchment areas were identified. In stage 3, in each health center catchment area, one village was selected using a simple random sampling approach. In the villages, every other household was selected using a systematic random sampling procedure, and either the head of the household or the wife was considered for the interviews.
The data collection method was developed after reviewing the literature from various countries and adapting it to the context. Additionally, the study was piloted with 50 samples in the local language. Following the pilot test, the measurement tools were modified before being used for actual data collection. Data were collected electronically using tablet/smartphone applications, ODK/KOBO, where structured questionnaires with precoded answers were uploaded. The questionnaire was developed after reviewing the different studies in the relevant literature and was then modified, based on the project outcome indicators. The data measurement tool was developed after rigorous review for various measurements [17][18][19]. In addition, we used the health belief model to assess COVID-19 vaccine hesitancy; a systematic review indicated that health belief models (HBM) is a useful model in predicting COVID-19 vaccine hesitancy [20]. Therefore, the errors associated with this part of the tool are minimal. Additionally, the research questionnaire was pretested and modified before the actual data collection took place. The questionnaire (Table S1) included supplementary information. To ensure reliability and validity, our team gave intensive training to the research data collectors, including a pretest; the purpose of the research was clearly explained to study participants and the data collection was supervised by the research team. In addition, data sources and methodological triangulations were used to make the evidence reliable and valid. The local language was used during the data collection interviews to ensure the intelligibility of concepts about vaccine hesitancy; data collectors needed to speak the local language and understand the culture of the community. The research was carried out in urban areas where the public is more aware of COVID-19 and its vaccine. Furthermore, the pandemic engaged the population when it occurred, encouraging everyone to learn about its control, prevention, and vaccines. Throughout the pandemic, the government and non-governmental organizations, all media (including social media), and various sectors provided information about COVID-19; therefore, the community had no difficulty understanding the study questions. In addition, the content validity of the instrument was assessed by different experts from each study area. All the above efforts improved the study's reliability, intelligibility, and face validity.

Qualitative Sample Size
A purposive sample that included a total of 35 key informant interviews was conducted (with key players from an immunization program, the Ministry of Health, six Regional Health Bureaus for immunization, and eight cities/sub-cities). These key informants included eight health extension, eight health development, and nine healthcare workers who participated in the COVID-19 vaccine campaign services. In addition, 12 focus group discussion (FGDs) among the young males and females and adult men and women were included. Additionally, a semi-structured interview guide was used to facilitate the key informant interviews to obtain insightful qualitative information regarding the key issues to be examined in this research. All interviews were tape-recorded after gaining consent from each baseline survey participant.

