COVID-19 Vaccine Acceptance and Hesitancy in Health Care Workers in Somalia: Findings from a Fragile Country with No Previous Experience of Mass Adult Immunization

Coverage of COVID-19 vaccines in Somalia remains low, including among health workers. This study aimed to identify factors associated with COVID-19 vaccine hesitancy among health workers. In this cross-sectional, questionnaire-based study, 1476 health workers in government and private health facilities in Somalia’s federal member states were interviewed face-to-face about their perceptions of and attitudes toward COVID-19 vaccines. Both vaccinated and unvaccinated health workers were included. Factors associated with vaccine hesitancy were evaluated in a multivariable logistic regression analysis. Participants were evenly distributed by sex, and their mean age was 34 (standard deviation 11.8) years. The overall prevalence of vaccine hesitancy was 38.2%. Of the 564 unvaccinated participants, 39.0% remained hesitant. The factors associated with vaccine hesitancy were: being a primary health care worker (adjusted odds ratio (aOR) = 2.37, 95% confidence interval (CI): 1.15–4.90) or a nurse (aOR = 2.12, 95% CI: 1.05–4.25); having a master’s degree (aOR = 5.32, 95% CI: 1.28–22.23); living in Hirshabelle State (aOR = 3.23, 95% CI: 1.68–6.20); not having had COVID-19 (aOR = 1.96, 95% CI: 1.15–3.32); and having received no training on COVID-19 (aOR = 1.54, 95% CI: 1.02–2.32). Despite the availability of COVID-19 vaccines in Somalia, a large proportion of unvaccinated health workers remain hesitant about being vaccinated, potentially influencing the public’s willingness to take the vaccine. This study provides vital information to inform future vaccination strategies to achieve optimal coverage.


Introduction
The COVID-19 pandemic caused severe and widespread disruption to essential health services in Somalia, even two years after the first case of COVID-19 was detected in the country on 16 March 2020 [1]. The reasons for this disruption include the lockdown and closure of health facilities because of absenteeism of health care workers who feared getting COVID-19. As a result, the country's routine immunization service has been interrupted, resulting in thousands of children missing routine immunization. This situation is alarming as the routine immunization coverage in Somalia was already low [2]. The country also has one of the smallest number of health workers in the world-fewer than 1 per 1000 population [3]. The number of health care workers in Somalia who contracted COVID-19 remains unknown. However, earlier estimates suggest that during the first 180 days of the outbreak, health workers had a 61% positivity rate and accounted for at least 5% of cases during the epidemic's peak in the country [3]. This is comparable to the finding of an earlier scoping review suggesting that 3.9% of COVID-19 cases worldwide might have been health workers during the early days of the COVID-19 pandemic [4].

Data Collection
Development and pilot testing of a structured and closed-ended questionnaire started in June 2021, and ethical clearance for the study was obtained. The questionnaire was adapted from the behavioral and social drivers of vaccination model of the World Health Organization (WHO) and was available in English and Somali [13][14][15]. It was tested on a group of health workers to determine its clarity and applicability. Issues identified during this process were addressed in the final version of the questionnaire used for data collection.
Data were collected from 20 February to 21 March 2022 by 30 trained data collectors using the questionnaire. The data collectors visited all six states and interviewed each of the health workers face-to-face after receiving their informed consent to participate in the study. The interviews were conducted in Somali. No interview was done over the telephone. The questionnaire had five sections: (i) participants' sociodemographic characteristics (such as, age, sex, profession, marital status, education, and work setting); (ii) history of COVID-19 and other health-related information such as underlying chronic health conditions; (iii) perceived concerns about the COVID-19 vaccine, including views

