Correlates of hepatitis B virus awareness and disease specific knowledge in Uganda: a meta-analysis and meta-regression of studies published between 2018 and 2025

Abstract

Background

Hepatitis B virus (HBV) remains a major public health challenge in Uganda, where prevalence varies by region and population group despite national vaccination and screening initiatives. Understanding awareness and knowledge levels is essential for guiding HBV prevention and control strategies. This study synthesized available evidence on the correlates of HBV awareness and disease specific knowledge among different Ugandan populations.

Methods

We systematically searched PubMed, African Journals Online (AJOL), and Google Scholar for cross-sectional studies reporting on HBV awareness or disease-specific knowledge in Uganda, published between January 2018 and December 2025. Study selection, data extraction, and quality appraisal were conducted independently by two reviewers using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Analytical Cross-Sectional Studies. Eighteen studies meeting the eligibility criteria were included in the final synthesis. Pooled estimates of HBV awareness and correlates of disease-specific knowledge were computed using a random-effects model in MedCalc software (version 10.10), with heterogeneity quantified using the I² statistic. Sources of heterogeneity were explored through subgroup analysis and sensitivity analysis. Meta-regression was performed to examine trends in HBV awareness by region, year of publication, and study population. Publication bias was assessed using Begg’s and Egger’s tests. Statistical significance was set at p < 0.05, and all estimates are reported with 95% confidence intervals.

Results

Eighteen studies met the inclusion criteria. The pooled prevalence of HBV awareness was 75.3% (95% CI: 61.5%–86.9%), with substantial heterogeneity observed across regions and population groups (I²=99.24%). Subgroup analysis revealed that awareness was significantly higher in Central Uganda compared to the northern, eastern, and western regions (p < 0.05), and among healthcare workers compared to the general population and antenatal care attendees (p < 0.05). Knowledge of specific HBV domains varied considerably: awareness of the causative agent, screening services, and vaccination availability was significantly higher among knowledgeable participants compared to those with no prior HBV knowledge (p < 0.05), whereas awareness of post-exposure prophylaxis and completion of the full HBV vaccination schedule remained low compared to those who had received at least partial vaccination (p < 0.05). Awareness of key transmission routes including sexual contact, needle sharing, and mother-to-child transmission was moderate, though knowledge of transmission through infected body fluids remained poor. Participants who were of higher education level were more likely to take up vaccination compared to lower education level (p < 0.05), and participants with a positive attitude toward vaccine effectiveness were significantly more likely to be vaccinated compared to those with a negative attitude (p < 0.05). The meta-regression analysis further revealed a paradoxical overall decline in HBV awareness over the study period.

Conclusion

Critical gaps still persist in HBV awareness and key knowledge areas highlighting the need for alternative strategies tailored to the social dynamics of the current generation during the delivery of health messages.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12879-026-14019-6.

Introduction

Hepatitis B virus (HBV) infection remains one of the most significant global public health challenges. The World Health Organization estimates 296 million persons globally to be living chronically with HBV infection and 820,000 deaths annually due to HBV associated complications such as liver cirrhosis, fibrosis and hepatocellular carcinoma [1]. The HBV transmission has continued to persist in the low-and middle-income countries (LMICs) despite the availability of an effective vaccine and an efficacious antiviral therapy [2]. Sub-Saharan Africa is classified as a high HBV endemic region with prevalence estimates between 6% and 10% maintained by mother-to-child transmission (MTCT), early childhood exposure to the virus, and adult horizontal transmission routes [3, 4]. In eastern Africa, the prevalence of HBV in the general population is estimated at 6.03% [5] and aligns with that reported in Sub-Saharan Africa. Similarly, Uganda is among the high endemic countries with disproportionate regional prevalence estimates varying from 1.9% in the southwestern region to more than 10% in northern Uganda [6, 7] with a population pooled prevalence estimate of 8.3% [8]. There is plenitude of information on studies targeting the biomedical model as a plausible intervention to HBV control and prevention in Uganda [9–14] but little success has been achieved in attaining the united nations (UN) HBV elimination strategy of 2030. Thus, scaling up public health interventions like health promotion practices, screening and vaccination campaigns which have yielded success stories elsewhere [15–17] is paramount. However, awareness and knowledge about HBV infection remain suboptimal and varies across population groups [3, 4, 7, 18–25]. This undermines the prevention efforts such as vaccination uptake, stigma reduction, timely initiation on antiviral therapy and treatment adherence which would otherwise reduce the risk of progression to terminal liver diseases. Additionally, cohort and regional specific variations in levels of HBV awareness and disease related knowledge have been reported [6, 19–21, 25–28] suggesting variations in the demand for the health education. While individual studies provide useful insights, the evidence is usually fragmented and inconsistent [29]. To date, there is no systematic review with meta-analysis and meta-regression that comprehensively quantifies HBV awareness at regional and population levels in Uganda. Additionally, gaps remain in understanding the correlates of HBV knowledge, such as awareness of transmission routes, prevention measures, screening services, availability of vaccination, and knowledge of PEP; areas where previous studies have demonstrated deficiencies even among HCWs who should pioneer the stewardship of community sensitization about infectious [3, 18]. Synthesizing data from studies conducted across diverse settings in Uganda would provide a clearer picture of national awareness levels, highlight regional and demographic disparities, identify factors associated with awareness, and guide targeted public health interventions. Such evidence is essential for informing policy, improving HBV prevention programs, and steering Uganda towards HBV elimination in tandem with the global WHO 2030 targets [30].

Methods

Protocol development and registration

The review protocol was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines 2020 [31]. The research questions, eligibility criteria, search strategy, data extraction strategies and the plan for statistical analysis were specified a prior in the protocol for transparency and reducing reporting bias. The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420251271349.

Research questions

1. What is the overall level of awareness of hepatitis B virus (HBV) infection among populations in Uganda?

2. How does HBV awareness and knowledge vary across different regions and population groups within Uganda?

3. What are the levels of knowledge regarding HBV transmission routes and prevention measures, including screening, vaccination, and post-exposure prophylaxis (PEP)?