Variable Measurements and Categorization
Knowledge of COVID-19's signs and symptoms: To measure the respondents' knowledge of the COVID-19 signs and symptoms, a total of 8 questions were posed. "Yes" responses were coded "1" and "no" responses were coded "0". The mean score was used to categorize the study participants as knowledgeable or not knowledgeable. Those who scored above the mean score were knowledgeable, while those who scored below the mean score were considered not knowledgeable.
Knowledge of sources of COVID-19 infection: To measure the respondents' knowledge of the sources of infection, a total of 3 questions were used. "Yes" responses were coded "1" and "no" responses were coded "0". The mean score was used to categorize participants as knowledgeable or not knowledgeable. Those who scored above the mean score were knowledgeable, while those who scored below the mean score were considered not knowledgeable.
Knowledge of COVID-19 infection risk groups: To measure the respondents' knowledge of the population groups at risk, a total of 3 questions were used. "Yes" responses were coded "1" and "no" responses were coded "0". The mean score was used to categorize participants as knowledgeable or not knowledgeable. Those who scored above the mean score were knowledgeable, while those who scored below the mean score were considered not knowledgeable.
Knowledge of COVID-19 prevention methods: To measure the respondents' knowledge of the means of transmission, a total of 6 questions were used. "Yes" responses were coded "1" and "no" responses were coded "0". The mean score was used to categorize participants as knowledgeable or not knowledgeable. Those who scored above the mean score were knowledgeable, while those who scored below the mean score were considered not knowledgeable.
Overall comprehension (knowledge) of COVID-19 prevention and control options: To measure the respondents' overall knowledge of COVID-19 prevention and control, a total of 30 questions were used. "Yes" responses were coded "1" and "no" responses were coded "0". The mean score was used to categorize participants as knowledgeable or not knowledgeable. Those who scored above the mean score were knowledgeable, while those who scored below the mean score were considered not knowledgeable.
Attitude toward COVID-19 prevention and control: The respondents were asked about their attitude toward COVID-19 prevention and control activities, using 7 questions with three Likert scale scales (agree, neutral, and disagree). The responses' mean value was calculated, and those who scored above the mean were recorded as having a favorable attitude, while those who scored below the mean were recorded as not having a favorable attitude toward prevention and control activities.
COVID-19 prevention and control measures: The respondents were asked about their levels of practicing prevention and control measures, using 7 questions with 3 Likert scale scales (never, sometimes, and often). The responses' mean value was calculated, and those who scored above the mean were recorded as having practiced the measures, while those who scored below the mean were recorded as not practicing them.
Knowledge of the COVID-19 vaccine: The respondents were asked to list all the COVID-19 prevention and control methods, and if the respondent mentioned the "COVID-19 vaccine" as a method, the respondent was considered knowledgeable. Otherwise, those who did not mention one mechanism for COVID-19 prevention and control were considered not knowledgeable. Attitude about the COVID-19 vaccine: The respondents were asked about their attitude toward taking the COVID-19 vaccine; "yes" responses were recorded as having a favorable attitude, while "no" responses were recorded as having an unfavorable attitude.
Vaccination status: Taking either of the available COVID-19 vaccines regardless of the stage was considered "vaccinated".
Vaccine hesitancy: Vaccine hesitancy was measured differently for vaccinated and nonvaccinated individuals separately. Individuals who have been vaccinated are considered to have "vaccine hesitancy" if they are hesitant or regret taking the vaccine after being vaccinated, whereas, for non-vaccinated individuals, the respondent was asked about their plan to take the vaccine and if they responded, "I have no plan", the response was recorded as "COVID-19 hesitancy".
Perceived COVID-19 infection susceptibility: A total of 4 questions with a 5-point Likert scale (1: strongly disagree to 5: strongly agree) were used to measure perceived COVID-19 infection susceptibility. Those who scored above the mean value were categorized as having "perceived COVID-19 infection susceptibility". Those who scored below the mean value were considered to not have perceived COVID-19 infection susceptibility.
Perceived COVID-19 infection severity: A total of 3 questions with a 5-point Likert scale (1: strongly disagree to 5: strongly agree) were used to measure perceived COVID-19 infection severity. Those who scored higher than the mean value were classified as having a high perception of COVID-19 infection severity. Those who scored at the mean or lower were considered to have a low perception of COVID-19 infection severity.
Perceived benefits of the COVID-19 vaccine to the community: A total of 2 questions with a 5-point Likert questionnaire (1: strongly disagree to 5: strongly agree) were used to measure the perceived benefits of the COVID-19 vaccine to the community. Those who scored higher than the mean value recognized the community benefit of the COVID-19 vaccine, while those who scored lower saw no COVID-19 vaccine community benefit.
Perceived COVID-19 vaccine benefits for individuals: To assess the perceived benefits of the COVID-19 vaccine's benefits for individuals, three questions on a five-point Likert scale (1: strongly disagree to 5: strongly agree) were used. Those who scored higher than the mean value were classified as having perceived benefits for individuals from the COVID-19 vaccine. Those who scored lower than the mean value were deemed to have no perceived COVID-19 vaccine benefit to individuals.
Perceived COVID-19 vaccine intake barriers: A total of 3 questions on a 5-point Likert scale questionnaire (1: strongly disagree to 5: strongly agree) were used to measure perceived barriers to receiving the COVID-19 vaccine. Those who scored above the mean value were categorized as having "perceived barriers" to receiving the COVID-19 vaccine. Those who scored at the mean or lower were deemed to have no perceived barrier to receiving the COVID-19 vaccine.
COVID-19 self-efficacy and the need to receive the COVID-19 vaccine: A total of 3 questions with a five-point scale Likert questionnaire (1: strongly disagree to 5: strongly agree) were used to measure the self-efficacy of the individuals regarding taking the COVID-19 vaccine. Those who scored above the mean value were categorized as having self-efficacy to receive the COVID-19 vaccine. Those who scored lower than the mean value were thought to have insufficient self-efficacy regarding receiving the COVID-19 vaccine.

Quantitative Analysis
The quantitative data were cleaned, coded, and analyzed using STATA 16. The descriptive statistics presented tables and figures. Additionally, binary logistic regression analysis was used. The results of the multivariable binary logistic regression analyses were presented using an adjusted odds ratios, with 95% CI and a p-value of less than 0.05 as the level of significance.

Qualitative Analysis
The qualitative data analysis involved the thematic coding of transcribed and translated in-depth interviews and focus group discussions. A hybrid coding approach, which included the process of creating pre-set and emergent codes were used. Before the data collection and the coding process, pre-set codes were derived from the objectives and questions. Accordingly, a codebook was prepared which included the list of pre-codes. Additionally, the ideas, concepts, actions, relationships, and meanings that emerged from the data that were different from the pre-set codes were used as the emergent codes. The data were then analyzed using a thematic approach by conducting an ongoing content analysis.

Ethical Considerations
Ethical clearance was obtained from the FMOH. Permission was secured from the respective regional, zonal, district, kebele, and community level leaders. Informed consent was obtained from each study participant after an explanation of the purpose of the assessment. Confidentiality was ensured from all data collectors and the principal investigator via coding numbers, not names, and by keeping the responses locked.

