This was a cross-sectional survey nested within the serosurvey component of a study assessing the impact of COVID-19 on primary health care service provision and utilisation in Kambia (Sierra Leone), Goma (DRC) and Masaka (Uganda): the “COVID-19 HWI study.” The COVID-19 HWI study has been described elsewhere [17]. In the serosurvey component, repeated SARS-CoV-2 serology was conducted among staff with patient contact roles at participating healthcare facilities to estimate their exposure to SARS-CoV-2, the incidence of infection during the study period, and the rate of antibody waning following infection. The vaccine acceptability survey described in this paper was conducted among healthcare facility workers when they were participating in the second or third serosurvey.

Kambia district, located in the northern province of Sierra Leone, covers an area of ∼3,100 km² and has ∼345,000 residents and 69 public health facilities (1 hospital, 15 community health centres, 16 community health posts, and 37 maternal child health posts). Goma, the capital of the North Kivu province of the DRC, covers an area of ∼76 km² and has ∼707,000 residents and 39 public health facilities (13 tertiary care hospitals and 26 health centres). Masaka, located in southern Uganda, covers an area of ∼1,603 km² and has ∼336,000 residents and 36 health facilities (2 hospitals, 25 public health centres and 9 private not-for-profit health centres). Twenty-nine public health facilities in Kambia were selected for participation in the COVID-19 HWI study through stratified random sampling to ensure proportional representation of health posts and health facilities [17]. All 25 public health centres in Masaka and 21 accessible health centres in Goma were selected for participation.

Details of the COVID-19 pandemic situation and COVID-19 vaccine introductions in the 3 study countries are shown in Table 1.

Selection of participants for the COVID-19 HWI serosurvey has been described previously [17]. Healthcare facility workers aged ≥18 years who were enrolled in the COVID-19 HWI serosurvey and available at the time that the vaccine acceptability survey was conducted were eligible for inclusion. Potential participants were provided with an information sheet about the study, and written informed consent was obtained from all individual participants included in the study.

Data were collected through structured, face-to-face questionnaire-based interviews conducted by trained study staff members. The questionnaire, which was developed by the study team specifically for this survey, was pre-programmed into computer tablets. Participants were asked about their knowledge and perceptions of, and attitudes towards, the COVID-19 pandemic and COVID-19 vaccines, their COVID-19 vaccination status and, if unvaccinated, whether they would accept COVID-19 vaccination if it were offered to them. The interviewer read each question to the participant and pre-coded answers were recorded in the electronic case report form.

Data were analysed using STATA version 16.0. Data were summarised descriptively and tabulated, stratified by country. Primary analyses were conducted among all healthcare facility roles combined. Secondary analyses were stratified by role, considering level and proximity of patient contact as follows [1]: doctors, clinical officers, nurses and midwives—highest contact [2]; clinical support staff (such as health attendants, antenatal care workers and community health workers)—medium contact; and [3] pharmacy and laboratory staff and non-clinical support staff (such as ambulance drivers, porters and receptionists)—lowest contact.

Multivariable logistic regression was used to examine factors associated with COVID-19 vaccine uptake or intent to get vaccinated. Regression analyses first evaluated associations with age, sex, education level and role in the healthcare facility. Second, adjusted for these variables, analyses examined whether vaccine uptake or intent to get vaccinated was associated with specific perceptions on the impact of the COVID-19 pandemic and the COVID-19 response in the country, as well as understanding and views of COVID-19 vaccines.

The study was reviewed and approved by the London School of Hygiene and Tropical Medicine Research Ethics Committee (Ref: 22726), the Uganda Virus Research Institute Research Ethics Committee (Ref: GC/127/821), the Uganda National Council for Science and Technology (Ref: H1430ES), the Comité National d’Ethique de la Santé, and the Sierra Leone Ethics and Scientific Review committee. All study procedures were performed in accordance with the ethical standards of the institutional and national research committees, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

All participants in Kambia, 185 (98.4%) from Goma and 225 (97.4%) from Masaka agreed that COVID-19 was an important public health problem in their country (Table 3). Results were similar when stratifying by role in the healthcare facility (Supplementary Tables S1–S3). Most participants perceived that many healthcare workers were getting sick (337/543; 62.1%), and that many people generally were becoming unwell (n = 455/543; 83.8%) and dying (n = 365/545; 67.2%), from COVID-19 in their country. However, 2 registered nurses from Goma and a training nurse from Masaka believed that COVID-19 was not a real disease (n = 2), or that there was no SARS-CoV-2 or COVID-19 in their country (n = 2).