Theoretical model

An integrated model anchored in the Socio-Ecological (SE) and Health Belief (HB) [32] to inform the knowledge, attitude and practices (KAP) was used to shape our systematic review with meta-analysis and meta-regression (Fig. 1). The integrated model places awareness about HBV at the centre stage and views awareness as a direct determinant of attitudes as well as key preventive practices such as screening, vaccination and uptake of PEP services. Any disease awareness is influenced by a multitude of factors at multiple levels. For instance, at the intrapersonal level, predictors of awareness include age, sex, education and health literacy. At the interpersonal level, determinants are the advices from health workers. At community level, awareness is influenced by the media exposure and local norms. At the institutional level, the predictors of awareness include access to ANC services and availability of screening/vaccination services. Finally, at the level of policy, awareness is influenced by national immunization programs and advocacy for health promotion campaigns. The health belief model (HBM) components including perceived susceptibility, severity, benefits, barriers, cues to action, and self-efficacy are nested inside the individual level of the social ecological model to explain how knowledge translates or fails to translate into action. The Socio-Ecological framework underscores the role of higher-level contexts such as community, institution and policy in moderating individual responses; for example, the same level of knowledge may produce different vaccination behaviors depending on facility access.

Fig. 1.

Integrated Socio-Ecological (SE) and Health Belief (HB) model to inform the knowledge, attitude and practices

The data from the primary records operationalized three key constructs of the integrated models, viz.: individual level intertwined within the key constructs of the HBM (perceived severity, perceived susceptibility, perceived barriers and perceived benefit), community level and policy levels. At individual level, the variables were: sex, marital status and education level which have a direct influence on the route of transmission, vaccination history, vaccine efficacy, vaccine safety and attitude towards HBV vaccination. At community level, the variables included regions as well as availability of screening and vaccination services in the health units. At policy-level variables included year of publication with a focus on pre/post major HBV campaign milestone, extended program on immunization (EPI) strategies and presence of regional health campaigns. The outcomes measures at individual level were: awareness of HBV, PEP, causative agent, transmission routes, screening services, vaccination, signs and symptoms, perceived risk, willingness to screen or willing to vaccinate. At community level, the outcome measurers were availability of screening and vaccination services in the health units. Finally, policy-level indicators include year of study focusing on pre/post major campaigns or EPI milestones and presence of regional health campaigns.

We further conducted a meta-regression analysis to examine trends in HBV awareness across the study period, stratified by region, year of publication, and study population. Although all included studies employed a cross-sectional design precluding the establishment of temporal relationships, the exploratory findings of the meta-regression generate testable hypotheses that may inform and guide future investigations.

Inclusion and Exclusion Criteria

Studies were eligible for inclusion if they reported on awareness, knowledge, or correlates of HBV disease-specific knowledge among any population group in Uganda, with quantitative outcomes on HBV awareness or knowledge. Only cross-sectional study designs were considered, as this represents the predominant and most feasible design for knowledge and awareness surveys in this context. Eligible studies were restricted to those published in English between 1st January 2018 and 31st December 2025, and conducted within Uganda.

Studies were excluded if they employed designs other than cross-sectional, including case reports, editorials, commentaries, letters, and qualitative-only studies, given that the focus of this review was on the prevalence of awareness, knowledge, and their correlates. Articles that did not report primary data on HBV awareness or knowledge such as review articles were also excluded. Additionally, studies published in languages other than English were not considered, nor were studies in which data on HBV awareness or knowledge could not be disaggregated from information on other viral hepatitis conditions.

Search Strategy

The review team searched for published records from African Journal Online (AJOL), Google scholar and PubMed. The PubMed was the primary biomedical database used to identify peer-reviewed articles on HBV awareness and disease-specific knowledge. The AJOL was included specifically to capture African-published literature that may not be indexed in PubMed, given the regional focus of this meta-analysis and meta-regression on Uganda. Google Scholar was used as a supplementary database to capture additional peer-reviewed studies, grey literature (including theses, dissertations, pre-prints) and studies from journals not indexed in PubMed or AJOL. This was conducted by three reviewers (HMK, AW, AK and MN) between 1st December 2025 to 1st January 2026 and any discrepancies were resolved during the weekly review meetings. Failure to resolve them called for an intervention by the senior reviewers including PO, HK and DN. Only records published between 1st January 2018 to 31st December 2025 in Uganda were selected.

The following search strings were used: “hepatitis B” OR “HBV” OR “hepatitis B virus” AND “awareness” OR “knowledge” OR “perception” OR “attitude” AND “Uganda”, with filters highlighting the study period spanning 2018–2025. Snowballing from the references of the articles to establish the studies citing them was done. Furthermore, manual searching of the references of systematic reviews on HBV in Uganda was done for additional records. Lastly, thesis publication as grey literature was searched for in Google scholar for more records.

The HBV awareness was broadly defined as evidence that a participant had ever heard of HBV. Three main approaches to measuring HBV awareness were identified in the included records. First, some studies used a binary measure, whereby participants were asked directly whether they had ever heard of HBV, with “Yes” classified as aware and “No” as unaware. Second, other studies employed a knowledge-probe approach, in which participants were asked a set of questions covering the organ affected by the virus, routes of transmission, and signs and symptoms of HBV disease; participants who correctly answered at least two out of three questions were classified as aware, while those who answered one or none were classified as unaware. Third, a scale-based approach was also used to assess awareness through a multi-item questionnaire scored on a scale of 0–100%, whereby a score of ≥ 50% indicated awareness and a score of < 50% indicated unawareness. Similarly, records that reported correlates of HBV awareness and disease specific knowledge in Uganda were eligible for inclusion in the data synthesis. The population, exposure, comparison and outcome (PECO) strategy was used throughout the selection process. The population was the Uganda population; the exposure was the awareness of HBV and all correlates of HBV related knowledge. The comparison was between the participants who were aware of HBV infection and it’s correlates with those who were naïve about HBV related knowledge. The outcome was the HBV related knowledge levels.

Risk of Bias Assessment

The methodological quality of all included studies was appraised using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Analytical Cross-Sectional Studies [33]. Three independent reviewers (H.M.K., N.M., and A.W.) participated in the risk of bias assessment, with discrepancies resolved through consensus. Regarding selection bias, studies employing probability sampling demonstrated comparatively lower risk, as they are more likely to yield samples representative of the broader target population. Conversely, studies relying on non-probability sampling were at greater risk of systematic differences between participants and the wider population, potentially limiting the generalizability of their findings. Although most studies adequately described the demographic characteristics of participants, an attribute facilitating comparability assessments across study groups, a notable subset failed to justify their sample sizes, raising concerns about the precision and rigor of the reported HBV awareness estimates.