Sociodemographic Characteristics
Of the total 1382 respondents calculated for the sample size, 1361 (98.5% response rate) study participants were included in the analysis. Regionally speaking, a relatively large proportion of the sample was obtained from the Somali Region 303 (22.3%) (see Table 2). Of the total number, 683 (50.2%) were male and 678 (49.8) were female respondents. More than three-quarters (76.7%) were adult men and women whose age was over 24 years, and about one-quarter (23.3%) of survey participants were youth/adolescent females and males (18-24 years old). In terms of the participants' religious affiliation, almost half, 641 (47.1%), of the study participants were Orthodox Christian religion followers, while almost two-fifths, 536 (39.4%), were Muslim. In this study, 508 (37.3%) had a monthly income of less than the median ETB 6500, and 705 (51.8%) had less than the median ETB 5000 monthly expenditure. Table 3 indicates that 657 (53.9%) survey participants had adequate knowledge of COVID-19 prevention and control activities, with statistically significant differences by region and age (p < 0.001). The maximum percentage with adequate knowledge was observed in the Sidama Regional State (71.55%), and the minimum percentage was observed in the Somalia Regional State (37%).

Knowledge, Attitude, and Practice Regarding COVID-19 Infection Prevention and Control
Further analysis showed that 59% of the respondents had adequate knowledge of means of transmission and 37.2% had adequate knowledge of prevention methods against COVID-19. Figure 1 below summarizes the level of knowledge for different components regarding COVID-19 prevention and control.
The study showed that almost three-quarters of the survey participants (74.1%) who were infected with COVID-19 may not have shown signs and symptoms. The most frequently mentioned signs and symptoms included coughing, sneezing, congestion, and a sore throat (84%). A total of six questions were used to assess the respondents' knowledge about the means of transmission. Taking the mean value as a cutoff point to classify the respondents' knowledge about COVID-19, about 63.6% had adequate knowledge of the transmission methods of COVID-19 infections. Regarding knowledge about COVID-19 transmission, 72.8% said that it was transmitted by touching contaminated objects, and 62.1% mentioned droplets as a means of transmission of COVID-19 infections. According to the study participants, the two most frequently mentioned conducive environments for COVID-19 were crowded places (90.5%) and people traveling across cities (89.4%). Moreover, the study respondents reported people aged 65 years or older (81.5%), with non-communicable diseases/comorbidities (59.3%), and people with known respiratory diseases (47.5%) to be at-risk population groups for COVID-19 infection severity and the related complications. About 8.5% of the study participants did not know who was at risk from COVID-19 infection. Adequate knowledge of COVID-19 prevention and control methods is essential to stop the spread of the pandemic. This study showed that most participants (88.2%) knew the importance of wearing masks, while 84.9% knew that regular hand-washing with soap and water and physical distancing (avoiding crowded/public places) were prevention control methods for reducing the transmission of COVID-19.  Further analysis showed that 59% of the respondents had adequate knowledge of means of transmission and 37.2% had adequate knowledge of prevention methods against COVID-19. Figure 1 below summarizes the level of knowledge for different components regarding COVID-19 prevention and control. The study showed that almost three-quarters of the survey participants (74.1%) who were infected with COVID-19 may not have shown signs and symptoms. The most frequently mentioned signs and symptoms included coughing, sneezing, congestion, and a sore throat (84%). A total of six questions were used to assess the respondents' knowledge about the means of transmission. Taking

Attitudes about Infection Prevention and the Control of COVID-19
The community's knowledge in itself was inadequate to prevent and control COVID-19. In addition, a favorable attitude toward the COVID-19 prevention and control options was essential to ensure practice. This study showed that 753 (55.3%) of survey participants had a favorable attitude regarding COVID-19 prevention and control activities. The qualitative findings were in line with the quantitative results. Even though many people knew about vaccinations for COVID-19, they may not have been vaccinated for various reasons, such as misconceptions, issues related to politics, and high living costs in the community: " . . . there are our neighbors who think that COVID-19 does not exist and give everything to their Lord. When you inform them to take preventive measures, they consider you are working against their religion. But we have seen many people die due to COVID-19" (21-year-old male youth from Bahirdar). Table 4 shows the details of the different attitude domains about COVID-19 prevention methods; for instance, nearly three-quarters (74%) of respondents believed that COVID-19 was government or media propaganda, and more than half (57.5%) believed that COVID-19 was a curse. As indicated in Table 5, below, the highest favorable attitude toward COVID-19 prevention options was observed in the Amhara Regional State and the lowest was in the Somali Regional State. About 53.4% of males and 57.2% of females had a favorable attitude toward COVID-19 prevention and control activities. Nearly half of the younger people (49.2%) and 57.2% of the adults had a favorable attitude toward COVID-19 prevention and control activities. Practices about COVID-19 Infection Prevention Table 6 shows a total of seven questions with three Likert scale items that were used to measure the level of COVID-19 prevention practices; only 39.0% of study participants were wearing masks often, 28.7% were socially distancing often, and 39.5% were using hand sanitizer often to prevent the spread of COVID-19. This study showed that 55.6% of the respondents followed inadequate COVID-19 prevention practices. The highest levels of COVID-19 prevention practices were observed in the Amhara Regional State (73.08), and the lowest were observed in the Somali Regional State (32.3%) (p < 0.001).

Level of COVID-19 Testing
Of the total study participants, 33.8% were tested for COVID-19 after the pandemic period began; of those who were tested, 36 (2.6%) were positive for COVID-19. The remainder (66.2%) were not tested for COVID-19 for different reasons. More than half (55.3%) mentioned an absence of COVID-19 symptoms, fearing the test (27.5%), and a lack of trust (15.5%), which were the main reasons for not taking COVID-19 tests.

