Hepatitis B virus (HBV) is the causative agent for liver inflammatory diseases, which, if not diagnosed timely and subsequently treated, are likely to progress to liver fibrosis, cirrhosis, and cancer. Despite the presence of a safe and highly efficacious vaccine, the HBV infection is one of the major public health problems (1) with 257 million people chronically living with the virus globally (2). In Africa, the burden stands at 6.1% (3) in the general population but varies with the study cohort (4). In east Africa, a prevalence rate of 6.025% has been recently reported (5). Locally, approximately 1.845 million Ugandans are chronically infected with the virus and are disproportionately distributed in the 10 regions of the country. These range from low through moderate to high endemicity with a national prevalence rate of 4.1% (6) (7). A genetic component has been implicated in regional difference in burden (8, 9). The lowest cases of HBV have been reported in western, Kampala, central, and mid-eastern regions with prevalence rates of 1.8%, 1.9%, 2.0%, and 2.1%, respectively, conventionally taken as low endemic regions. On the other hand, the highest cases have been noted in the mid-north and north-east with prevalence rates of 4.6% and 4.4%, respectively (6). To achieve the ambitious sustainable development goal (SDG) of eliminating HBV infection by 2030 (2), scaling up antiviral treatment becomes inevitable. Unfortunately, accessing the diagnostic procedures to determine treatment eligibility is inadequate and not economically affordable in resource-constrained settings like ours.

Traditionally, liver biopsy has been used as the gold standard in staging the liver diseases and determining the therapeutic relevancy. However, the assay is expensive for implementation in resource constrained parts of Africa (10). Besides, progressive biopsies to monitor the extent of liver damage and stage fibrosis appears not to be clinically feasible (11). Elastography has been endorsed by the World Health Organization (WHO) as a plausible alternative (12). However, it requires sophisticated and expensive equipment restricting its extensive application to countries with developed economies (13). Alternatively, the viral load-based polymerase chain reaction (PCR) testing approach has been adopted, but still, it is very expensive for use in resource-limited settings (14). Notwithstanding the limitations surrounding the current diagnostic procedures to establish treatment eligibility among the chronic hepatitis B (CHB) patients, the WHO has recommended the screening of the general populations for the HBV in the moderate to high endemic countries in order to come up with the number of chronically infected patients that need to be enrolled on treatment (15). Accordingly, the WHO is targeting 80% of the chronic HBV-infected patients that need the anti-viral therapy to be indeed enrolled on treatment by 2030 (16). Besides, the United Nations (UN) Sustainable Development Goal-3 (SDG-3) target 3.3 highlights the combating of hepatitis B virus by 2030 (17). Nonetheless, the rapid diagnostic tests for the HBsAg and HBeAg with high sensitivity have bridged the screening gap to establish the HBV serostatus and virus replication kinetics, respectively, in communities and hospital outpatient settings (18). Unfortunately, these do not determine treatment eligibility. However, incorporation of other correlates of liver damage including AST (aspartate amino transferase), ALT (alanine amino transferase), GGT (gamma glutamyl transferase), ALP (alkaline phosphatase), ALB (albumin), PT (platelet count), and TB (total bilirubin) (19) has been ordinarily used to establish the HBV treatment candidature (20) (21). For example, the APRI (aspartate aminotransferase:platelet ratio) scores have for long guided the treatment decisions. Unfortunately, the liver function tests (LFTs) are non-specific since many cases of drug-induced or alcohol-persuaded liver diseases have been implicated in their alterations (22, 23). Therefore, these cannot be used in isolation to establish treatment eligibility in chronic HBV patients. The currently used algorithm for treatment eligibility that is recommended by the Uganda ministry of health is based on the following parameters: complete blood count to get the platelet count, LFTs particularly ALT and AST, as well HIV serological testing. The HBV viral load and abdominal ultrasound scan are recommended depending on their availability. In addition, the HBeAg test may be done but the guidelines do not advocate for it as a basis for the decision to treat HBV chronic patients in Uganda (24). The disclaimer given by the Ministry of Health pertaining to the inclusion of HBeAg in the algorithm for the decision making to treat or not to treat the HBV chronically infected patients underestimates its potential in scaling up treatment candidacy in communities.

Nonetheless, the ALT in synergy with HBeAg has proved to be a suitable option for the treatment eligibility and has been exploited in the development and validation of a novel HBV infection treatment eligibility algorithm called treatment eligibility in Africa for the hepatitis B virus (TREAT-B) (14, 25). It is a simple score designed based on the ordinary laboratory tests extensively available in the peripheral laboratories in the low- to medium-income countries (LMICs), without depending on the viral load (HBV DNA levels), Fibroscan (elastography) or liver histopathology (liver biopsy). In this model ( Figure 1 ), the total points of TREAT-B are obtained by summing up the HBeAg score as negative (0 point) or positive (1 point) and ALT score in the ranges; <20 IU/L = 0 point, 20–39 = 1 point, 40–79 = 2 points, or ≥80 = 3 points. Thus, TREAT-B scores range from a minimum of 0 for HBeAg-negative and ALT <20 IU/L to a maximum of 4 for HBeAg-positive and ALT ≥80 IU/L (25).