Measurement bias was also a salient concern. The ascertainment of HBV awareness was predominantly based on structured questionnaires; however, details regarding their standardization and pre-testing procedures were frequently absent from primary records, undermining confidence in the validity and reliability of the measurement instruments. Information bias further cannot be discounted, given the reliance on self-reported awareness and knowledge. In particular, healthcare workers may have been inclined to overreport their knowledge due to social desirability pressures. Additionally, retrospective data collection in some studies introduced the possibility of recall bias, which may have contributed to either overestimation or underestimation of HBV awareness levels and their correlates.

At the aggregate level, considerable heterogeneity across studies in data collection procedures and outcome measurement further elevates the overall risk of bias. Importantly, the exclusive use of cross-sectional designs across all included studies precludes the establishment of causal inferences. Accordingly, the findings of this synthesis should be interpreted with due consideration of these methodological limitations.

Data analysis

Data extraction from the original articles was independently conducted by HMK., ND., NM and AW. The key information extracted for analysis included the first author, year of publication, major geographical region of the country where the study was conducted, study population, sample size, study design, sampling technique, and the level of awareness of hepatitis B virus (HBV) infection. The key effect measure used in the meta-analysis was relative proportion expressed as a percentage. In studies that disaggregated information on HBV specific knowledge, the following data were extracted: - route of transmission & its correlates, vaccination status and correlates of vaccination, screening status and its correlates, causative agent, being at risk, ever heard of PEP and signs & symptoms. For each study, raw numerators (awareness) and denominators (sample size) were extracted from which pooled proportions were calculated along with the 95% CI. Heterogeneity was evaluated by using the I² statistic, with values above 50% indicating substantial heterogeneity. Due to the high heterogeneity in the primary studies, the random-effects meta-analysis (REM) model (DerSimonian–Laird method) was used to pool the correlates of awareness and the HBV specific knowledge [34]. Subgroup analyses were performed based on geographical region (Central, Eastern, Northern, Western Uganda), study population (healthcare workers, out patients, pregnant women, community members, students, and key populations), and sampling method. These subgroup meta-analyses allowed estimation of pooled awareness within more homogeneous groups giving insights into contextual factors influencing HBV knowledge across Uganda.

Potential publication bias was evaluated qualitatively through visual inspection of the funnel plot. Additionally, the Egger’s and Begg’s p values were computed to quantitatively evaluate the publication bias in the primary records. however, interpretation was limited due to the relatively small number of included studies. Symmetrical appearance of the funnel plot and a p value ≥ 0.05 demonstrated the absence of publication bias. In contrast, where asymmetry was observed and/or a calculated p < 0.05, it was attributed primarily to differences in study size, sampling methods, inherent interest by authors to publish manuscripts with only significant values and settings rather than true publication bias.

Sensitivity analyses were conducted by omitting a study with the largest sample size in each category from the pooled model to determine the robustness of the overall estimates. These analyses indicated that the direction and magnitude of the pooled awareness estimates remained stable, suggesting that no single study disproportionately influenced the results. All analyses were performed using the statistical software, MedCalc (version 10.10) available at http://www.medcalc.org and GraphPad Prism (version 10.5.0 (774) at 95% CI and a p < 0.05 was taken to be statistically significant.

Results

Study identification

Initially, the database search yielded 2,139 records; 2,041 from PubMed and 97 from AJOL and 1 record from Google scholar as a thesis publication. 339 were removed as duplicate records leaving 1,800 records for screening. After screening for titles and abstracts, 1,650 records were excluded leaving 168 records for full text review. Of these, 107 records were excluded leaving 61 records for evaluation of their eligibility. Of these, 46 were excluded with specific reasons: 16 were not conducted in Uganda while 27 neither reported awareness of HBV nor HBV related knowledge. Finally, 18 records were eligible for inclusion in data synthesis. (Fig. 2).

Fig. 2.

The PRISMA flow chart for eligibility of studies: Abbreviation: - AJOL: African Journal Online

Characteristics of Eligible Studies for inclusion in the data synthesis

A total of 18 studies published between 2018 and 2025 satisfied the eligibility criteria for inclusion in this systematic review and meta-analysis (Fig. 3). These studies enrolled 7,408 participants drawn from the target populations in the four major regions of Uganda (Central, Eastern, Northern and Western). Ten studies were published between 2018 and 2021 whilst eight studies were published between 2022 and 2025 (Supplementary File S1).

Fig. 3.

Characteristics of included studies: - A: by year of publication, B: by region where the study was conducted, C: by the study group, D: by sampling technique

All included studies were cross-sectional in design since they are the standard and most commonly used approach for measuring prevalence of knowledge and awareness in populations. Moreover, the research questions of this review pertained to the prevalence of HBV awareness and correlates of knowledge, rather than causal or longitudinal outcomes. Besides, all available primary studies from Uganda on this topic were cross-sectional in design, reflecting the nature of the evidence base.

The studies demonstrated a wider regional representation across the country. Nine studies were conducted in the central region, three in the northern region, two in the western region, and three in the eastern region. One study used data from both central and northern region with clearly separate results by region. This ensured a balanced regional coverage, although the central region contributed the highest number of studies.

The study populations varied greatly. Five studies focused on healthcare workers (HCWs), a group at heightened occupational risk of HBV infection and often expected to have higher disease awareness. An additional four studies targeted pregnant women attending antenatal care (ANC), who represent a critical population due to the risk of mother-to-child transmission (MTCT). Three studies were conducted among general community residents to provide insights into population-level HBV awareness. Five studies involved other specialized groups, including men who have sex with men (MSM), female sex workers, medical students and outpatient adults increasing the scope of HBV awareness and disease specific knowledge assessment beyond conventional populations of HCWs and ANC mothers. Sample sizes varied markedly across the studies, ranging from smallest of 91 participants in a eastern Uganda among HCWs [35] to the largest sample size of 1,020 participants in an outpatient population from the eastern region from a study by [28]. These variations contributed to differences in the precision of awareness estimates, with larger studies providing more stable estimates than the small ones.

The sampling techniques reported in the included studies were also variable. Nine studies employed non-probability sampling methods, such as convenience or purposive sampling, particularly in community and facility-based surveys. The remaining nine studies utilized probability sampling approaches.

Overall, the eligible studies used diverse methods, from different regions, conducted on diverse populations across Uganda. This provided a comprehensive dataset with reliable pooled estimates that facilitated elaborate comparisons by region, study population, sampling technique, and year of publication.

