Assessing hepatitis B vaccination status and related factors among healthcare workers in the Central Health District of Bujumbura, Burundi: a cross-sectional study

¹Department of Environmental Health, Faculty of Health Sciences, University of Johannesburg, Johannesburg, South Africa, ²Department of Optometry, Faculty of Health Sciences, University of Johannesburg, Johannesburg, South Africa

^&Corresponding author
Thokozile Ingrid Metsing, Department of Optometry, Faculty of Health Sciences, University of Johannesburg, Johannesburg, South Africa

Introduction    

Hepatitis B virus (HBV) infection remains a major global public health challenge despite the availability of safe and effective vaccines [1-3]. According to the World Health Organization (WHO), an estimated 254 million people were living with chronic HBV infection worldwide in 2022, with more than one million deaths occurring annually from HBV-related complications, including cirrhosis and hepatocellular carcinoma [4,5]. The burden of disease is particularly high in sub-Saharan Africa, where HBV remains highly endemic and prevalence frequently exceeds 8% in the general population [4,5]. In Burundi, the reported prevalence was 8.43%, putting the country at the higher end [6]. Healthcare workers (HCWs) are among the occupational groups at greatest risk of HBV infection because of their frequent exposure to blood and body fluids during patient care [3,7]. Occupational exposures such as needlestick injuries, sharps injuries, and contact with contaminated biological materials substantially increase the risk of infection [8,9]. The risk of HBV transmission following percutaneous exposure to infected blood may be as high as 30%, considerably higher than the corresponding risks for human immunodeficiency virus (HIV) and hepatitis C virus (HCV) [8,9]. Vaccination remains the most effective strategy for preventing HBV infection among healthcare workers [4,10,11]. The WHO recommends HBV vaccination for all healthcare workers, and completion of the recommended three-dose schedule provides protective immunity in more than 90% of vaccinated individuals [4,10,11].

Nevertheless, vaccination coverage among healthcare workers remains low in many African countries [12-14], with reported full vaccination coverage ranging from 9% in Uganda [15], 15.4% in Malawi [12], 16% in Somalia [16], 18% in Tanzania [17], 36.9% in Ethiopia [18], and 42.6% in Nigeria [19], while coverage remains below 10% in the Democratic Republic of the Congo [20]. These findings highlight persistent gaps in occupational protection across sub-Saharan Africa [10-12]. In Burundi, HBV vaccination was introduced into the national Expanded Program on Immunization (EPI) in 2004 [21]. However, unlike childhood immunization, there is currently no national occupational health policy that ensures systematic HBV vaccination or routine pre-employment vaccination for healthcare workers [22,23]. Although HBV remains endemic in Burundi, with an estimated prevalence of 8.43% [6], little is known about HBV vaccination coverage among healthcare workers. To our knowledge, no published study has specifically assessed HBV vaccination status and associated factors among healthcare workers in Burundi [12, 21-26]. Generating such evidence is essential for informing occupational health policies and strategies aimed at improving HBV prevention among healthcare workers. This study aimed to determine HBV vaccination coverage and identify factors associated with HBV vaccination uptake among healthcare workers in the Central Health District of Bujumbura, Burundi.

Methods     

Study design and setting: this analytical cross-sectional study was conducted among healthcare workers between November 2025 and January 2026. The study took place in the Central Health District of Bujumbura, Burundi. The district comprises a mix of public and private hospitals and health centers, and was purposively selected because it hosts a high concentration of healthcare professionals and a diverse range of health facilities.

Study population: the study population consisted of healthcare workers involved in direct patient care, including doctors, nurses, laboratory technicians, hygienists, and other clinical personnel working in health facilities within the Central Health District of Bujumbura. Eligible participants were healthcare workers born before 2004 who were actively working in selected facilities during the study period and provided written informed consent. Within selected facilities, all eligible healthcare workers present during the data collection visits and willing to participate were included. Administrative staff and HCWs born in 2004 and after were excluded. In addition, HCWs who were absent, on leave, or declined participation were not enrolled; reasons and numbers of non-participants were not recorded. A list of 51 health facilities was obtained from the district authorities, and 22 facilities were selected using simple random sampling because financial and logistical constraints prevented inclusion of all facilities. The participant selection process is summarized in Figure 1. The minimum sample size was estimated using Epi Info, based on a 95% confidence level, a 5% margin of error, and an assumed vaccination prevalence of 50%, resulting in a required sample size of 341 participants [27-29]. To account for potential non-response, the target sample size was increased to 375. Within the 22 randomly selected health facilities, all eligible healthcare workers present during the data collection visits and willing to participate were recruited using convenience sampling.

