Background
The prevalence of chronic Hepatitis B virus (HBV) infection remains unequally distributed globally, and the highest disease burden is found in southeast Asia and sub-Saharan Africa (SSA) [
1,
2]. The high endemicity in these regions is of concern because HBV infection is a significant risk factor for the development of end-stage liver disease, including liver cirrhosis and hepatocellular carcinoma (HCC). It is estimated that globally, over 1.3 million deaths in 2017 were because of liver cirrhosis alone. Studies in West Africa have shown that HBV accounts for 38-59% of cirrhosis and chronic liver disease cases [
3‐
6]. The incidence of HCC continues to increase worldwide, and the highest rates of morbidity and mortality occur in sub-Saharan Africa [
7,
8], with HBV infection accounting for 55-70% of HCC cases in the region [
3,
9].
To reduce the morbidity and mortality associated with liver cirrhosis and liver cancer, measures to target risk factor reduction, including the elimination of viral hepatitis, especially in highly endemic countries, are of utmost importance. In view of this, the World Health Assembly set out targets for viral hepatitis elimination by 2030 [
10]. To achieve these elimination targets, interventions such as effective vaccination of viral hepatitis, prevention of mother-to-child transmission, injection, blood and surgical safety, harm reduction practices in injecting drug users and treatment of viral hepatitis are recommended [
11]. There is a need to scale up testing and treatment of HBV infection in sub-Saharan Africa. This in turn requires up-to-date estimates of the disease burden and an accurate picture of the challenges to testing and treatment that exist in countries with high disease burden.
Viral hepatitis poses a significant public health threat in Ghana, and the country is considered to have high endemicity of Hepatitis B infection (i.e., prevalence ≥8%) [
12,
13]. In 2019, there were over 3500 deaths due to viral hepatitis with about 50% of liver cancer deaths attributable to HBV and 14% attributable to HCV [
14]. The national seroprevalence of HBsAg in a systematic review by Ofori-Asenso and Agyeman in 2016 was reported to be 12.3% [
13], whilst a more recent review by Abesig et al. published in 2020 estimated the prevalence at 8.4% in the adult population [
12].
There are several limitations in estimating a nationally reflective HBV prevalence from studies in Ghana, including a lack of data from all administrative regions in the country, the use of estimates based on studies mostly conducted in voluntary and replacement blood donors or specific patient groups, and limited data from children and adolescents. Furthermore, following the creation of new administrative regions by the government of Ghana in 2019, in which the number was increased from 10 to 16 [
15], a large-scale study which describes the current national burden encompassing these new regions is yet to be published. Consequently, there are still gaps that remain in estimation of the district and consequently regional-level, age and gender specific burden of HBV infection in Ghana.
This study investigated the testing patterns for hepatitis B virus infection in Ghana, estimated the age- gender- and region- specific prevalence of HBV infection. Findings from this study provides information that will guide the formulation of policies and interventions to improve hepatitis B testing and treatment services in Ghana.
Methods
Study design
A nationwide cross-sectional study to describe testing patterns and burden of HBV infection using existing hospital-based registers across multiple sites in Ghana.
Sampling approach
Using purposive sampling, Ghana Health Service affiliated healthcare institutions and HBV related non-governmental organizations (NGO’s) were selected for data collection. Selection was based on zoning of Ghana’s 16 administrative regions into the northern, middle, and southern belt. In each zone, two regional hospitals, 2 district-level hospitals, 1 faith-based facility, 1 HBV-related NGOs, public health reference laboratories (PHRL), and teaching hospitals present in the zone were selected by convenience sampling and approached for data collection. Out of 26 sites approached, 22 (84.6%) provided data.
Data collection
From February 2021 to December 2021, paper-based laboratory, blood bank and delivery ward register records spanning entries from 1st January 2016 to 1st January 2021 were reviewed across study sites. Field work in each institution took 2 to 4 days. Data reviewed were reported in registers either as monthly or yearly aggregated results, or as results for each person tested on a case-by-case basis. Data included number of positive cases, number of negative cases and total number tested. Much of the data was from paper-based records that were retrieved from archives. In two centres, data were obtained from electronic health information management systems instead of paper-based records. Where available, data on age and gender within the records were obtained.