Knowledge, Attitude, and Vaccination Practices Regarding the COVID-19 Vaccine
Community knowledge and attitudes about the COVID-19 vaccine are a prerequisite to taking the vaccine. In this study, 733 (53.9%) had adequate knowledge about the COVID-19 vaccine as one of the infection prevention methods, and 720 (52.9%) had a favorable attitude about the importance of the COVID-19 vaccine as a COVID-19 prevention and control strategy. A statistically significant association was observed among the regions (see Table 7) and from having adequate knowledge about the COVID-19 vaccine as a prevention option (p < 0.001). Among the Regional States, the highest knowledge level was observed in the Sidama Regional State (75.65%), and the lowest frequency level was observed in the Somali Regional State (33.0%). Forty-seven percent of respondents had a favorable attitude toward the COVID-19 vaccine. Regional variations regarding attitudes toward the COVID-19 vaccine were observed; the highest favorable attitude about the vaccine was observed in the Amhara Regional State (52.56%), while the lowest was observed in the Oromia region (39.01%).
Only 29.0% of the total survey participants were vaccinated with at least one dose of vaccine since the vaccine had been introduced. Regional variations were observed in vaccination status; the highest vaccination status level was observed in the Sidam Regional State (55%) and the lowest in the Somali Regional State (14.5%). The vaccination status was also higher among adult population groups (32.0%) in comparison to young people (20.0%). As indicated in Table 7 below, there is a statistical difference in vaccination status based on the region (p < 0.001) and age (p < 0.0.001) of the survey participants.
The proportion of unvaccinated study respondents was as high as 71%, for different reasons. According to the self-reported reasons, these included a lack of trust in the vaccine (18.0%), religious reasons (16.8%), lack of time to attend the appointment (11.7%), and lack of access or eligibility issues with the vaccine (38%). Of those vaccinated study participants, 65.1% of them did not know what type of vaccine they received. Only 15.7%, 13.2%, 5.3%, and 0.8% of survey participants knew that they received the Johnson and Johnson, Astra Zeneca, and Pfizer vaccines, respectively. Only 8.4% of the study participants chose the type of vaccine and the remaining received whatever vaccine was available without knowing the existing alternatives.

Self-Reported Vaccine Adverse Effects
Of the total vaccinated survey participants, 222 (56.2%) reported vaccine-related mild to moderate adverse effects in this survey. As indicated in Figure 2, of those who reported vaccine adverse effects, 57.7%, 46.8%, and 44.6% reported fever, headache, and fatigue as adverse effects, respectively. available without knowing the existing alternatives.

Self-Reported Vaccine Adverse Effects
Of the total vaccinated survey participants, 222 (56.2%) reported vaccine-related mild to moderate adverse effects in this survey. As indicated in Figure 2, of those who reported vaccine adverse effects, 57.7%, 46.8%, and 44.6% reported fever, headache, and fatigue as adverse effects, respectively.

COVID-19 Vaccine Hesitancy
The findings of this study identify that of the 966 unvaccinated individuals, 729 (75.5%) hesitated to receive the vaccine. Similarly, the level of regret after receiving the COVID-19 vaccination was assessed for the vaccinated individuals. Of those vaccinated individuals, 148 (37.5%) regretted receiving the vaccine (Figure 3).

COVID-19 Vaccine Hesitancy
The findings of this study identify that of the 966 unvaccinated individuals, 729 (75.5%) hesitated to receive the vaccine. Similarly, the level of regret after receiving the COVID-19 vaccination was assessed for the vaccinated individuals. Of those vaccinated individuals, 148 (37.5%) regretted receiving the vaccine (Figure 3).
Johnson, Astra Zeneca, and Pfizer vaccines, respectively. Only 8.4% of the study participants chose the type of vaccine and the remaining received whatever vaccine was available without knowing the existing alternatives.

Self-Reported Vaccine Adverse Effects
Of the total vaccinated survey participants, 222 (56.2%) reported vaccine-related mild to moderate adverse effects in this survey. As indicated in Figure 2, of those who reported vaccine adverse effects, 57.7%, 46.8%, and 44.6% reported fever, headache, and fatigue as adverse effects, respectively.

COVID-19 Vaccine Hesitancy
The findings of this study identify that of the 966 unvaccinated individuals, 729 (75.5%) hesitated to receive the vaccine. Similarly, the level of regret after receiving the COVID-19 vaccination was assessed for the vaccinated individuals. Of those vaccinated individuals, 148 (37.5%) regretted receiving the vaccine (Figure 3).  We disaggregated vaccine hesitancy prevalence by the region, gender, and age of the survey participants. The highest frequency of vaccine hesitancy was observed in the Afar region (78.5%), while the lowest frequency was observed in the Sidama region (47.82%). As indicated in Table 8 below, there was a statistical difference in vaccine hesitancy based on region, but there was no statistically significant difference based on the age and gender of survey participants in the study area.