Data collection: a structured questionnaire in French was administered electronically via Google Forms to collect data. The questionnaire featured closed-ended, multiple-response, true/false/don't know, and likert-scale items. It addressed socio-demographic characteristics, HBV vaccination status, hepatitis B knowledge, attitudes toward vaccination, and reasons for non-vaccination. A one-day training session was conducted for the six data collectors prior to data collection. During this session, data collectors were instructed on the study objectives, ethical considerations, informed consent procedures, inclusion criteria, and administration of the structured questionnaire using Google Forms. The training ensured a shared understanding of all questionnaire items and standardized data collection procedures. Prior to data collection, the questionnaire was pilot tested with six healthcare workers from health districts not included in the study to assess its clarity, relevance, and completion time. Minor revisions were made based on participant feedback, including simplification of selected questions and reduction of the Likert scale from five to three response categories. The questionnaire was later administered on smartphones. Responses were automatically stored in Google Sheets, then exported to Microsoft Excel and IBM SPSS version 29 for cleaning and analysis. Throughout the data collection period, the completeness and consistency of the submitted questionnaires were reviewed daily using Google Sheets.Data collection took place in 22 randomly selected health facilities within the Central Health District. The six data collectors visited the selected health facilities and invited eligible HCWs to participate. To maximize participation, each facility was visited two to three times over approximately 10 days, covering both morning and afternoon shifts. Before completing the questionnaire, participants were provided with an explanation of the study and subsequently asked to give informed consent. Due to concerns about confidentiality, particularly amid longstanding political tensions, some participants were allowed to provide only their initials on the consent form. No names or identifying information were collected in the questionnaire. Potential sources of bias were addressed through random selection of health facilities, repeated visits to maximize participation, use of a standardized pretested questionnaire, training of data collectors, and daily quality checks of electronic data. Residual selection and recall bias are acknowledged.

Definitions: the primary outcome variable was HBV vaccination status, categorized as vaccinated (having received at least one dose of HBV vaccine) or unvaccinated (having never received any dose). Participants who reported receiving three or more doses of HBV vaccine were classified as fully vaccinated. Independent variables were defined as follows: age was recorded in completed years and categorized into predefined age groups; sex was classified as male or female; professional category referred to the participant's role in the health system (doctor, nurse, laboratory technician, hygienist); religion was self-reported and categorized according to major religious affiliations; years of professional experience were calculated from the date of entry into professional practice and grouped into categories; type of health facility was classified as public or private hospitals and health centers; facility location referred to the geographical area within the district where the facility was situated; knowledge of hepatitis B was assessed using seven questions covering HBV transmission, prevention, occupational exposure, and vaccination. Each correct response was worth 1 point, whereas incorrect or "don't knowʺ responses received 0 points. Individual item scores were summed to obtain a total knowledge score ranging from 0 to 7. Consistent with cut-off values used in previous studies, participants were classified as having poor knowledge (0-3 points), moderate knowledge (4-5 points), or good knowledge (6-7 points). The internal consistency of the knowledge scale was acceptable (Cronbach's ɑ= 0.702). Attitude toward HBV vaccination was assessed using five statements measured on a three-point Likert scale (agree, neutral, disagree). Negatively worded items were reverse-coded so that higher scores reflected more positive attitudes toward vaccination. The individual item scores were summed to generate a total attitude score ranging from 5 to 15. Based on cut-off values adapted from previous studies, participants were classified as having a negative (5-8), neutral (9-11), or positive (12-15) attitude toward HBV vaccination. The attitude scale demonstrated acceptable internal consistency (Cronbach's ɑ= 0.742).

Statistical analysis: data were exported from Google Sheets to Microsoft Excel and analyzed using IBM SPSS Statistics version 29. The dataset was examined for completeness before analysis. No missing data were identified; therefore, all 379 participants were included in the analyses. Descriptive statistics were used to summarize participants' socio-demographic and professional characteristics, HBV vaccination status, and knowledge and attitude levels. Continuous variables were summarized using means and standard deviations (SD) or medians and ranges, as appropriate, while categorical variables were presented as frequencies and percentages. Pearson's chi-square tests were initially performed to examine associations between HBV vaccination status and categorical explanatory variables. Univariable logistic regression analyses were then conducted to estimate crude odds ratios (CORs) and 95% confidence intervals (CIs) for factors associated with HBV vaccination. Variables with a p-value <0.20 in the univariable analysis, along with variables considered epidemiologically relevant based on prior literature and the study's conceptual framework, were simultaneously entered into a multivariable logistic regression model to identify factors independently associated with HBV vaccination. Age was evaluated in the univariable analysis but was not retained in the final model because of its strong correlation with years of professional experience, thereby avoiding multicollinearity. Adjusted odds ratios (aORs) with corresponding 95% confidence intervals were reported. Model performance was assessed using the Omnibus test of model coefficients, the Hosmer-Lemeshow goodness-of-fit test, Nagelkerke's R², classification accuracy, and the area under the receiver operating characteristic (ROC) curve. Multicollinearity among independent variables was evaluated using variance inflation factors (VIFs), with values below 5 indicating no important multicollinearity. Statistical significance was set at p < 0.05.