Awareness of HBV

Awareness of HBV was pooled from 15 studies comprising a total sample of 5,789 participants, yielding an overall pooled awareness of 75.3% (95% CI: 61.5–86.9%) (Fig. 4). To assess the robustness of this estimate, publication bias was evaluated and none was detected, as demonstrated by the symmetry of the funnel plot (Supplementary file S2) and the non-significant results of both Egger’s and Begg’s tests (p > 0.05) (Table 1).

Fig. 4.

Pooled awareness proportion of HBV in Uganda from 21,018 − 2025: The awareness was indicated as a proportion with corresponding 95% confidence interval. The proportion estimate of each study is indicated by the solid blue square. The size of the square represents the weight contributed by each study in the meta-analysis. The pooled proportion for both fixed effects and random effects is shown by the diamonds

Table 1.

Analysis of HBV awareness from 2018 to 2025 by study period, region, cohort and sampling technique

Awareness			Analysis of HBV awareness		Analysis of heterogeneity and Publication bias
	No	Sample	Proportion%, (95%CI)	P value	I2%,	Model	Egger’s p Value	Begg’s p value
Overall	15	5,789	75.3 (61.5 to 86.9%)		99.24	Random	0.458	0.0703
Year of publication
2018–2021	8	3,166	75.6(56.6 to 90.5%) Ref		99.0	Random	0.6263	0.5212
2022–2025	7	2,623	75.0(52.1 to 92.2%)	0.598	99.37	Random	0.042	0.065
Region
Central	8	2.772	85.2(64.4 to 97.7%) Ref		99.42	Random	0.159	0.258
Northern	4	1,076	75.9(60.9 to 88.2%)	< 0.001	96.36	Random	0.325	0.174
Western	2	515	55.5(39.2 to 70.6%)	< 0.001	90.53	Random	0.001	0.317
Eastern	2	1,423	47.3(25.6 to 69.6%)	< 0.001	98.4	Random	0.001	0.317
Study group
HCWs	4	1,058	95.0(74.1 to 99.3%) Ref		99.02	Random	0.001	0.042
ANC	4	1609	67.6 (45.8 to 86.0%)	< 0.001	98.79	Random	0.8485	0.7098
Community	3	914	73.2 (53.3 to 89.2%)	< 0.001	97.36	Random	0.1559	0.6015
Sex workers	2	785	67.9 (32.4 to 94.4%)	< 0.001	99.07	Random	0.001	0.317
OPD	2	1,423	47.3 (25.6 to 69.6%)	< 0.001	98.41	Random	0.001	0.317
Sampling Technique
Probability	9	3,931	72.1(55.2 to 86.2%)		99.17	Random	0.519	0.113
Non-probability	6	1,935	78.0(53.9 to 96.8%)	< 0.001	99.37	Random	0.953	0.687

Abbreviations: HCWs: Health Care Workers, ANC: Antenatal Care, OPD: Outpatients Department

As shown in Table 1; Fig. 5 the national pooled awareness of HBV was consistent with the pooled awareness in northern region and year of publication but higher than that reported in the western and eastern regions, as well as among antenatal care attendees, sex workers, and outpatients. In contrast, healthcare workers and residents of the central region demonstrated higher levels of HBV awareness compared with the national pooled estimate. Disaggregating the data by regions, the highest level of awareness was observed in the central region, with a pooled estimate of 85.2% (95% CI: 64.4 to 97.7%) derived from eight studies comprising 2,772 participants.

Fig. 5.

Forest plot showing overall and sub-group pooled awareness: - Abbreviations: HCWs: Health care workers, ANC: Antenatal care, OPD: outpatient department

This was followed by the northern region, which had a pooled awareness of 75.9% (95% CI: 60.9 to 88.2%) based on four studies with a combined sample size of 1,076.

In contrast, substantially lower awareness levels were reported in the eastern region, with a pooled estimate of 47.3% (95% CI: 25.6 to 69.6%) from two studies involving 1,423 participants, and in the western region, where pooled awareness was 55.5% (95% CI: 39.2 to 70.6%) based on two studies with 515 participants. Comparative analysis showed that HBV awareness in the central region was significantly higher than that of all other regions (p < 0.05).

When stratified by study population, HBV awareness was highest among healthcare workers (HCWs), with a pooled estimate of 95.0% (95% CI: 74.1 to 99.3%) from four studies involving 1,058 participants. This was followed by community-based populations, which demonstrated a pooled awareness of 73.2% (95% CI: 53.3 to 89.2%) across three studies with a combined sample size of 914. Awareness among sex workers (including men who have sex with men and female sex workers) was slightly lower, at 67.9% (95% CI: 32.4 to 94.4%) based on two studies with 785 participants. The lowest awareness was observed among out-patient department (OPD) attendees, with a pooled estimate of 47.3% (95% CI: 25.6 to 69.6%) derived from two studies including 1,423 participants. Comparative analysis indicated that awareness among HCWs was significantly higher than that of all other study groups (p < 0.05) (Supplementary files S3 -S5).

With respect to sampling methodology, studies employing non-probabilistic sampling reported significantly higher HBV awareness, with a pooled estimate of 69.6% (95% CI: 54.9 to 82.4%) from nine studies encompassing 3,931 participants, compared with studies using probabilistic sampling, which yielded a pooled awareness of 60.4% (95% CI: 41.1 to 78.1%) based on four studies with a total sample size of 1,323 (p < 0.05). There was a substantially high heterogeneity (I² > 95) across all subgroup analyses, therefore, a random-effects meta-analysis model was applied to generate pooled estimates. For all the analyses above, there was no evidence of publication bias detected based on funnel plot inspection for asymmetry (Supplementary files S6-S8) and Begg’s test (p > 0.05). However, Egger’s test suggested possible bias for the western and eastern regions, likely attributable to the limited number of studies included for each region, which may have reduced statistical power.

Correlates of HBV awareness

Data were available for meta-analysis of several correlates of HBV awareness, including knowledge of routes of transmission, the causative agent, risk status, availability of screening services, post-exposure prophylaxis, and signs and symptoms (Table 2; Figs. 6 and 7). For analytical purposes, these factors were dichotomized as aware or unaware. Notably, awareness of these HBV-specific knowledge attributes was substantially lower than the overall pooled awareness of HBV.

Table 2.