Reasons for Vaccine Hesitancy
Of those vaccinated individuals, 14.4% were not interested in recommending the vaccine to other people. Vaccinated survey participants were asked about their perceptions of the main reasons of those in the community who did not receive the vaccine. It was reported that 40.2% did not know about the vaccine, while the remaining 39.2% had no trust in the vaccine. Similarly, unvaccinated study participants (966) gave different reasons for being hesitant, including not trusting the vaccine (56%), the fear of long-term effects (47.2%), and negative reports about the vaccine (43.6%). The details are indicated in Figure 4 below. survey participants. The highest frequency of vaccine hesitancy was observed in the Afar region (78.5%), while the lowest frequency was observed in the Sidama region (47.82%). As indicated in Table 8 below, there was a statistical difference in vaccine hesitancy based on region, but there was no statistically significant difference based on the age and gender of survey participants in the study area.

Reasons for Vaccine Hesitancy
Of those vaccinated individuals, 14.4% were not interested in recommending the vaccine to other people. Vaccinated survey participants were asked about their perceptions of the main reasons of those in the community who did not receive the vaccine. It was reported that 40.2% did not know about the vaccine, while the remaining 39.2% had no trust in the vaccine. Similarly, unvaccinated study participants (966) gave different reasons for being hesitant, including not trusting the vaccine (56%), the fear of long-term effects (47.2%), and negative reports about the vaccine (43.6%). The details are indicated in Figure 4 below.   Of the total unvaccinated survey participants, 138 (25.1%) planned to receive the vaccine in the near future. Among those who showed an interest in receiving the vaccine, 75.4% did not know which vaccine they should receive, and about 11.6% were interested in receiving the Johnson & Johnson COVID-19 vaccines.

Myths and Misconceptions about the COVID-19 Vaccine
Political conspiracy: In addition to being suspicious of the vaccine's efficacy, the community had heard various rumors and misconceptions, such as politicians' agendas and religious views, resulting in the low utilization of the vaccine. The above reasons make vaccination programs challenging. For example, a 30-year-old female EPI coordinator and KII participant from FMOH stated: "Different rumors are circulating in the community. For example, the community raised issues like the vaccine is fake, the vaccine is political, the politicians have brought it for themselves, COVID-19 does not exist in Ethiopia, the vaccine is 666, and the vaccine is micro-chips. Things went wrong, and we had a big challenge when we started the vaccination programs" (EPI coordinator KII participant from FMOH).
Religion: According to the qualitative findings, there was a tendency to give different reasons for linking the COVID-19 vaccine to religion and belief. One of the KII participants from a health center in Yeka sub-city stated: "There is a strong belief in the community that what saves them is not wearing face masks or taking the vaccine, but it is their belief/religion".
The above statement also narrated the issue of religion and the COVID-19 vaccine by a community FGD participant; a 27-year-old male FGD participant from Batu stated: " . . . .as to my understanding, it is also related to religious reasons. I believe in God, and I live by the will of God rather than the vaccine. So, why do I take the vaccine?" Fear of Side Effects: The qualitative findings indicated that the fear of side effects was one of the main reasons for not taking the vaccine. Fear of vaccination is mentioned as one of the main reasons for being hesitant to receive COVID-19 vaccinations at all levels. An adult 45-year-old FGD participant from Bahir Dar stated "I am afraid of being vaccinated. I did not get vaccinated because I have seen many people get sick after taking the vaccine".

Perceived Susceptibility to COVID-19 Infection
Those who perceived their susceptibility to the COVID-19 infection were more likely to receive the vaccine. Of the total survey participants, 754 (55.4%) had a perceived susceptibility to COVID-19 infection. Regional variations were observed; the maximum perceived susceptibility was observed in the Oromia region (81%) and the minimum susceptibility was observed in the Afar Regional State (12.4%) (p < 0.001).

Perceived Severity of COVID-19 Infection
In the survey, 44.5% considered that COVID-19 infection and its complications were not severe. The qualitative study indicated a low risk perception in the community regarding COVID-19, affecting the vaccine intake. "Another main challenge is a low-risk perception on COVID-19 infection in our society. People believe that I already had a COVID-19 infection even though it did not show signs and symptoms, and I have no risk of reinfection again. So, why do I take [the] COVID-19 vaccine?" (KII at MOH).
Similarly, participants mentioned that there was no problem concerning COVID-19 prevention and control measures. The problem is associated with the low perceived risk of the disease. "I don't think there is a significant knowledge gap about how COVID-19 is transmitted and the prevention measures. People hear about it from many media. But we can see a huge gap regarding the preventive measures. There is a low risk perception among the community. As a result of this, hand washing, hand sanitizer usage, and wearing of face mask are very poor" (KII EPI unit at Hawassa).
Another 19-year-old respondent from Addis Ababa added, "Many people see the disease as simple. The previous panic and anxiety related to the disease have been lost. This has a great impact on the uptake of COVID-19 vaccines".