Ethical considerations: ethical approval for the study was obtained from the Burundi National Ethics Committee (Decision No. CNE/45/2025). Additional authorization was granted by the Ministry of Health, the Central Health District Office, and the heads of participating health facilities. All participants were informed of the study objectives and provided written informed consent before participation. Participation was voluntary, and respondents could withdraw from the study at any time without consequences. To protect confidentiality, no names or personally identifiable information were collected, and all data were stored securely and accessed only by authorized members of the research team.

Results     

Characteristics of participants and vaccination status: Table 1 shows that 379 healthcare workers participated in the study. The mean age was 39.4 ± 8.4 years. Females constituted 65.7% of the study population. Nurses represented the largest professional group (66.5%), followed by laboratory technicians (18.7%), hygienists (7.4%), and doctors (4.5%). Approximately half of the participants worked in public hospitals (52.0%). Most participants were employed in Buyenzi (41.7%) or Rohero (35.6%), and 58.8% identified as Catholic. Overall, 73.4% demonstrated good knowledge of hepatitis B, and 53.3% had a positive attitude toward HBV vaccination. A total of 178 (47.0%) healthcare workers had received at least one dose of the hepatitis B vaccine, whereas 158 (41.7%) had completed the recommended three-dose vaccination schedule (Table 1).

Reasons for non-vaccination: the distribution of reported reasons for non-vaccination is presented in Figure 2. Among the 201 healthcare workers who had not received any dose of the hepatitis B vaccine, the most frequently reported reason for non-vaccination was lack of awareness that the hepatitis B vaccine was available for healthcare workers in Burundi (56.2%), followed by vaccine cost (21.9%) and lack of access to vaccination services (14.4%). Fear of side effects was reported by 4.0% of participants, while only 0.5% believed that vaccination was unnecessary and 2.5% reported other reasons.

Factors associated with HBV vaccination: univariable logistic regression analysis (Table 2) showed that health facility type, health facility location, religion, knowledge level, and attitude toward HBV vaccination were significantly associated with HBV vaccination, whereas sex and professional category were not. In the multivariable logistic regression analysis, health facility type, health facility location, sex, and religion were independently associated with HBV vaccination, whereas knowledge and attitude toward HBV vaccination were no longer statistically significant after adjustment. Compared with healthcare workers employed in public hospitals, those working in private hospitals (aOR: 0.15, 95% CI: 0.07-0.33; p < 0.001), public health centers (aOR: 0.008, 95% CI: 0.001-0.053; p < 0.001), and private health centers (aOR: 0.035, 95% CI: 0.015-0.083; p < 0.001) had significantly lower odds of HBV vaccination. Similarly, compared with healthcare workers working in Buyenzi, those employed in Bwiza (aOR: 0.14, 95% CI: 0.06-0.37; p < 0.001) and Rohero (aOR: 0.40, 95% CI: 0.20-0.78; p = 0.007) were less likely to be vaccinated. Female healthcare workers had lower odds of vaccination than male healthcare workers (aOR: 0.39, 95% CI: 0.21--0.72; p = 0.003), while Protestant healthcare workers had lower odds of vaccination than Catholic healthcare workers (aOR: 0.45, 95% CI: 0.25-0.80; p = 0.007). The final multivariable model demonstrated good fit and excellent discriminatory ability (Omnibus test χ²= 197.06, p < 0.001; Hosmer-Lemeshow test χ² = 7.73, p = 0.461; Nagelkerke R²= 0.541; overall classification accuracy = 81.3%; ROC area under the curve = 0.939), with no evidence of multicollinearity (VIF range: 1.02-1.07). The adjusted odds ratios and corresponding 95% confidence intervals are presented in Table 2, and the magnitude and direction of these associations are illustrated in Figure 3.