Views on the impact of the COVID-19 pandemic varied considerably across the 3 study sites (Table 3). For example, 184 (79.7%) participants in Masaka believed that healthcare services in their country were suffering because of the pandemic, compared with just 39 (20.7%) from Goma. Most (n = 119; 96.0%) participants from Kambia considered that the COVID-19 response was detrimental to their country’s economy, compared to just 79 (42.0%) from Goma.

Most participants (89.7% overall; 86.4% of clinicians, nurses and midwives) considered that COVID-19 vaccines were important to control the pandemic; this ranged from 75.0% of participants in Goma to 100.0% in Kambia (Table 3; Supplementary Tables S1–S3). Among the participants who did not agree that COVID-19 vaccines were important for control of the pandemic, the most commonly cited reasons were that there were other priorities that were more important for the country (n = 15), that COVID-19 could be controlled with other measures (n = 15) and that COVID-19 vaccines might not work (n = 15). Overall, 58 (46.8%) participants from Kambia, 21 (11.2%) from Goma and 154 (66.7%) from Masaka believed that COVID-19 vaccination should be compulsory in their country.

Whilst the results above indicate that overall support for COVID-19 vaccines was high (albeit lower in Goma compared to the other study sites), 393 (72.4%) participants (ranging from 49.4% in Masaka to 91.1% in Kambia) had concerns that the vaccines were new and experimental (Table 3). Findings were similar when restricted to clinicians, nurses and midwives (Supplementary Tables S1–S3). Overall, 250 (46.0%) participants had concerns that COVID-19 vaccines could cause SARS-CoV-2 infection or COVID-19, 167 (30.8%) had concerns that they could cause infertility and 283 (52.1%) had concerns that they could be harmful in pregnancy; all of these concerns were more common in Kambia than in Goma or Masaka (Table 3). Findings were similar when restricted to clinicians, nurses and midwives (Supplementary Tables S1–S3). Only 174 (54.9%) participants from this restricted group considered that COVID-19 vaccines could cause anaphylaxis, despite this being a known risk associated with vaccines.

At the time of the survey, 48 (38.7%) participants from Kambia and 89 (38.5%) from Masaka had received ≥1 COVID-19 vaccine dose (Table 4; Figure 1). In Kambia, vaccine uptake (≥1 dose) was highest among clinicians, nurses and midwives (n = 17/29; 58.6%), followed by clinical support workers (n = 25/71; 35.2%), followed by other workers (n = 6/24; 25.0%) (37.9%) (Chi² = 7.1, p = 0.029; Supplementary Table S1). There was no difference in uptake between different roles in Masaka (Supplementary Table S3). Most unvaccinated participants from Kambia (n = 75/76; 98.7%) and Masaka (n = 136/142; 95.8%) said that they would accept a COVID-19 vaccine if offered it, meaning that reported willingness to accept the vaccine—including either uptake of ≥1 dose by the time of the survey or intent to get vaccinated—was very high (99.2% in Kambia; 97.4% in Masaka). In Goma, no study participant was vaccinated against COVID-19 at the time of the survey (despite the vaccine being available to healthcare facility workers at that time) and only 81 (43.1%) participants (and 43.6% of clinicians, nurses and midwives) said that they would accept the vaccine if offered it.

Reasons for accepting or not accepting the COVID-19 vaccine (with “acceptance” referring to either vaccine uptake or intent to get vaccinated) are shown in Table 4. Among the 96 participants from the 3 study sites who reported that they would not accept the vaccine, the most common reasons were fear of side effects (n = 80; 83.3%) and concerns that the vaccines were new and experimental (n = 81; 84.4%). Approximately one fifth of participants who would not accept the vaccine believed that they were not at risk of catching SARS-CoV-2 (n = 21; 21.9%) or becoming seriously ill or dying from COVID-19 (n = 19; 19.8%), and almost a third believed that the vaccines did not work (n = 32; 33.3%).