Disease specific knowledge of HBV infection on route of transmission, causative agent, risk of infection, PEP and signs & symptoms

Knowledge on HBV	Response			Analysis of disease specific knowledge % (95%CI)		Analysis of heterogeneity and Publication bias
		No	Sample	Proportion% (95%CI)	P value	I2%,	Model	p value
								Egger	Begg
Route of transmission, n = 3448	Aware	8	1806	49.5 (38.6 to 60.2%)		97.52	Random	0.342	0.535
a) Contact with body fluids		2	662	35.8 (2.2 to 82.087%)		99.40	Random	0.001	0.127
b) Sexual		3	947	65.1 (40.4 to 86.12%)	< 0.001	98.38	Random	0.299	0.452
c) Needle sharing		3	947	67.9 (42.8 to 88.6%)	< 0.001	98.46	Random	0.509	0.113
d) MTCT		3	947	63.04 (45.9 to 78.6%)	< 0.001	96.60	Random	0.996	0.608
	Unaware		1642	50.5 (39.7 to 61.3%)	0.558	97.53	Random	0.330	0.706
Causative agent, n = 1417	Aware	3	917	67.97 (31.6 to 94.8%)		99.24
	Unaware		500	32.0 (5.19 to 68.36%)	< 0.001	98.81	Random	0.858	0.602
Being at risk, n = 2829	Aware	7	1720	57.9 (39.9 to 74.8%)
	Unaware		1109	42.12 (25.2 to 60.1%)	< 0.001	98.95	Random	0.997	0.758
PEP, n = 1303	Aware	4	362	27.6 (11.6 to 47.4%)
	Unaware		941	72.4 (52.6 to 88.4%)	< 0.001	97.2	Random	0.0816	0.114
Signs and symptoms, n = 1783	Aware	3	1029	60.6 (51.2 to 69.6%)
	Unaware		754	39.4 (30.4 to 48.8%)	< 0.001	93.1	Random	0.04	0.117

Abbreviations: - PEP: Post Exposure Prophylaxis, MTCT: Mother-to-Child Transmission, CI: confidence intervals

Fig. 6.

Forest plot showing overall and correlates of awareness: - Abbreviations: PEP: Post exposure prophylaxis, MTCT: Mother to child transmission

Fig. 7.

The HBV related knowledge: A: awareness of route of transmission, B: awareness of the causative agent, C: awareness of being at risk, D: awareness of post exposure prophylaxis, E: awareness of signs and symptoms. Abbreviations: - ANC: antenatal care, HCWs: Health care workers, MSM: men who have sex with men, OPD: outpatient department, FSWs: Female sex workers

For those who demonstrated awareness of the route of transmission, we further dug deep in the data from the primary records to establish their knowledge of the particular route of transmission.

Regarding awareness of route of transmission, there was suboptimal knowledge with only 49.5% (95% CI = 38.6 to 60.2%) demonstrating awareness of the HBV transmission pathway.

The reported routes were mainly sexual contact, needle sharing and MTCT. When the data was disaggregated by study cohort, the MSM showed the strongest understanding of the route of transmission. In contrast, community-based studies demonstrated only moderate awareness of the route of transmission. Pertaining the awareness of the causative agent for the infection, 67.97% (95%CI = 31.6 to 94.8%) were aware of the causative agent significantly more than those who were naïve about the causative agent (p < 0.001). By study cohort, knowledge of HBV as a viral causative agent was highest among the HCWs. However, it was sub optimal among the non-healthcare workers. Furthermore, there was fair knowledge on being at risk with 57.9% (95%CI = 39.9 to 74.8%) aware of being at risk significantly more than those who were novice about being at risk (p < 0.001). Awareness of personal risk to infection was generally low across most non-HCW cohorts. Thus, the ANC mothers, OPD attendees, MSM, and FSWs frequently underestimated their susceptibility to HBV infection, even in contexts where exposure risk is known to be high like among the MSM and FSWs. Regarding the availability of pre-exposure prophylaxis (PEP) services only 27.6% (95%CI = 11.6 to 47.4%) were aware of the availability of PEP services significantly fewer than those who were unaware of the PEP services (p < 0.001). Knowledge about PEP was the lowest across all cohorts. Non-the-less, HCWs posted moderate awareness of the availability of PEP. Finally, the signs and symptoms of HBV infection were known by 60.6% (95% CI = 51.2 to 69.6%) significantly more than who were unaware of the signs and symptoms (p < 0.001). Paradoxically, most knowledgeable about the signs and symptoms were the community members.

As earlier reported, substantial heterogeneity remained across studies (I² >95%). and the random effect model meta-analysis was used to pool the awareness. Generally, there was no evidence of publication bias when evaluated using both Egger’s and Begg’s tests for publication bias (p > 0.05).

Awareness of HBV vaccination services and the correlates of vaccination

Data were available for meta-analysis on several correlates of HBV vaccination, including awareness of HBV vaccination, general vaccination status, and vaccination coverage stratified by gender, level of education, and marital status, as well as attitudes toward the HBV vaccine (Table 3; Figs. 8 and 9).

Table 3.

Awareness of HBV vaccination and the correlates of vaccination

Knowledge on HBV vaccination	Response			Analysis of knowledge on HBV vaccination % (95% CI)		Analysis of heterogeneity and Publication bias
		No	Sample	Proportion (95%CI)	P value	I2%,	Model	Egger’s p Value	Begg’s p value
Aware of HBV Vaccination, n = 4062	Yes	7	2332	65.7(40.6 to 86.9%)
	No		1630	34.3(13.1 to 59.4%)	< 0.001	99.62	Random	0.3262	0.2386
Have been vaccinated, n = 2979	Yes	6	797	26.7% (10.6 to 46.9%)
	No		2182	73.3(53.1 to 389.4%)	< 0.001	99.26	Random	0.472	0.1885
Vaccination status by gender, n = 535	Male	3	285	52.7(35.99 to 69.1%)
	Female		250	47.3(30.93 to 64.0%)	0.2131	90.84	Random	0.9113	0.6015
Vaccination status by level of education, n = 458	≤ primary	3	34	12.1(0.4 to 49.17%)
	> primary		424	87.9(50.8 to 99.6%)	< 0.001	98.08	Random	0.4808	0.6015
Vaccination status by marital status, n = 635	Married	4	188	39.9(9.4 to 75.8%)
	Not married		447	60.1(24.2 to 90.6%)	< 0.001	98.66	Random	0.4371	0.4969
Attitude towards HBV vaccine, n = 1140	Effective	3	459	57.5(18.6 to 91.5%)
	Not effective		691	43.04(8.1 to 82.74%)	< 0.001	99.31	Random	0.483	0.601

Abbreviations: - HBV: hepatitis B virus, CI: confidence intervals

Fig. 8.