Information on types of HBV tests available including point of care (POC) tests such as rapid diagnostic tests (RDTs) for HBsAg and HBV serologic profile, and centralized tests such as Enzyme Linked Immunosorbent Assay (ELISA) were collected for each site for the review period (2021). Additionally, we evaluated the crude number of HBsAg, HBV serological profile including HBV core antibody (HBcAb) and HBV envelope antigen (HBeAg), and HBV DNA testing conducted per facility. We performed testing capacity to determine the proportion of sites visited conducting various types of HBV-related tests, and further evaluated cost of testing. From labour ward delivery registers, we collected data on HBsAg testing of pregnant women, and the proportion of pregnant women who had undertaken a test by the time of delivery across all study sites. Data were extracted using a data extraction form designed for data capture, and extracted data was entered into a Microsoft Excel template.
Statistical analysis
Patient characteristics including age and sex were described using means with standard deviation and median with interquartile range for continuous variables, and frequencies (percentages) for categorical variables. Chi squared test was used to compare proportions in independent groups of categorical variables. We used multivariable logistic regression to identify factors associated with a positive HBsAg test result and adjusted for age (continuous) sex (male, female), year (categorical) and study region (categorical). All tests were two-sided and a p value of less than 0.05 was considered statistically significant. Data analysis was performed using Stata, version 17; StataCorp software.
Discussion
In our evaluation, although registers did not capture reasons for conducting HBV testing, facility personnel reported the primary reason for testing was risk-based or provider-initiated assessment, which is typical of facility-based testing. NGOs included in the study conducted HBsAg testing by population-based screening periodically, depending on funds available, and these were generally considered insufficient in frequency and coverage to identify a significant number of the populace. Studies assessing testing practices in low- and lower middle-income countries found that in general, HBV testing is doctor-initiated, with only few countries in Africa stating that there is a population- or community-based screening programme in place [
16,
17]. In Ghana, HBV testing is risk-based [
18], with no current national or population-based screening programme in place. This method of testing includes screening based on clinical presentation or risk behaviours such as injection drug use. Consequently, estimates derived from risk-based testing are biased and higher than estimates from a random serosurvey. Population-based screening in regions with HBV prevalence of > 2% is recommended by the WHO, especially among pregnant women [
19]. Since Ghana has high endemicity for HBV infection, it is imperative that such an agenda is put into action, with subsequent linkage to care for those testing positive.
In the present study, the proportion of pregnant women who had undertaken HBV screening as at the time of delivery was also determined using labour ward registers. We found that for the year 2020, 94.3% of pregnant women delivering in health facilities had received an HBsAg test, and that the rate of testing among pregnant women had steadily increased between 2017 and 2020. Among pregnant women in Ghana, HBsAg testing is required as part of routine testing for antenatal care registration. Mother-to-child transmission is a major route of HBV transmission in low- and middle-income countries. To achieve elimination goals for HBV transmission, WHO recommends all infants receive hepatitis B birth dose vaccination, to screen all pregnant women for HBsAg and if positive to deliver HBIG if available to their infants. Additionally, it is recommended to deliver appropriate maternal antiviral prophylaxis by screening for evidence of high viral load of HBV based on HBV DNA or if not available HBeAg testing [
20]. Some studies have suggested that the rate of HBV screening among pregnant women in regions of high endemicity are low [
21,
22], therefore the high and increasing proportion of pregnant women tested in Ghana since 2017 is promising. However, it must be noted that our data were obtained from health facilities. Women delivering outside of health facilities may have different health-seeking behaviours and different access to prenatal care and HBV testing rates. Further research is needed to determine the reasons for various health seeking behaviours, to establish barriers to antenatal care and delivery in health centres, to determine the current proportion of women giving birth outside healthcare centres, and to determine the rates of HBV and MTCT in these groups.
Also, data are lacking regarding the proportion of HBsAg+ women who were linked to care for their health, or who received further testing for HBeAg or HBV DNA. A pilot study conducted in the Eastern Region of Ghana found that only 6 and 1% of women received HbeAg and HBV DNA testing, respectively following an HbsAg positive test [
23]. To benefit from HBV testing, it is therefore imperative that the promising screening rates are complimented by comparable rates of serological and HBV DNA testing, and linkage to care, if PMTCT of HBV is to be successful. Furthermore, data are needed in Ghana regarding the proportion of HBV-exposed babies who receive appropriate care (HBV birth dose vaccination and HBIG), as well as outcomes of infants born to HbsAg+ mothers who receive these interventions.