When asked, hypothetically, what factors might influence participants’ decisions on whether or not to receive a COVID-19 vaccine, the most commonly selected factors included which country the vaccine was developed in (n = 334; 61.5%) and where and for how long the clinical vaccine trials to evaluate it were conducted (n = 341; 62.8% and n = 355; 65.4%, respectively), although results varied slightly across the 3 study sites (Table 4). Among the 114 unvaccinated participants who said that they would not accept a COVID-19 vaccine or were unsure if they would accept one, most said that, hypothetically, their decision on whether to get vaccinated might be influenced by where and for how long the clinical vaccine trials were conducted (n = 78; 68.4% and n = 74; 64.9%, respectively), and how many people had been vaccinated before them (n = 69; 60.5%).

In Kambia, theoretical willingness to be vaccinated against COVID-19 was high for vaccines developed in the United States (n = 119; 96.0%), the United Kingdom (n = 122; 98.4%), Europe (n = 112; 90.3%) or Russia (n = 113; 91.1%), but was lower for vaccines developed in China (n = 80; 64.5%) (Supplementary Figure S1). A similar pattern was observed in Goma (the United States: n = 95; 50.5%, United Kingdom: n = 92; 48.9%, Europe: n = 86; 45.7%, China: n = 57; 30.3%, Russia: n = 69; 36.7%). In Masaka, willingness to be vaccinated against COVID-19 was highest for vaccines developed in the United States (n = 192; 83.1%) and the United Kingdom (n = 185; 80.1%), and lower for those developed in Europe (n = 114; 49.4%), China (n = 97; 42.0%) or Russia (n = 100; 43.3%).

Since almost all participants in Kambia and Masaka had either received ≥1 COVID-19 vaccine dose or reported their intention to receive the vaccine, analyses of factors associated with willingness to receive a COVID-19 vaccine were limited to participants from Goma (Table 5). Adjusted for age, sex and type of role within the healthcare facility, participants with lower education level were more likely to report intent to receive the COVID-19 vaccine than those with higher education level (OR 3.17, 95%CI 1.01–9.96; p = 0.040) (Table 5). Adjusted for age, sex, education and type of role within the healthcare facility, intention to get vaccinated was associated with participant perceptions that the COVID-19 pandemic, or the country’s response to the pandemic, was having a negative or detrimental affect on healthcare services (OR 5.69, 95%CI 2.45–13.21; p < 0.001), that the COVID-19 response was resulting in the neglect of other diseases (OR 8.61, 95%CI 3.24–22.89; p < 0.001), and that the COVID-19 response was detrimental to the economy (OR 2.24, 95%CI 1.21–4.17; p = 0.010). Intent to be vaccinated was also associated with participants’ perceptions/knowledge that COVID-19 vaccines protect against SARS-CoV-2 infection or can stop people from getting very ill from COVID-19 (for each, OR 3.69, 95%CI 1.73–7.87; p < 0.001), and that the vaccines are important for control of the pandemic (OR 18.19, 95%CI 5.23–63.30; p < 0.001).

Whilst many participants from Goma had concerns that COVID-19 vaccines were new and experimental (n = 166; 88.3%), that they might not work (n = 42; 22.3%), and that they may cause side effects or harm (n = 129; 68.6%), these concerns were not associated with lower intention to get vaccinated in multivariable analyses (Table 5). Thus, whilst these concerns may have played an integral role in decision making among participants who did not intend to get vaccinated, the concerns were similarly prevalent among participants who reported intent to get vaccinated (i.e., with the latter group intending to get vaccinated despite these concerns).

Additional exploratory analyses evaluated factors associated with uptake of ≥1 vaccine dose in Masaka, with the hypothesis that unvaccinated participants may be those who had delayed getting vaccinated (as the survey was conducted 6–7 months after vaccine introduction); none of the variables evaluated were associated with uptake (Supplementary Table S4).

In 2019, the WHO declared vaccine hesitancy as one of the ten greatest threats to global health [36]. With the recent development, trialling and roll-out of COVID-19 vaccines, topics surrounding vaccine decision-making, misinformation and ‘anti-vax’ movements have gained even greater prominence in both the scientific literature and the media. This survey provides important results on COVID-19 vaccine acceptability among healthcare facility workers in the months following vaccine introduction in study sites from 3 SSA countries, which may be informative for future vaccine introductions in the region, particularly within the context of an outbreak or pandemic. The results demonstrate the importance of educating healthcare staff on vaccine development processes, and on evidence surrounding vaccine efficacy/effectiveness and safety, both to promote vaccine uptake in this group and to ensure appropriate and accurate communication regarding vaccines between healthcare staff and their patients or the wider communities. Results also demonstrate the importance of careful and accurate messaging surrounding vaccines and vaccine introductions at a global level.