Forest plot showing the overall awareness and correlates of awareness of HBV vaccination

Fig. 9.

Knowledge, attitudes towards vaccination and vaccination status among the different cohorts: A: overall awareness of HBV vaccination, B: vaccination status, C; complete vaccination, D: vaccine effectiveness. Abbreviations: - ANC: antenatal care, HCWs: Health care workers, MSM: men who have sex with men, FSWs: Female sex workers

Overall, knowledge of HBV vaccination and its associated correlates was generally lower than the overall pooled HBV awareness. However, individuals with education beyond the primary level demonstrated both higher awareness of HBV vaccination and higher vaccination uptake compared with other educational groups. 65.7% (95% CI = 40.6 to 86.9%) from a sample of 2,432 knew HBV vaccination and these were significantly more than those who were naïve about HBV vaccination (p < 0.001). However, only 26.7% (95% CI = 10.6 to 46.9%) from a sample of 797 reported to have received full vaccination. These were significantly fewer than those who were not fully vaccinated (p < 0.001). There was almost equal proportion of complete vaccination for both male and female among studies eligible for inclusion in the data synthesis. When we compared the level of education and complete vaccination status, participants who attained an education level > primary were significantly more likely to be vaccinated (87.9%, 95% CI = 50.8 to 99y.6%) compared to those with ≤primary level of education (p = 0.001).

Regarding marital status, the unmarried were significantly more vaccinated (60.1%. 95% CI = 24.2 to 90.6%) compared to the married persons (39.9%, 95% CI = 9.4 to 75.8, n = 188) (p < 0.001). Finally, 57.5%, 95% CI = 18.6 to 91.5, n = 459) were optimistic that the vaccine is effective and these were significantly more than the pessimists about the effectiveness of the vaccine.

Overall, HCWs exhibited the highest levels of HBV-vaccination related knowledge and the most favorable attitudes towards vaccination. Nevertheless, incomplete vaccination among a proportion of HCWs indicates an unmet need for vaccination in such riskier groups. Among mothers on ANC, knowledge about HBV vaccination and its correlates was generally suboptimal. Key populations, MSM and FSWs, demonstrated predominantly low levels of HBV knowledge and mixed or negative attitudes towards vaccination. As earlier noted, the heterogeneity remained high across studies (I² >95%) and the random effect model was used to pool the prevalence across the sub-groups.

Awareness of HBV screening services and the correlates of screening

The following correlates of HBV screening had data for meta-analysis: -awareness of the availability of screening services, ever been screened, gender and marital status (Table 4; Figs. 10 and 11). Overall, knowledge of HBV screening and its associated correlates was lower than the overall pooled awareness. Non the less, when disaggregated by gender, more female reported to have been screened before.

Table 4.

Awareness of HBV screening and the correlates of screening

Correlate of screening	Status			Analysis of HBV awareness % (95% CI)		Analysis of heterogeneity and Publication bias
		No	Sample		P value	I2%,	Model	Egger’s p value	Begg’s p value
Aware of HBV screening, n = 1990	Yes	6	1098	64.1(33.9 to 89.1%)
	No		892	35.9(10.9 to 66.1%)	< 0.001	99.47	Random	0.4721	0.1198
Have been screened, n = 1897	Yes	4	454	27.3(6.4 to 55.9%)
	No		1443	72.7(44.1 to 96.4%)	< 0.001	99.39	Random	0.337	0.174
Screening status by gender, n = 366	Male	3	159	30.3(4.4 to 66.5%)
	Female		207	69.7(33.5 to 95.6%)	< 0.001	97.57	Random	0.669	0.602
Screening status by marital status, n = 367	Married	4	146	50.7(23.1 to 78.1%)
	Not married		221	49.3(24.2 to 90.6%)	0.793	96.07	Random	0.552	0.6015

Abbreviations: - HBV: hepatitis B virus, CI: confidence intervals

Fig. 10.

Awareness of HBV and correlates of HBV screening

Fig. 11.

Awareness of screening services and screening status among the different cohorts: A: overall awareness of HBV screening, B: screening status. Abbreviations: - ANC: antenatal care, FSWs: Female sex workers, HCWs: Health care workers, MSM: men who have sex with men

The pooled proportion of participants who were aware of the availability of HBV screening services was 64.1% (95% CI: 33.95 to 89.12%), based on six studies comprising a total sample of 1,098 participants. These were significantly more than those who were unaware of the availability of these services (p < 0.001).

Of the pooled screened sample, 69.7% (95% CI = 33.5 to 95.6%) were female significantly more than the screened males. Finally, screening status did not differ significantly by marital status (p > 0.05).

Overall, HCWs demonstrated the highest knowledge on the availability of HBV screening services as well as the highest level of HBV screening. Similarly, MSM demonstrated outstanding knowledge on HBV screening services despite the absence of records on their actual screening. None-the-less, incomplete screening among a proportion of HCWs indicates an unmet need in the control and prevention of HBV in this high-risk group. For the other specific groups including ANC mothers and community members, knowledge about HBV screening was largely low. Critical populations including FSWs, demonstrated mainly low levels of both awareness of the availability of HBV screening actual screening (Fig. 12). Consistent with earlier sub-group analyses, the heterogeneity remained across studies (I² >95%) and the random effect meta-analysis was used to pool the prevalence for each subgroup.

Fig. 12.

Meta-regression analysis showing the awareness of HBV infection (%): - A: overall awareness from 2018 to 2025, B: awareness of HBV infection for studies published from 2018 to 2021, C: awareness of HBV infection for studies published from 2022 to 2025, D: awareness of HBV infection for central region, E: awareness of HBV for northern region and F: awareness among ANC mothers

Meta-regression analysis

The meta-regression analysis reveals a paradoxical pattern in reported HBV awareness across studies published between 2018 and 2025 (Fig. 12). It is important to note that, because the included studies used cross-sectional designs, these findings reflect differences between studies conducted at different time points rather than true temporal trends within populations, and should therefore be interpreted as exploratory and hypothesis-generating. Studies published between 2018 and 2021 reported comparatively higher awareness levels, with minimal between-study variation explained by publication year (R²=0.02). By contrast, studies published between 2022 and 2025 were associated with lower reported awareness, with publication year explaining a moderate proportion of between-study variance (R²=0.38). This pattern may reflect differences in study populations, geographic sampling, methodological approaches, or question framing across time periods, rather than a true decline in population-level awareness.