The TREAT-B score strategy snugly fits into the resource-constrained setting unlike other treatment approaches (26) that require specially trained staff and the sophisticated equipment because both ALT and HBeAg are readily accessible and are economically viable in nearly all LMICs (16). Besides, the TREAT-B algorithm requires measurement of ALT using one blood sampling unlike other diagnostic procedures that require frequent visits to the hospital for assessment of ALT (25). Moreover, TREAT-B can use values of ALT adapted from the local laboratory and the cutoff can be adjusted from ≥2 to ≥3 at a relatively lowered sensitivity without compromising specificity in extreme resource poor setting limiting the number of chronic HBV-infected patients under treatment and minimizing those who may not need the lifelong treatment (27). Our study therefore sought to estimate the proportion of the chronic HBV patients eligible for treatment among a sample of HBsAg seropositive treatment-naïve hospital attendees screened at Kibuku health center IV and Kigtum general hospital based on the TREAT-B algorithm using the cutoff of ≥2.

Uganda is a country of moderate to high endemicity for the HBV infection with a projected prevalence of over 4.1% in the general adult population ranging from 0.8% in the south-west to 4.6% in the mid-north (6). The burden is highest among persons living with HIV (7.4%) (30) and is disproportionately distributed in the different regions ranging from 4.7%–8% in Southern Uganda to 6%–12% in Northern Uganda (6, 12). Unfortunately, treatment eligibility candidature for HBV infection is not well documented in our communities. In 2019, the Uganda Ministry of Health developed a national strategic plan for viral hepatitis and released guidelines for the diagnosis and treatment of chronic hepatitis B virus infection. These guidelines are based on the complex and expensive tests and are hence difficult to implement in remote and rural areas (31). With an estimated 1,845,000 people chronically infected with HBV in Uganda (6), based on the Polaris Observatory Collaborators (POC) data (32), it is projected that 647,000 (34.97%) people require antiviral therapy against the hepatitis B virus. However, only 3,000 people (<1%) are currently receiving the antiviral therapy (32) because of the mainly asymptomatic presentation of CHB infection making case detection challenging (33). The WHO target on eradication of the HBV requires a drastic shift from 8% in 2015 to 80% in 2030 of those eligible for treatment to definitely get the therapy (16). To realize this ambitious target, 517, 600 HBV-infected individuals in Uganda should be enrolled on treatment by 2030. Therefore, over the next decade, an additional 517,600 people will have to be enrolled on the antiviral therapy against HBV. Achieving this will not be plausible in the absence of cost-effective and resource-constrained compatible treatment eligibility algorithms. The appraisal of the LFTs as non-invasive markers dates back from the 1970s (20). However, many of them lack specificity since they are produced by multiple organs in response to altered function (34, 35). Presently, the control of the HBV epidemic in Uganda largely depends on collective infant hepatitis B virus immunization program called the Uganda National Program on Immunization (UNEPI) introduced in 2002 (31). However, studies elsewhere have shown that HBV immunization alone will be inadequate to achieve the ambitious United Nations (UN) and World Health Organization (WHO) goal of HBV eradication in the next decade (25). Thus, it is essential to scale up interventions that target screening and treatment at a manageable cost in resource-limited settings to reduce the transmission by suppressing viral load (36). Consequently, the TREAT-B algorithm as an accurate alternative to the current complex treatment criteria that rely on invasive liver biopsy, HBV viral load testing, and elastography is believed to have a profound potential in complementing the current immunization strategy in order to eliminate HBV by 2030.

Here we report the estimation of treatment eligibility among the CHB-infected patients using the TREAT-B strategy previously described by Yusuke et al. (25). Overall, in a sample of 227 CHB-infected treatment-naïve patients that participated in our study from both low and high endemic regions, 56.8% were eligible for treatment using the TREAT-B cutoff ≥2. Comparing the treatment eligibility by HBV endemicity, in a sample of 132 from low endemic and 95 from the high endemic, we found a large proportion of patients in need of antiviral treatment using the cutoff ≥2 of 69.8% and 30.2%, respectively. To the best of our knowledge, this is the first study in Uganda to analyze treatment eligibility using the TREAT-B algorithm among CHB-infected treatment-naïve persons attending the outpatient clinic in a large government hospital from a high endemic region and those coming for screening during the Ministry of Health screening program prior to vaccination from the district health center IV in a low endemic region.