Discussion     

This study aimed to determine hepatitis B virus (HBV) vaccination coverage among healthcare workers (HCWs) in the Central Health District of Bujumbura and to identify factors independently associated with vaccination uptake. Overall, less than half of the participants had received at least one dose of the HBV vaccine, and only 41.7% had completed the recommended three-dose vaccination schedule. Multivariable logistic regression identified health facility type, health facility location, sex, and religion as independent predictors of vaccination, whereas knowledge and attitudes were not independently associated with vaccination after adjustment. Among unvaccinated healthcare workers, lack of awareness about the vaccine, vaccine cost, and limited access to vaccination were the most frequently reported reasons for non-vaccination. The overall HBV vaccination coverage observed in this study is suboptimal considering the occupational risk faced by healthcare workers. Although vaccination coverage in the present study was higher than that reported among healthcare workers in several African countries, including Uganda (9%) [15], Tanzania (16.4%) [17], Somalia (25.7%) [30], Senegal (17.6%) [13], Qatar (16.75%) [31], the Democratic Republic of the Congo (6.7%) [20], and Ethiopia (36.9%) [18], it remains considerably lower than the vaccination coverage reported in South Africa (66.9%) [32], Kenya (82%) [33], and Rwanda (96%) [34]. The differences in vaccination coverage across countries may be attributed to variations in national occupational health policies, vaccine availability and affordability, and the extent to which healthcare worker vaccination is mandated or supported by employers or the government [30,31,34,35]. The type of health facility was the strongest independent predictor of HBV vaccination. Compared with healthcare workers employed in public hospitals, those working in private hospitals, public health centers, and private health centers had significantly lower odds of being vaccinated.

Similar findings have been reported in the Democratic Republic of the Congo [20], Tanzania [17], and Ethiopia [18], where healthcare workers in public facilities were more likely to be vaccinated. Although our findings may reflect differences in institutional factors, such as vaccine availability and employer support, these variables were not directly assessed in the present study; therefore, this interpretation should be made with caution. Health facility location also remained independently associated with HBV vaccination after adjustment. Healthcare workers employed in Bwiza and Rohero had significantly lower odds of vaccination than those working in Buyenzi. These geographical disparities suggest that access to vaccination is uneven across the Central Health District and that locally tailored interventions may be needed to ensure equitable vaccine delivery. Future studies should investigate district-level organizational and health system factors that may explain these differences. Female healthcare workers had significantly lower odds of vaccination than male healthcare workers after adjustment for other variables. Although sex was not associated with vaccination in the univariable analysis, its significance in the multivariable model suggests that sex may interact with other factors influencing vaccine uptake. The association between sex and HBV vaccination has been inconsistently reported across studies. While studies from Tanzania [17] and Ethiopia [18,36] observed higher vaccination uptake among male healthcare workers, no significant association was found in a study conducted in Qatar [31].The reasons underlying this disparity could not be determined from the present study but may include differences in occupational roles, employment conditions, access to workplace vaccination, or competing family responsibilities that influence healthcare utilization. Additional qualitative research would be useful to better understand gender-related barriers to HBV vaccination among healthcare workers in Burundi.

Religion was independently associated with vaccination. Protestant healthcare workers had significantly lower odds of vaccination than Catholic healthcare workers, whereas no significant difference was observed among Muslim participants. Although religion itself is unlikely to directly influence vaccination behavior, religious affiliation may reflect differences in community norms, health beliefs, or exposure to health promotion activities. Given the relatively small number of Muslim participants, these findings should be interpreted cautiously. Further research is required to determine whether the observed association represents a true behavioral difference or residual confounding. Other studies have also identified religion as a key factor, with non-christians being less likely to be vaccinated than other religious groups [3]. Knowledge and attitude toward HBV vaccination were associated with vaccination in the univariable analysis but were no longer significant after adjustment. Similar findings have been reported elsewhere: Shindano et al. observed high knowledge alongside low vaccination coverage [20], whereas a study in Ghana found that poor vaccine knowledge was associated with lower uptake [37]. Good knowledge of HBV and a positive attitude toward HBV were reported in Nigeria, but the vaccination rate was low [19]. Together, these findings suggest that knowledge alone may not ensure vaccination when structural barriers, including limited vaccine availability, persist [34-37]. In Burundi, the results support the establishment of a national HBV vaccination program for healthcare workers across public and private health facilities.