Regionally, studies from the central region reported comparatively higher awareness overall, while those from the northern region consistently reported lower awareness levels. Among ANC mothers a priority group for PMTCT studies did not suggest upward shifts in awareness, which warrants particular attention regardless of whether this reflects a true trend.

Taken together, these cross-sectional patterns are suggestive of population-specific gaps in HBV awareness and point to an urgent need for renewed, context-specific strategies. Longitudinal and repeated cross-sectional studies are needed to confirm whether these patterns reflect genuine changes in awareness over time.

Sensitivity analysis

A sensitivity analysis was performed to examine the robustness of the pooled estimates of hepatitis B virus (HBV) awareness by assessing the influence of individual studies with the largest sample sizes (Fig. 13). One study with the largest sample size was removed in each relevant analysis and the pooled estimates were re-calculated. Overall HBV awareness was re-assessed after excluding [28]. The overall pooled HBV awareness remained almost unaltered and the heterogeneity remained high (I² = 99.24%) (Supplementary file S9). Similarly, there was no evidence of publication bias as assessed qualitatively by the funnel plot symmetry and quantitatively by the Egger’s and Begg’s tests (p > 0.05) (Supplementary file S10).

Fig. 13.

Forest plot for sensitivity analysis with one study with the highest sample size omitted: The awareness was indicated as a proportion with corresponding 95% confidence interval. The proportion estimate of each study is indicated by the solid blue square. The size of the square represents the weight contributed by each study in the meta-analysis. The pooled proportion for both fixed effects and random effects is shown by the diamonds

Awareness in the Central Region was re-estimated following the removal of [36]. Awareness among healthcare workers was evaluated after excluding [21]. In addition, for studies that employed non-probability sampling methods, HBV awareness was re-analyzed after excluding [37]. The exclusion of a single study with the largest sample size did not alter the pooled estimates of HBV awareness across all analyses. The heterogeneity remained consistently high (I² > 50%), and there was no meaningful change in evidence of publication bias following study removal.

Discussion

This study synthesized evidence from eighteen records conducted in Uganda between 2018 and 2025 to evaluate awareness and knowledge specific correlates of HBV infection among different population groups. Overall, the pooled awareness of HBV was high, with approximately 75% of the respondents reporting awareness of the infection. This suggests good exposure to health information and recognition of HBV as a public health concern [38]. However, the pooled awareness reported in this study is higher than previously reported [3, 4, 39–41] and should be interpreted cautiously because of the inclusion of health care workers in computing the pooled awareness could have biased the overall awareness.

In the subgroup analysis, data were disaggregated by region, study population, and other relevant groupings. Regional analysis revealed that pooled HBV awareness was significantly higher in the Central Region compared to other regions (p < 0.05), suggesting reduced informational barriers in this area [26]. Regional disparities in HBV awareness are consistent with findings reported in literature [8, 42–44] underscoring the need for region-specific resource allocation and targeted public health interventions in the management and prevention of HBV infection. With respect to study population subgroups, healthcare workers demonstrated significantly higher pooled HBV awareness estimates relative to other study populations (p < 0.05). This finding is plausibly attributable to their comparatively greater exposure to Continuous Medical Education (CME) programs and Infection Prevention and Control (IPC) training, both of which are structured to reinforce knowledge of bloodborne pathogens, including HBV, within clinical and occupational settings. Such elevated awareness among healthcare workers is consistent with findings reported across other sub-Saharan African settings, suggesting that institutionalized professional training frameworks serve as meaningful drivers of HBV-related knowledge in this population [45, 46].

Despite the high overall and subgroup pooled awareness estimates, significant gaps in HBV awareness and disease-specific knowledge persist across certain study populations and domains of prevention and control. Notably, deficits in awareness were observed among pregnant women attending antenatal care (ANC) and female sex workers (FSWs), findings that are consistent with those reported in earlier studies [47–52]. Such gaps are of particular concern given that accurate knowledge of transmission pathways, prophylactic measures, and vaccine effectiveness are fundamental prerequisites for informed preventive behavior and effective HBV control at both the individual and population levels [24, 46, 53–56]. These findings corroborate with two cardinal constructs of the Health Belief Model (HBM), namely perceived benefits and cues to action, both of which are instrumental in translating awareness into protective health behaviors. Within this theoretical framework, adequate HBV awareness may be conceptualized as a proximal cue to action, potentially activating health-seeking behaviors such as vaccination uptake. However, the concurrent presence of negative attitudes toward vaccine effectiveness constitutes a significant perceived barrier a construct equally central to the HBM that may attenuate the behavioral impact of awareness by undermining individuals’ confidence in the protective utility of vaccination. This interplay between perceived benefits and perceived barriers underscores the insufficiency of awareness alone as a determinant of vaccine uptake, and highlights the necessity of targeted behavioral interventions aimed at addressing attitudinal barriers to HBV vaccination at the population level. Conversely, good level of awareness has been reported in knowledge specific areas including the causative agent, symptoms, being at risk, screening and vaccination in conformity with previous work [57, 58]. However, gaps continue to persist in the implementation of actual vaccination and screening respectively. This may suggest that while information is reaching the population, it is not translating into practice. None the less, vaccination status was better among the individuals with higher than a primary education, the females and the unmarried.

Better education alludes to better health literacy that may facilitate vaccine uptake. From an HB Model perspective, the better educated are more likely to have increased perceived susceptibility and perceived benefits of vaccination whilst reducing perceived barriers, ultimately strengthening self-efficacy to act. Similar reports have been reported in Uganda and other sub-Saharan African settings [3, 4, 23].

The better HBV screening level of the females is most likely attributable to the routine screening of women during antenatal care visits consistent with earlier reports [59, 60]. According to the SE model, this serves as an institutional-level context for health education. However, as a forementioned, higher screening status among pregnant women does not necessarily translate into better knowledge about HBV and associated preventive interventions. This anomaly underscores the missed opportunities in the healthcare systems and suggests that awareness should be done alongside the counselling services [23]. Finally, the unmarried individuals demonstrating higher awareness than their married counterparts appear paradoxical since previous studies have reported otherwise [61, 62]. Non the less, the un-married people are generally young and live in the urban with better access to health information. However, the data on age and residence was not available for meta-analysis to justify our hypothesis.