A positive HBsAg result is followed by serological profile testing, HBV DNA testing, and assessments of hepatic structure and function to determine whether treatment criteria are met [
24]. In the present study, we noted a significant mismatch between HBsAg+ results and HBV serological profile and DNA testing. Limitations in testing for HBV serological profile and HBV DNA in LMICs are often a result of cost of assays and lack of government financing of HBV testing, lack of or non-functioning equipment outside of large referral centres and delays caused by centralized laboratory systems [
25,
26]. Such limitations make it difficult to determine which patients meet treatment criteria, and the opportunity to initiate treatment or limit transmission is potentially lost. Conversely, patients who do not require treatment may be initiated on therapy without enough laboratory information. We found that HBV DNA testing was limited to the largest teaching hospitals. Decentralisation of HBV testing, and creation of simple and understandable algorithms for care in LMICs can expand HBV testing and treatment to meet elimination targets. To elaborate, current orthodox algorithms require use of HBV DNA level cut offs based on Hepatitis B e Antigen status, however, simpler algorithms such as TREAT-B which does not require HBV DNA, have been demonstrated to be less costly and applicable in low-resource settings for determining treatment eligibility [
27‐
29]. Another proposed method is to link HBV care to already established national HIV and tuberculosis (TB) control programmes. For example in the national TB programme, PCR testing sites are widely available across Ghana [
30] in the form of GeneXpert® systems. Exploring the use of these systems under dialogue and collaboration with government and funding agencies, may be one way of increasing PCR testing capacity for HBV.
In our study, the crude and pooled HBsAg seroprevalence estimates excluding blood donors and pregnant women were 8.4 and 11.4% respectively. Abesig et al. in their systematic review of studies conducted between 2015 and 2019, found a national prevalence in the adult population of 8.4% with lower prevalence in the northern part of Ghana (5.7%) compared to the south (8.9%) [
12].
Conversely, we found higher region-specific seroprevalence in the Northern regions of Ghana, specifically in the Upper East (15.6%), Upper West (18.0%), Savannah (22.7%) and Northern regions (21.6%). This higher region-specific prevalence in our study, compared to the recent systematic review by Abesig et al. could be because of the differences in the study population between the two studies. In their paper, estimates from the Northern regions were derived from studies in pregnant women, whilst studies in southern Ghana included outpatients, patients with HIV and patients with jaundice. Our study largely comprised data hospital attendants and some community-based screening. There may be multiple factors that contribute to the higher HBV seroprevalence in Northern Ghana found in our study. Disparities in the level of health care compared with the south, challenges in implementing practices promoting maternal and child health, insufficient healthcare staff, lower educational level, socio-cultural factors, and higher poverty rates [
31] are likely to contribute to the increased burden of HBV in these areas. Eliminating HBV in Ghana needs focused policies that will address these gaps in healthcare in Northern Ghana, including increasing testing and treatment capacity, and strategies to prevent mother-to-child-transmission.
In the present study, we found that the HBsAg seroprevalence in children < 5 years was 1.9%. Modelled estimates of the global prevalence in children < 5 years range between 1.3 - 3.4%, and it is estimated that the HBsAg seroprevalence in Ghana for children < 5 years is between 0.9 - 1.4% [
2]. This indicator is important, as it can be used to measure the success of Hepatitis B vaccination programmes, as well as for estimation of the cumulative incidence of chronic HBV infection [
2]. Absolute elimination targets include a seroprevalence of < 0.1% in children under 5 years by 2030 [
32], therefore our estimate suggest that Ghana has not yet reached this target and has just 8 years to do so.
We found that the prevalence among adolescents (15 – 19 year olds) was 6.9%, which was much lower than a previously reported estimate of 14.3% in 2016 [
12]. This is important, because Hepatitis B vaccination at 6, 10 and 14 weeks of age was introduced into the expanded program for immunisation (EPI) in Ghana in 2002. Therefore, the early recipients of this program are in the 15-19 years age group of this current study. Consequently, it is interesting to compare HBsAg seroprevalence among adolescents in our study with that of the 2016 study, since it may reflect a significant decline in HBsAg seroprevalence in a vaccinated cohort. A pilot study conducted in Ghana, supports this finding, in that the HBcAb seroprevalence among school children was higher (6.1%) in children born before the introduction of childhood HBV vaccination, compared with after (2.6%) its introduction [
33]. Furthermore, when the yearly coverage for HBV vaccination among 1-year olds in Ghana in compared over the past two decades, it is noted that in 2002, when the EPI began in Ghana, coverage was 80% compared with 94% in 2012 and 98% in 2022 [
34]. This is important to reflect that increasing vaccine coverage in Ghana has likely contributed to the decreasing prevalence among age-groups below 20 years, as seen in our data.