Data Collection
Development and pilot testing of a structured and closed-ended questionnaire started in June 2021, and ethical clearance for the study was obtained. The questionnaire was adapted from the behavioral and social drivers of vaccination model of the World Health Organization (WHO) and was available in English and Somali [13][14][15]. It was tested on a group of health workers to determine its clarity and applicability. Issues identified during this process were addressed in the final version of the questionnaire used for data collection.
Data were collected from 20 February to 21 March 2022 by 30 trained data collectors using the questionnaire. The data collectors visited all six states and interviewed each of the health workers face-to-face after receiving their informed consent to participate in the study. The interviews were conducted in Somali. No interview was done over the telephone. The questionnaire had five sections: (i) participants' sociodemographic characteristics (such as, age, sex, profession, marital status, education, and work setting); (ii) history of COVID-19 and other health-related information such as underlying chronic health conditions; (iii) perceived concerns about the COVID-19 vaccine, including views on its safety and effectiveness; (iv) sources of information on COVID-19 vaccines that the health worker thought credible and trusted; and (v) intention to get a COVID-19 vaccine for unvaccinated health workers. For part (v), the health workers were asked, "If not vaccinated, do you intend to get vaccinated?" and the responses were "Yes", "No", and "I am currently undecided". Date were collected using the KoboToolbox mobile data collection platform.
We also included a question on perceived negative treatment because of being a health worker. In some sub-Saharan countries it has been reported anecdotally that health workers were stigmatized and seen as the source of COVID-19 virus transmission because of the nature of their work and their exposure to COVID-19 patients. We aimed to understand if health workers in our study thought that they were stigmatized. If this were the case, this might influence health workers' decision to accept the COVID-19 vaccine.
For the purpose of this analysis, health workers were considered to be vaccinated if they had received two doses of a COVID-19 vaccine or had received a single dose and were waiting for the second dose. A third (booster) dose was not mandatory in Somalia at the time as many high-risk populations were yet to complete their primary vaccines. Respondents were defined as COVID-19 vaccine hesitant if they had refused or delayed accepting the COVID-19 vaccine despite its availability at designated centers or health facilities where they worked; hence, they were unvaccinated at the time of the survey [12].

Data Analysis
Data review and cleaning were done with Microsoft Excel, input errors were removed, and statistical analysis was conducted using SPSS, version 27 (IBM Corp., Armonk, NY, USA).
Sociodemographic data were summarized as categorical variables, disaggregated by vaccination status. Descriptive statistics (mean and standard deviation (SD), frequency, and percentage) are presented. Sociodemographic characteristics of the health workers who were vaccinated were compared with the characteristics of those who were not vaccinated using the chi-squared test.
We conducted a bivariate analysis with hesitancy as the dependent variable and sociodemographic characteristics as the independent variables. We combined health workers who were undecided and who did not plan to get vaccinated as a single category called "vaccine hesitant" [12]. We compared the characteristics of health workers in the vaccinehesitant group with those who said that they were willing to get vaccinated at a later time. The results were presented as odds ratios (OR) and 95% confidence intervals (CI). A p-value of <0.05 was considered statistically significant.
A multivariable logistic regression analysis was also conducted to assess predictors of vaccine hesitancy in the health workers. The health workers who were not vaccinated at the time of this study, irrespective of their intention to do so at a later stage, were classified as vaccine hesitant (which is consistent with the WHO definition of vaccine hesitancy). We included all clinically important variables in our model. Results are presented as adjusted OR (aOR) with 95% CIs. A p-value of <0.05 was considered statistically significant.

Ethical Considerations
The study was approved by the Federal Ministry of Health & Human Services of the Government of Somalia. Data collection was anonymous, and participants' identities were kept confidential. The participants were informed of the study's objectives and assured of the anonymity and confidentiality of their data. Written informed consent was collected from the participants before starting the questionnaire. Table 1 shows the sociodemographic characteristics of the 1476 health workers according to their vaccination status. The mean (SD) age of the participants was 34 (11.8) years. The median age was 30 years (interquartile range (IQR) 25-38). A significant difference was seen between the median age of those who were vaccinated and those who were not vaccinated (32 (IQR: 26-

Vaccination Status by Sociodemographic Characteristics
At the time of the study, 61.8% (912/1476) of the health workers were vaccinated (had either had two doses of a COVID-19 vaccine or had a single dose and were waiting for the second dose), while 38.2% (564/1476) were unvaccinated. Significant differences were found between the vaccinated and unvaccinated health workers for all the variables examined except for sex (Table 1).

Exposure to COVID-19
As shown in Table 2, 54.7% (807/1476) of health workers had had COVID-19. Of these respondents, 82.8% (668/807) had been confirmed positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in a laboratory test. The remaining 17.2% were not specifically asked why they thought they had had COVID-19 as this was not a focus of our study. Most health workers (60.1%; 887/1476) said that they had cared for or treated a patient with COVID-19, and 60.8% (898/1476) said that they had been in contact with a friend or family member who was positive for COVID-19.

Reasons for COVID-19 Vaccine Hesitancy
The two most important reasons cited by the unvaccinated health workers for their unwillingness to take the COVID-19 vaccine were: (i) need for more information about the vaccine's safety and effectiveness (39.4%; 222/564) and (ii) the perception that the COVID-19 vaccine has side effects (31.7%; 179/564) ( Table 5).