Viewed through the lens of the Socio-Ecological Health Behaviour (SE-HB) framework, the cross-sectional patterns identified in this review raise exploratory questions about the sustained reach and engagement of HBV health communication efforts in Uganda. While this review cannot directly evaluate the effectiveness of national awareness campaigns, the lower awareness reported in more recent studies may suggest that current health promotion approaches merit re-examination. Drawing on the theoretical framework used in the current study, one plausible hypothesis is that prolonged, uniform messaging may reduce its salience over time, consistent with habituation theory and literature on cue-to-action fatigue [63, 64]. However, it must be emphasised that these are theoretical propositions, not empirical findings of this review. Whether message fatigue, reduced community engagement, or structural gaps in health service integration contribute to the observed patterns remains to be tested in future primary studies.

Separately, evidence consistently suggests that health promotion interventions are more likely to produce durable change when reinforced through community participation and integrated into routine health [65]. Taken together, these considerations point to the value of exploratory, formative research to understand the social and communicative dynamics shaping HBV awareness among current Ugandan populations particularly among high-risk groups such as ANC mothers before attributing the observed patterns to any specific programmatic cause.

Regional and cohort-specific declines in HBV awareness particularly among ANC mothers is suggestive of persistent policy-level gaps as reported by [66] and [67]. These findings suggest the need to adopt local interventions that strengthen local health support systems and individual-level inspirations. Interestingly, even among HCWs, knowledge about HBV transmission, prevention and control remained suboptimal [18]. Similar anomalies from health professionals have been reported in Nigeria and Ethiopia, emphasizing gaps in medical training and CME [22].

Finally, the meta-regression analysis examining publication year as a moderator of HBV awareness revealed a modest positive association with awareness levels between 2018 and 2021. However, the between-study variance was only 2% (R²=0.02) indicating that temporal factors played a negligible role in HBV awareness within this period. In contrast, for studies published between 2022 and 2025, a negative association with awareness was detected with the between-study variance of 38% (R²=0.38), suggesting that temporal factors may play a more meaningful role in HBV awareness within this more recent period. None the less, meta-regression with a limited number of eligible studies is compromised by the low statistical power and is susceptible to bias limiting its use in hypothesis-generating.

Conclusion

This systematic review and meta-analysis reveals critical gaps in HBV awareness across Uganda, with particularly pronounced deficiencies in knowledge of screening opportunities, vaccination services, post-exposure prophylaxis availability, and routes of transmission. These domain-specific gaps constitute meaningful structural barriers to Uganda’s national efforts to curtail HBV transmission and advance progress toward elimination targets.

The findings underscore the need for health education interventions that transcend general awareness to systematically address specialised knowledge domains, honed to the contexts of high-risk and underserved populations including antenatal care attendees and communities in the northern region. It must be acknowledged, however, that all included studies were cross-sectional in design; the patterns observed across study publication periods therefore reflect differences between study populations and settings rather than confirmed temporal trends, and should be interpreted as exploratory rather than directional. These patterns nonetheless raise substantive programmatic questions: whether existing health communication strategies are effectively reaching the most vulnerable populations; whether messaging retains salience across socially and geographically diverse communities; and whether integration of HBV awareness into routine health service delivery remains adequate. Resolving these questions will necessitate longitudinal and repeated cross-sectional studies with the capacity to track awareness trajectories within defined populations over time.

Recommendations

To address the persistent gaps in HBV awareness and knowledge identified in this systematic review with meta-analysis and meta-regression, several recommendations are proposed. First, Uganda’s Ministry of Health may consider strengthening region-specific health education campaigns, with priority given to underserved areas such as the northern and eastern regions, where awareness levels appear comparatively lower. Second, integration of HBV education into routine antenatal care warrants prioritization, given the observed associations between healthcare contact and awareness levels in the included studies. Third, incorporating HBV awareness content into vaccination campaigns and maternal health services may help extend the reach of existing health messaging infrastructure. Fourth, continuous professional development programs for healthcare workers should be enhanced, with particular emphasis on accurate HBV transmission routes, infection prevention measures, and PEP protocols, given that knowledge gaps were documented even among clinical populations in the reviewed studies. Fifth, investment in community-based health promotion interventions leveraging community health workers, peer educators, and local radio programming may be associated with broader reach and improved uptake of health messages, particularly in rural settings. Finally, aligning the delivery format of health messages with the communication preferences and generational needs of target populations is recommended. Together, these measures represent a context-informed approach to strengthening Uganda’s HBV prevention landscape and supporting the country’s progress toward WHO hepatitis elimination targets.

Limitations

The very high between-study heterogeneity (I² >95%) observed across all pooled analyses limits the interpretability of the pooled prevalence estimates and warrants explicit caution in drawing conclusions Secondly, inclusion of the HCWs could have overestimated the HBV awareness levels. So, the results must be interpreted while considering this methodology limitation. Finally, our search for records was limited to only PubMed, African Journal Online and Google scholar which have compromised data richness in our meta-analysis.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AJOL

African Journal Online

ANC

Antenatal Care

CME

continuous medical education

COVID-19

Corona Virus Disease-19

EPI

Expanded Program on Immunization

FSWs

Female sex workers

HB

Health Belief

HBM

Health Belief Model

HBV

Hepatitis B virus

HCWs

Healthcare Workers (HCWs)

JBI

Joanna Briggs Institute

KAP

Knowledge, Attitude and Practices

LMICs

Low-and Middle-Income Countries

MOH

Ministry of Health

MSM

Men who have sex with men

MTCT

Mother-to-Child Transmission

NOS

Newcastle–Ottawa Scale

OPD

Outpatient department

PECO

Population, Exposure, Comparison and Outcome

PEP

Post-Exposure Prophylaxis

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

Prospective Register of Systematic Reviews

REM

Random-Effects Meta-Analysis

SE

Socio-Ecological

UN

United Nations

Author contributions

H.M.K. and P.O. conceived the idea. H.M.K., M. N., A.W. and A.K. participated in the search for the articles from the data bases. H.M.K., D.N., M.N and A.W. extracted, presented, analyzed and discussed the results. H.M.K. wrote the final manuscript draft. P.O., D.N., F.M., M.N and H.S. reviewed the manuscript draft. All the authors read and approved the publication of the manuscript.

Funding

The authors declare that they did not receive any specific funding for the study.

Data availability

All data generated or analyzed during this study are included in this published article and as supplementary materials.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.