An important contribution of this study is the identification of the principal reasons for non-vaccination (Figure 2). More than half of the unvaccinated HCWs (56.2%) reported being unaware that the hepatitis B vaccine was available for healthcare workers in Burundi, while vaccine cost and limited access to vaccination were also frequently cited. In Somalia [30], Mahmoud et al. found that the majority of respondents cited vaccine unavailability as the primary reason for non-vaccination. These findings indicate that the low vaccination coverage observed in this study may be driven primarily by modifiable health system barriers rather than vaccine hesitancy. Burundi currently lacks an occupational HBV vaccination program for healthcare workers, and vaccination is not routinely provided free of charge to all health personnel [22]. The findings support the need for strengthened occupational health policies that ensure equitable vaccine availability, improve awareness among healthcare workers, and reduce financial barriers through government-funded vaccination programs. Such interventions would align with the World Health Organization's recommendations for protecting healthcare workers from vaccine-preventable occupational infections [4]. This study has several strengths. To the best of our knowledge, it is the first to assess HBV vaccination coverage and associated factors among healthcare workers in Burundi, providing important baseline evidence to inform occupational health policy and future research. The study included healthcare workers from both public and private health facilities and used univariable and multivariable logistic regression to identify independent predictors of vaccination while accounting for potential confounding. The final model demonstrated good calibration, excellent discrimination, and no evidence of multicollinearity, supporting the robustness of the findings. However, several limitations should be acknowledged. The cross-sectional design precludes causal inference; vaccination status was self-reported and could not be verified using vaccination records; and the study was conducted in a single health district, which may limit the generalisability of the findings to other regions of Burundi. In addition, occupational exposure characteristics, vaccine availability, and institutional vaccination policies were not directly assessed. Finally, although convenience sampling within health facilities has been widely used in similar studies [38-42], this approach may have introduced selection bias by excluding healthcare workers who were absent, on leave, or otherwise unavailable during data collection, thereby potentially limiting the representativeness of the findings. Future studies should employ probability-based sampling at the health facility level, conduct multicentre studies incorporating objective verification of vaccination status, and use multilevel analytical approaches to provide a more comprehensive understanding of the determinants of HBV vaccination among healthcare workers in Burundi.

Conclusion     

Hepatitis B virus vaccination coverage among healthcare workers in the Central Health District of Bujumbura was low, with less than half of participants completing the recommended three-dose vaccination schedule. Vaccination uptake was independently associated with health facility type, health facility location, sex, and religion, while knowledge and attitude were not independently associated after adjustment for potential confounders. Lack of awareness of the availability of the hepatitis B vaccine for healthcare workers, vaccine cost, and limited access to vaccination were the most frequently reported barriers among unvaccinated healthcare workers. These findings highlight important disparities in HBV vaccination uptake among healthcare workers and provide evidence to inform occupational health policies and targeted interventions to improve equitable access to HBV vaccination in Burundi.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Conception and study design: Billy Joseph Kubenga Tshishiku, Thokozile Ingrid Metsing. Data collection: Billy Joseph Kubenga Tshishik. Data analysis and interpretation: Billy Joseph Kubenga Tshishik and Thokozile Ingrid Metsing. Manuscript drafting: Billy Joseph Kubenga Tshishiku. Critical revision of the manuscript: Thokozile Ingrid Metsing. Supervision: Thokozile Ingrid Metsing. Guarantor of the study: Billy Joseph Kubenga Tshishiku and Thokozile Ingrid Metsing. All the authors have read and agreed to the final manuscript.

Acknowledgments     

The authors thank the Burundi Ministry of Health, the Burundi National Ethics Committee, the Central Health District of Bujumbura, and the heads of participating health facilities for granting permission to conduct the study. The authors also thank all healthcare workers who participated in the study and the data collection team for their contributions to data collection and quality control.

Tables and figures     

Table 1: sociodemographic, professional, and hepatitis B vaccination characteristics of healthcare workers in the Central Health District of Bujumbura, Burundi (N = 379)

Table 2: univariable and multivariable logistic regression analyses of factors associated with hepatitis B vaccination among healthcare workers in the Central Health District of Bujumbura, Burundi (N = 379)

Figure 1: flow diagram of participant selection and inclusion among healthcare workers in the Central Health District of Bujumbura, Burundi

Figure 2: reported reasons for non-vaccination among unvaccinated healthcare workers in the Central Health District of Bujumbura, Burundi (n = 201)

Figure 3: forest plot of adjusted odds ratios (aORs) and 95% confidence intervals for factors independently associated with hepatitis B vaccination among healthcare workers in the Central Health District of Bujumbura, Burundi (N = 379)

References