Trusted Sources of Information
Unvaccinated health workers who expressed willingness to get vaccinated reported that mobile telephone caller tune messages (heard by a caller after making an outgoing call to another person), international organizations, and the government COVID-19 hotline were the most trustworthy sources of information about COVID-19 and COVID-19 vaccines that could influence their decision to take the vaccine (Figure 2). Somalia, like many other African and Asian countries, replaced the standard ringtone with COVID-19 awareness and prevention messages when the pandemic started [16]. These messages are in Somali, and the Ministry of Health has approved the content.

Predicators of COVID-19 Vaccine Hesitancy
In the multivariable logistic regression analysis to determine predictors of vaccine hesitancy in health workers, the following factors were significantly associated with hesitancy: being a primary health care worker (adjusted OR (aOR) = 2.37, 95% CI: 1.15-4.90)

Predicators of COVID-19 Vaccine Hesitancy
In the multivariable logistic regression analysis to determine predictors of vaccine hesitancy in health workers, the following factors were significantly associated with hesitancy: being a primary health care worker (adjusted OR (aOR) = 2.37, 95% CI: 1. 15-4.90) Table 6). No association was found between vaccine hesitancy and perceived negative treatment as a health worker in the regression analysis.  Vaccine hesitancy in health workers was not significantly associated with sex, age, marital status, working in public or private health facilities, having chronic health conditions, and treating or caring for COVID-19 patients (Table 6).