The TREAT-B strategy uses only ALT and HBeAg tests, both of which are easily accessible and can be carried out in resource-limited and rural areas of Uganda making their incorporation in district hospital and health centers plausible. Using the TREAT-B score cutoff ≥2, a large proportion of patients from the low endemic and high endemic region obligated treatment indication. Consequently, using the TREAT-B score, some CHB-infected patients may receive unnecessary antiviral therapy posing significant medical and cost implication burdens since once initiated on antiviral therapy, the treatment is for life with potential toxicity (25, 14). In our study, nonetheless, this can be mitigated by using a score of ≥3, which though compromises the sensitivity has no effect on specificity and can be applied in areas where resources are severely constrained, restricting the number of patients enrolled onto antiviral therapy and minimizing cases of unnecessary lifelong treatment (25). Unfortunately, although the TREAT-B uses both ALT and HBeAg for a decision to treat or not, which are readily available and can potentially be measured as point-of-care (POC) tests, the detection limit of HBeAg rapid diagnostic tests has been questioned and needs to be improved (37).

Our result of 56.8% of HBV-infected patients indicated for treatment is higher than the estimate of 32% reported elsewhere in Asia, 31.1% in Zambia (38), 41.6% in Vietnam (39), and 41.0% among the Burkinabe of West Africa (14) while slightly lower than the eligibility of 64% reported in Hong Kong (40). The differences in the proportion of chronic HBV patients eligible for treatment reported in our study and other studies reported in Zambia and Gambia may be due to differences in the implantation of the prevention and control strategies by the local and regional governments. In Uganda, the data on the circulating genotypes is scanty (41) but in Zambia and Gambia, the predominant genotypes are A1 (38) and E, respectively. These differ in terms of disease severity. Genotype A1 has been allied with more severe disease compared to E (42).

In Uganda, 647,000 CHB patients are qualified for treatment eligibility (32). Therefore, our results suggest that using the simplified TREAT-B score, 1,047,960 hepatitis B chronically infected persons among our study populations will receive treatment, suggesting that 400,960 (21.7%) would be unnecessarily enrolled on the antiviral therapy. Certainly, this has financial implications to already poorly funded health systems in Uganda especially at a time when most of the resources have been diverted to COVID-19 management. However, widening antiviral therapy coverage has a profound potential towards reducing infection kinetics (43). Notwithstanding, our study gave hospital-based estimates whose findings should be interpreted with caution and may overestimate the proportion of persons eligible for treatment. Hence, population-based studies on the TREAT-B eligibility are warranted for more robust results.

Our findings suggest a significantly large number of patients eligible for treatment by the TREAT-B algorithm from the low endemic region compared to the high endemic region (p < 0.05). This is contrary to the study by Veldhuijzen et al. (33) who reported a profound need for treatment eligibility among high endemic migrants compared to those low endemic parts. The high treatment indication for participants from the low endemic region can be accounted for by several observations from our results and reports from literature, the first being the observed significant association between treatment eligibility and alcohol use among the low endemic participants. The odds of treatment eligibility were increased among alcohol users from the low endemic region (p < 0.05). This could be in part responsible for the elevation of ALT as earlier reported by Lin et al. (44) and Vanderlinde et al. (22). Secondly, the participants eligible for the TREAT-B indication from the low endemic region had their AST 9.45 times elevated compared to those not eligible for treatment. The AST has been implicated in alcoholic liver injury. Moreover, alcohol and hepatitis B virus infection are the top causal agents of liver cancer globally (45). This finding is consistent with the earlier reports on the prevalence of alcohol hepatitis in Uganda by Opio et al. (46) in a national referral hospital based study and from the Uganda Non-Communicable Diseases (NCDs) prevalence survey that highlighted a 10% prevalence of liver diseases due to alcohol misuse (47). Besides, alcohol use in Uganda is very high and available data show that Uganda is one of the top countries globally with the highest consumption of alcohol (48). For the high endemic region, the results are inconsistent with the reported trend of alcohol use and liver-related disease (49). This could be attributed to the self-reported alcohol consumption. The participants could have been dishonest with their alcohol intake status. Thirdly, the participants eligible for treatment from the low endemic region had their γ-GT 22.4 times more elevated than those from the high endemic region. This is consistent with the findings by Ban et al. (50). The elevations in γ-GT observed among the exposed subjects to HBV could be predictive of liver-related damage. The γ-GT is the most specific and thus accurate surrogate marker of liver disease, also originating from the kidney as well as from the intestines. Despite this, most of the γ-GT in serum is from hepatocytes (51). In the current study, the γ-GT levels were significantly higher among the low endemic eligible participants for treatment compared to their counterparts from the high endemic region (p < 0.0001). Nonetheless, sclerosing cholangitis, cholecystitis, and alcoholism have also been implicated in causing elevations in γ-GT giving false-positive rate when γ-GT is used alone (52, 53, 54, 55). Fourthly, the participants eligible for treatment from the low endemic region reported significantly higher odds of being born from home rather than from a health facility. This could probably have increased the chance of infection during birth that has over the years contributed to liver diseases (56). Besides, despite the Ugandan Ministry of Health’s ban on the traditional birth attendants in 2010, 80% of the women in rural areas still prefer their services to the formal health services (57). Thus, our results suggest that the Ministry of Health should strengthen and implement the ban imposed upon the traditional birth attendants to curb the vice and reduce the risks of mother-to-child HBV transmission. Fifthly, the observed significant hypoalbuminemia among the low endemic participants eligible for treatment is a marker of liver disease (58) while the reduced odds of hypoalbuminemia among the high endemic participants is indicative of reduced likelihood of liver damage confirming the low treatment eligibility observed in our study. Finally, eligible participants for treatment indication from the high endemic region had significantly low ALP (p < 0.05). The ALP has been implicated in drug-induced liver injury (23). Thus, our results suggest that in the high endemic region, the relative importance of drug abuse in liver disease indication is minimal. In addition, the hyperbilirubinemia observed among the treament indicated that participants from the high endemic region could be attributed to other conditions other than viral hepatitis including dietary toxins (59) and parasitic infections (60) like fascioliasis and hydatidosis (61, 62, 63, 64, 65).