In our study, the HBsAg seroprevalence among pregnant women (6.4%) and blood donors (5.0%) was lower than previously reported pooled estimates of 7.4 -13.1% and 7.2 - 11.6% respectively from systematic reviews [
12,
13]. This may suggest a decline in HBV burden, since studies used to derive these pooled estimates were conducted between 1995 and 2017, compared with our data which spanned 2016 – 2021. Furthermore, on multivariable analysis, our findings suggested that the odds of testing HBsAg positive were significantly lower for each year between 2017 and 2020, compared with 2016.
The HIV-HBV seroprevalence of 6.1% in this study is lower than a previously reported estimate for SSA [
35]. In Ghana, two studies reported the co-prevalence as 6.1 and 12.3% in the Central and Ashanti regions respectively [
36,
37]. Screening for HBV in PLHIV is important because there is a higher risk drug related liver toxicity as well as HCC development in these patients [
38] Furthermore, screening is necessary for appropriate drug selection in HIV treatment, and to identify HBV negative individuals, for whom HBV vaccination must be offered to reduce the chances of infection.
The strengths of this study include the wide coverage of administrative regions and facility type from where data was obtained, which allowed some comparisons between regions and helped to obtain a potentially nationally representative picture of the patterns of HBV testing in Ghana. It is however noted that sites were selected using non-probability sampling, and that precise comparisons were limited due to the nature of data available. Nonetheless, for the first time, we were able to determine seroprevalence in regions such as the Upper West and Savannah regions, which were not previously available. Furthermore, we provided an updated estimate of seroprevalence in children < 5 years.
Our study was not without limitations. We used secondary data for this assessment, therefore there was some missing data. Additionally, some data, such as the proportion of HBV exposed babies who receive appropriate PMTCT were unavailable because they were not routinely recorded. Furthermore, for regional data, it is possible that a few patients visited health facilities outside their region of residence, therefore this may have had minimal impact on the regional prevalence. Limitations notwithstanding, we believe the large dataset obtained allows for reliable analysis and our subsequent estimates. We also note that a higher pooled seroprevalence from our study may be because data was obtained from hospital laboratory registers. However most previous studies, including those used in the systematic reviews were similarly based on hospital-level data, with few community or population-based studies. We therefore believe our data is comparable to these.
Conclusion and recommendations
In conclusion, Ghana remains a country with a high burden of hepatitis B virus infection. Northern regions demonstrate a higher HBsAg seroprevalence, and a relatively lower proportion of pregnant women undertake HBsAg testing before delivery, compared with regions in middle and southern Ghana. Furthermore, PCR testing is not widely available outside of large teaching hospitals, which limits diagnostic work-up, and there is no subsidy for this test, nor antiviral medication in the National Health Insurance scheme, with patients still currently paying out-of-pocket. As a highly endemic country for HBV, increased testing and treatment of HBV remains a key strategy in overcoming the burden of HBV and its sequelae. Consideration of population-based screening methods may be helpful if systems are put in place to ensure linkage to care. Where gains have been made in testing, for example among pregnant women, this must be complimented with follow through testing for serological profile and HBV DNA, linkage to appropriate care, and PMTCT practices such as birth-dose vaccination. This remains a key part of reducing HBV transmission, as well as preventing the long-term sequelae of chronic HBV infection, such as liver cirrhosis and hepatocellular carcinoma. The foreseeable challenges with implementing these strategies in an LMIC such as Ghana includes funding gaps, therefore a funding plan for elimination is important and necessary to allow planning for appropriate budgetary allocations. Government agencies must be encouraged and political will secured, if LMICS such as Ghana are to be able to undertake such programmes to overcome the burden of disease. Testing sites must move beyond HBV RDT testing alone, and access to serological profile testing and PCR must be expanded especially to Northern Ghana, where burden of disease is highest, so that treatment decisions can be readily made. Furthermore, hepatitis reporting systems and registers should be firstly improved to facilitate data capture of important elimination indicators and standardised across the country to allow for comparability.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.