Perceived Individual and Community Benefits Regarding COVID-19 Vaccines
Evidence has shown that when individuals believe that receiving the vaccine is important to prevent COVID-19 from spreading in their community, the probability of being vaccinated is very high. As indicated in Tables 9 and 10 of the total participants, 794 (54.2%) perceived that receiving the COVID-19 vaccine had individual benefits, and 739 perceived that receiving COVID-19 had community benefits. Regional variations were observed in the perception of the study participants regarding the importance of the vaccine to the community. The highest (64.4%) importance was observed in Addis Ababa, and the lowest (44%) was observed in the Sidama Regional State (p < 0.001). Similarly, regional variations were observed in the perceptions of the respondents about the importance of the vaccine on the individual; the highest (70.4%) was observed in Addis Ababa, and the lowest (42.4%) in the Afar Regional State (p < 0.000) ( Table 9).

Perceived Barriers to Receiving COVID-19 Vaccines
This study showed that perceived barriers to receiving COVID-19 vaccines were reported by 836 (61.4%) respondents. Regional variations were observed in perceived barriers to COVID-19 vaccine uptake (p < 0.001); the highest (75.9%) perceived barrier to receiving the vaccine was observed in the Afar Region and the lowest (42.7%) in the Amhara Region (p < 0.001).

Self-Efficacy about the COVID-19 Vaccine
The assessment measured self-efficacy about receiving COVID-19 vaccines, and 720 (52.9%) had high self-efficacy about receiving the vaccine. The highest self-efficacy was observed in Addis Ababa (56.9%) and Sidama (56.9%) Regional States, and the lowest was observed in the Amhara Regional States (42.7%). A statistically significant association was observed between the regions and self-efficacy about the vaccine (p < 0.001).

Factors Associated with Vaccine Hesitancy
Using the city/sub-city administration as a cluster, data analysis was tested for a multilevel binary logistic regression model to identify the factors associated with vaccine hesitancy. However, in the empty model, the intra-class correlation coefficient (ICC) (1.07%) was less than 10%; in this case, classical regression was recommended to identify the factors associated with COVID-19 vaccine hesitancy.
To control the confounder variables and identify the independent factors with the dependent variable of vaccine hesitancy, variables with a p-value of less than 0.05 during the bivariate binary logistic regression were included in the model. The variables included in the final model were the respondent's region, educational status, prevention practices, attitudes toward the COVID-19 vaccine, vaccination status, perceived susceptibility, severity, individual benefit, perceived barriers, and self-efficacy.
Using a p-value of less than 0.05, region, practices about prevention methods, attitudes about the vaccine, vaccination status, perceived community benefit, perceived barriers regarding vaccination, and self-efficacy were significantly associated with vaccine hesitancy.
Respondents from the Sidama Regional state had odds of 53% (AOR, 95% CI: 0.47 (0.26, 0.84) of being less likely to be hesitant than residents from the Somali Regional state. The odds of study participants being from the Afar Regional state were 1.96 (AOR, 95% CI: 1.96 (1.19, 3.23)) times more likely to be hesitant in comparison with study participants from the Somali Regional State.
The odds of study participants that had inadequate preventive practices showed that they were 1.37 (AOR, 95 CI: (1.01, 1.87)) times more likely to be hesitant than study participants with adequate preventive practices. In addition, the odds of being hesitant among study participants with an unfavorable attitude toward the vaccine showed that they were 1.76 (AOR, 95%CI: 1.32, 2.37) times more likely to be hesitant to receive the COVID-19 vaccine than study participants with a favorable attitude. The odds of nonvaccinated individuals being hesitant showed that they were 1.72 (AOR, 95%CI: 1.27, 2.34) times more likely to be hesitant than vaccinated individuals. The odds of study participants having lower self-efficacy were about 8.79 (AOR, 95% CI: 5.48, 14.07) times more likely to be hesitant than those with higher self-efficacy (See Table 10). * p < 0.05, ** p < 0.01, *** p < 0.001. **** f 2 = large effect size (> 0.5).