Discussion
This is the first study in Somalia to report on COVID-19 vaccine acceptance and hesitancy among health care workers in the country and explore the factors associated with their hesitancy. Before the COVID-19 pandemic, Somalia's immunization program offered only two vaccines for adults-yellow fever for travelers and tetanus toxoid for pregnant women.
Despite health workers being a priority group for vaccination against COVID-19 in the country during the initial phase of vaccine deployment, the acceptance rate, after 12 months of vaccine roll-out, was 61.8% among the health workers. Our study found health workers in Hirshabelle State were significantly more likely to be unvaccinated. This state is one of the newly established states (in 2017), and many parts are inaccessible because they are controlled by insurgent forces. As a result, access to health care is severely disrupted and primary health care services are almost non-existent. Thus, security challenges and other health system problems could be the reason for the poor COVID-19 vaccine coverage in this state.
In the multivariable logistic regression analysis to determine predictors of vaccine hesitancy in health workers, having a master's degree compared with a secondary school certificate was significantly associated with vaccine hesitancy. It could be that these health workers were delaying vaccination until more data on vaccine safety were available.
Among the vaccine-hesitant health workers, 33.3% were willing to accept the vaccine, and 39.0% were not willing or were undecided. Our study was conducted during the initial phase of vaccine deployment in Somalia, while most published studies available on vaccine hesitancy and acceptance in health care workers around the world predate the deployment of COVID-19 vaccines. For example, in Egypt, a study on COVID-19 vaccination perception and attitudes before the vaccination programme reported that 51% of health workers were undecided about taking the vaccine, 28% were hesitant, and 21% said they would take it [17]. In a similar study in Nigeria before the COVID-19 vaccine was rolled out, the hesitancy rate was 50.5% (95% CI: 45.6-55.3%) [18]. Similarly in Ethiopia, before the COVID-19 vaccine was rolled out, 60.3% of health workers surveyed were vaccine hesitant [19]. However, another study in Nigeria during the initial phase of vaccine deployment found that only 8% health workers interviewed were vaccine hesitant [20]. Our finding of a 61.8% vaccine acceptance rate mirrors the findings of a systematic review of COVID-19 vaccine acceptance in health workers in Africa, which showed that the mean (SD) acceptance rate increased marginally from 55.5% (5.6) in 2021 to 60.8% (5.3) in 2022, resulting in an overall mean COVID-19 vaccine acceptance rate of 58% in Africa in 2022 [21]. The prevalence of vaccine hesitancy in health workers worldwide ranged from 4.3% to 72.0% [11].
This information on COVID-19 vaccine hesitancy and acceptance is important for Somalia. In the absence of any such studies in the past in the general population, our findings will be valuable for understanding and measuring changes over time in the attitude and perception of the general population and health workers regarding COVID-19 vaccines or during the roll-out of other adult vaccines in mass immunization programs. Studies have shown a strong association between vaccination of health workers for COVID-19 and high population coverage. This suggests that achieving high vaccination coverage for COVID-19 and other adult vaccines in health workers by addressing hesitancy could lead to optimal vaccination coverage of the same vaccines in the general population [22].
The two most important reasons cited by the unvaccinated health workers for their hesitancy were the need for more information on the vaccine's safety and efficacy, and the potential side effects of the COVID-19 vaccine. This is similar to the findings of other studies [21]. Other studies on vaccine hesitancy in health workers found concerns about vaccine safety, efficacy, and side effects were the main reasons for COVID-19 vaccination hesitancy [8].
The vaccinated and unvaccinated health workers in our study differed significantly in their perceptions of the COVID-19 vaccines' safety and effectiveness. Significantly more vaccinated health workers thought that their getting vaccinated helped protect others and would recommend the COVID-19 vaccine to eligible patients. This is consistent with many studies where the authors have suggested that accessing reliable information, rather than education, was a better determinant of vaccine acceptance [23]. This finding has an important policy implication. Wider dissemination of information using the sources that are most trusted by the health workers in the country would help improve vaccination coverage in health workers and eventually in the general population.
In our study, just over two thirds of health workers reported that mobile telephone ring tone messages were the most trustworthy source of information about COVID-19 and COVID-19 vaccines. Like many other African and Asian countries, Somalia replaced the standard ringtone with COVID-19 awareness and prevention messages. In addition, social media was considered a trusted and credible source of information that could influence health workers' decision to take the COVID-19 vaccine. This is similar to findings in Ethiopia [19], where health workers frequently viewed social media as the best source of COVID-19 information. Therefore, these sources can be used to amplify messages and develop a risk communication plan that is tailored to address myths and misinformation about COVID-19 vaccine hesitancy and hence contribute to improved coverage.
The factors significantly associated with vaccine hesitancy in our health workers were: being a primary health care worker or a nurse; not having had COVID-19; and not having received any training on COVID-19. This is similar to the findings of other studies [24]. However, unlike another study [25], we did not find any positive association between chronic health conditions and vaccine acceptance. This could be due to the fact that the mean age of our study participants was 34 years, and most reported having no chronic health conditions. Our finding that primary-level health workers and nurses were more likely to be vaccine hesitant needs special attention. Most health workers administering the COVID-19 vaccines as well as managing the essential health services in Somalia belong to these groups. They are also the frontline health workers in the country and are the backbone of COVID-19 vaccine deployment in the country. Therefore, addressing their concerns and hesitancy will serve to advance the vaccination program and improve COVID-19 vaccination coverage in the country, as has been found in Rwanda [26].
Understanding the dynamics of vaccine acceptance and hesitancy has always been important, especially in fragile countries with weak health systems, which are vulnerable to epidemics caused by high-threat pathogens and pandemics such as COVID-19. Vaccines play a key role in ending such epidemics or pandemics. Vaccination of health workers has been shown to protect their families, friends, and patients and reduce infection and transmission in health care settings [27]. However, vaccine hesitancy has been shown to exist among the general population and health workers [28], and it is important to explore interventions to overcome this hesitancy [29]. In our study, being a primary health care worker was associated with vaccine hesitancy. This is a concern as in the national health system of Somalia, primary health care workers are the first contact for a patient. This means that a patient has to go first to a primary health care worker and then to a hospital. Hence, primary health care workers are likely to be in contact with many patients and so are susceptible to infection from COVID-19 patients attending primary health care centers. It is therefore important that they are vaccinated so they do not get sick themselves and do not pass on the infection to primary health care attendees. In addition, they can reach many people to advocate for vaccination. Thus, it is particularly important to tackle hesitancy among primary health care workers. It is therefore vital to ensure that the concerns of Somali health workers about vaccines are addressed and that they are vaccinated.

Limitations and Strengths
Because of the design of the survey, we could not establish causality nor observe trends over time. In addition, our findings may be susceptible to non-response bias, as health workers who agreed to participate may have disproportionately different characteristics compared with non-respondents. We did not collect data from Somaliland, which represents 25% of the Somali population. Despite these limitations, this study has some strengths. This is the first study in Somalia to determine predictors of vaccine hesitancy in health care workers. Unlike many other studies on vaccine hesitancy that used telephone interviews or web-based surveys, we collected data in face-to-face interviews conducted by trained interviewers. Hence, there were opportunities to cross-check responses during the interview, limiting the biases. In addition, this study was conducted at a time when the COVID-19 vaccines were being rolled out in the country, so the findings on acceptance and hesitancy are from real-life experience.

Conclusions
The findings of this study present a unique opportunity to the policy planners and public health officials in the country to address the factors contributing to or associated with vaccine hesitancy and improve COVID-19 vaccine uptake among both health workers and the general public. The findings can be used to help improve vaccination coverage in health workers for any adult vaccine in the future, such as seasonal influenza and other pandemic vaccines, and hence help enhance Somalia's epidemic and pandemic readiness.