Discussion
The main purpose of the current COVID-19 vaccine hesitancy survey was to assess the knowledge level, attitudes, and prevention and control practices regarding COVID-19, and the level of vaccine hesitancy and associated factors in Ethiopia. This study is one of the few community-based studies to assess the Ethiopian community's perceptions about COVID-19 vaccine hesitancy and the associated factors using a large sample size covering urban settings from six regions in the country.
A variety of measures have been implemented to control the pandemic by improving the community knowledge, attitudes, and practices regarding COVID-19 prevention and control [16], despite the findings of this study indicating that only 53.9% of participants had adequate knowledge about COVID-19 prevention and control activities. Similarly, the qualitative findings highlight the finding that the community was not making COVID-19 a priority issue. Studies in Ethiopia inconsistently reported the magnitude of having adequate knowledge about COVID-19 prevention and control, which ranged from 25.0% to 88.2% [21]. This difference might be related to the study population characteristics, where the current study is a community-based study and the other study included preparatory and university students who have better media access to learn about COVID-19, including school-based interventions. In addition, our study included a large sample size with various population dynamics across the country, in comparison to other studies. In contrast, the current study findings about the knowledge of the community regarding COVID-19 prevention and control is lower than studies conducted in China (82.34%) [22] among undergraduate students and with study respondents from Cameroni (84.19%) [23]. This might be attributed to contextual differences, such as information access and literacy levels between countries.
Favorable community attitudes toward COVID-19 prevention and control play a major role in designing effective community-based interventions, including COVID-19 vaccination as one prevention and control strategy. However, in this study, 55.3% of survey participants had a favorable attitude toward COVID-19 prevention and control activities. The qualitative findings support the argument that the community considered COVID-19 to be a political agenda to sell the vaccine. This implies that a significant number of individuals had negative attitudes about the prevention and control activities of COVID-19, including receiving the vaccine as a prevention and control strategy.
Different studies reported inconsistent findings on the magnitude of favorable attitudes toward COVID-19 prevention and control activities, which ranged from 56.6% to 94.8% [21]. This variability could be explained because of population differences between healthcare workers and students. Both healthcare workers and students are exposed to COVID-19 information because of the nature of their susceptibility to infection.
In the current study, only 44.5% of the study participants used adequate COVID-19 prevention and control practices, including COVID-19 vaccination. Similar to knowledge and attitudes, various studies in Ethiopia reported inconsistent findings on the magnitude of practices about COVID-19 prevention and control activities. The current study's findings about COVID-19 prevention and control activities are higher than studies conducted in other parts of Ethiopia, such as Arbaminch (23.5%) [21] and Gedeo (39.5) [24]. These discrepancies might be due to differences in community awareness as a result of mass media and social media.
In terms of the efforts regarding COVID-19 prevention and control battles, the high proportion of the community with adequate knowledge and favorable attitudes is a prerequisite to receiving the vaccine. However, in the current study, only 53.9% of study participants had adequate knowledge about the COVID-19 vaccine as one of the infection prevention approaches, and only 47.1% of the study participants had a favorable attitude toward the importance of the COVID-19 vaccine as a COVID-19 prevention and control strategy. These findings indicate a wide knowledge and attitude gap about COVID-19 vaccination, which may be attributed to a low perceived susceptibility among pastoralist study groups. This implies that efforts should be targeted at increasing knowledge levels and building positive attitudes in parallel with vaccination campaigns in the community. These findings suggest that intensive advocacy, social mobilization, and community awareness activities in collaboration with partners should be strengthened.
The distribution and intake of vaccines are shaped by challenging political, economic, social, diplomatic, and health-related matters; in this study, less than one-third of the total survey participants were vaccinated with at least the first Pfizer dose, the Johnson & Johnson, SINO, or AstraZeneca vaccine since the vaccine was first introduced in Ethiopia on 13 March 2021. Having a comorbidity, vaccine promotion campaigns, and the level of perceived risk for COVID-19 infections were shown as justifications for receiving the vaccine. This study's results are slightly higher than the national level of coverage, where more than 21.5 million people have received at least one dose and more than 20.5 million people have been fully vaccinated [2]. This implies that extra effort should be made to increase vaccine uptake, including expanding vaccination sites and ensuring the effective use of available stocks.
In the current study, which has been conducted in major towns and cities in each region, 64.4% of the study participants were hesitant about receiving the COVID-19 vaccination. A study among healthcare workers in Ethiopia indicated similar findings, which showed that 60.3% of healthcare workers were hesitant about receiving the COVID-19 vaccine [25]. Alternatively, our study showed a higher uptake of vaccines in Addis Ababa [3] in comparison to other parts of Ethiopia and abroad, where the study participants were not willing to get vaccinated. For example, areas of Arbaminch (23.5%) and Gedeo (39.5%), and other countries such as Ghana (34.9%), and Malaysia (22.7%) showed higher levels of vaccine hesitancy [3,7]. This difference might be associated with major cities, where the inhabitants have a better understanding of COVID-19 infection risks and the community has more information. Different reasons were mentioned for being hesitant, including s lack of trust regarding the vaccine (21%), doubts about the long-term side effects (18.1%), and religious grounds (13.6%). The qualitative part of the study explored the reasons for being hesitant, including misconceptions and rumors about the COVID-19 vaccine, the fear of side effects, a lack of awareness, and misinformation and misunderstandings about the vaccine. Rumors and conspiracy theories can contribute to vaccine anxiety and those rumors about vaccination campaigns being used for political purposes are not new [26]. Community conversations with the close involvement of stakeholders, including community elders, religious leaders, and the administrative structure are necessary for groups to work closely to decrease vaccine hesitancy.
An individual's desire to avoid illness related to COVID-19 depends on the belief that receiving the vaccine will prevent, or cure, the illness or reduce the severity of the disease. Ultimately, an individual's course of action to be vaccinated often depends on the person's perceptions of their susceptibility to infection, disease severity, individual and community level benefits, barriers regarding being vaccinated, and self-efficacy [4,20]. In this study, only 55.4% had perceived their susceptibility regarding the COVID-19 infection, 55.5% had perceived that the COVID-19 infection was severe, 54.2% perceived that the COVID-19 vaccine had benefits, 61.4% harbored mental barriers regarding being vaccinated, and 52.9% of the study participants had positive self-efficacy about receiving the COVID-19 vaccine.
Different factors showed a significant association with vaccine hesitancy via the regression analysis. Those who had low levels of practice regarding COVID-19 prevention and control and poor attitudes had a high probability of being hesitant, which finding had been supported by a study conducted in Addis Ababa [3]. The geographical area where each respondent lived was also one of the factors that affected vaccine hesitancy, which may have created vaccine coverage imbalances among the regional states. Those who believed that being vaccinated had an importance to the community had a lower chance of being hesitant, which reflects the herd immunity concept of immunization. To safely achieve herd immunity against COVID-19, a substantial proportion of the population would need to be vaccinated, lowering the overall amount of virus to be available to spread to the whole population. Promoting the importance of herd immunity is helpful to keep vulnerable groups who cannot be vaccinated safely (for example, due to health conditions such as allergic reactions to the vaccine) so that they remain protected from the disease [12,[27][28][29]. The presence of barriers was a significant variable that affected COVID-19 vaccine hesitancy and the qualitative study highlights this point. The findings of this study highlight a combination of factors that restrict the rapid uptake of vaccines.
Self-efficacy was an important factor in wanting to receive the vaccine and was significantly associated with vaccine hesitancy in the current study. Similarly, a study conducted among university students showed that self-efficacy significantly predicts adherence to precautionary measures regarding COVID-19 prevention and control measures, including receiving the vaccine [27][28][29].

Implications of the Study
This study is crucial to deepen our understanding of COVID-19 vaccine hesitancy. The study findings from this survey can inform policymakers and administrators about the opportunities and constraints in vaccinating the community. The findings may thus contribute to developing a strategy for controlling the pandemic by addressing those factors significantly affecting vaccination uptake. Thus, it helps vaccination program implementers and decision-makers to make wise investments to reduce the impact of COVID-19 and enhance vaccination uptake. This includes contextualizing messaging during intervention or other interventions. Moreover, this study provides policymakers and program implementers with important insights into the effective review of COVID-19 policies and the allocation of resources, both human and financial, to improve the COVID-19 situation, with a special emphasis on vaccination.

Strengths and Limitations of the Study
This study's strengths included the use of standard measurement tools after pilot testing with a large sample size and the use of rigorous analysis at various stages. A large sample size was considered and, based on population structure, the proportional allocation was attended by age (youths and adults) and gender (male and female). Additionally, the study samples are from diverse study settings, including agrarian and pastoralist communities. The limitation of this study is that it shows no causality of effects, due to the nature of the cross-sectional design. Additionally, this study is only limited to urban settings but does not address some of the factors related to urban settings, such as refugees and internally displaced people. Hence, the generalization is limited to people residing in the urban setting of Ethiopia.

Conclusions
Based on this study, the knowledge, attitudes, and practice levels regarding COVID-19 prevention and control are inadequate for reducing hesitancy about COVID-19 vaccinations and improving vaccine coverage. There is a statistically significant association between the knowledge level, attitudes, and practices regarding COVID-19 prevention and control and the gender and age of the respondents. Variations were observed among regional states in terms of knowledge, attitudes, and practices regarding COVID-19 prevention and control. A low level of knowledge was reported from the pastoralist region.
The vaccination status is very low in Ethiopian urban cities across the country. Regardless of the optimum distribution of vaccines in the urban setting, more than half of the survey participants were hesitant about receiving the COVID-19 vaccine, and different reasons and misconceptions were reported. Educational status, knowledge levels regarding prevention, prevention practices, unfavorable attitudes toward the vaccine, receiving the vaccine, perceived community benefits, perceived barriers regarding receiving the vaccine, and self-efficacy were identified as factors associated with vaccine hesitancy at a p-value of less than 0.05, after controlling for potential confounder variables. Therefore, to improve vaccine coverage and reduce the current high level of COVID-19 vaccine hesitancy, we recommend making efforts to improve the knowledge level, attitudes, and control and prevention practices regarding the importance of receiving the COVID-19 vaccine. Rumors, misunderstandings, and misconceptions were reported as a reason for vaccine hesitancy; there should be vaccine promotion activities in collaboration with local institutions, including religious institutions. Establishing participatory engagement and open debates that include minorities and marginalized communities should be a priority early in the vaccine roll-out. Furthermore, there should be clear communication protocols for communicating with the public about adverse events. Additionally, before commencing massive immunization programs, an effective communication strategy should be designed. This strategy should consider all the different languages being spoken within a country and focus on context-specific messaging to build upon personal and positive stories and promotion on social media. Moreover, this study provides policymakers and program implementers with important insights for the effective review of the COVID-19 policies, for the allocation of financial resources to improve COVID-19 awareness, with a special emphasis on vaccination. In summary, to achieve high acceptance and uptake, evidence-based and behaviorally informed strategies should be designed and used, for instance, focusing on building trust in COVID-19 vaccines before people form an opinion against them, highlighting the consequences of inaction during consultations with health professionals, and emphasizing the social benefits of vaccination. Other strategies, such as engaging the religious leaders and gatekeepers, including the health extension workers in Ethiopia, are critical.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/vaccines11040774/s1, Table S1: Questionnaire-A Study on COVID-19 Vaccine Hesitancy. Funding: This research was funded by the Packard Foundation through Amref Health Africa in Ethiopia; the grant number of the project is R223 and the APC was funded by R223.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Ethics Committee of the Federal Ministry of Health (protocol code ETCO/Admin/026/22) and Regional Offices (protocol code AAHB/12725/109) and 1 April 2022.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy reasons.