Additionally, treatment indication using the TREAT-B score was significantly higher among those participants who reported household contact with an HBV-infected person for both the low and high endemic regions (p < 0.05). This finding is consistent with earlier reports from related studies by Shedain et al. (66) in Nepal, Muljono (67) in Indonesia, Pereira et al. (68) in Brazil, and Sofian et al. (69) in Iran, which implicated HBV infection to HBV seropositivity of a family member including the mother, father, or a sibling. Moreover, presence of an HBV carrier member in the family has been reported to increase the risk of HBV transmission by 11%–57% in studies by Ragheb et al. (70) in Egypt, Lobato et al. (71) in Brazil, and Mohammad Alizadeh et al. (72) and Sofian et al. (69) in Iran. This is due to the highly contagious nature of the virus, being 100 times more contagious than HIV, and its ability to remain on surfaces for 7 days (56). Thus, our results suggest that HBV carriers should be screened at the household level to reduce intra-familial transmissions. Finally, our advocacy for TREAT-B algorithm appears to be in fair conformity with the observed treatment indication from literature for age 30 years and above (73), AST elevation (74), alcohol use (75), and γ-GGT (76) for both participants from the low and high endemic regions.

The results from our study have thoughtful relevance to global HBV management strategy by filling the knowledge gap pertaining to treatment eligibility for chronically infected HBV patients. Being cheap, the method is economically viable with profound potential for use in resource-limited settings. Accordingly, because of the already weak health systems in most of the Sub-Saharan African countries that have been constrained by the COVID-19 pandemic, these economic compatible approaches should be prioritized.

Our study could not go without limitations. Firstly, the study findings are not representative of the Ugandan general population because we confined the study to outpatients in two hospitals: one in Eastern Uganda and another in Northern Uganda. Secondly, the cross-section nature of the study with only one assessment of the LFTs could have provided an overestimation of people eligible for treatment. Besides, one-off elevation of liver enzymes could not have been quite informative because factors other than HBV can be attributed to liver inflammation. Thirdly, hepatitis Delta could be prevalent among the study participants. However, we were not able to screen for HDV. Fourthly, the intake of alcohol was self-reported and there is a likelihood of under-reporting alcohol intake. Moreover, the influence of other drug abuse was not investigated. Fifthly, the retrospective nature of the study could have been associated with recall bias. Finally, we did not establish other causes of liver inflammation such as drug persuaded injury and exposure to afro toxins, which can potentially elevate the predictors of liver inflammation.

There is a need for screening of the exposed patients to HBV using point-of-care tests in order to establish treatment eligibility in a bid to eliminate HBV in the next decade as projected by the sustainable development goal 4. The concordance observed between the TREAT-B eligibility scores and the correlates of liver damage justify the appropriateness of this technique in bridging the diagnostic gap between the orthodox liver assessment methods like liver biopsy and expensive methods like magnetic resonance imaging (MRI) and computed tomography (CT).

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by the Research and Ethics Committee of the School of Biomedical Sciences, College of Health Sciences, Makerere University (reference number SBS-REC-708) and the Uganda National Council for Science and Technology (UNCST) (reference number HS575ES). The patients/participants provided their written informed consent to participate in this study.

HuK and HS conceived the idea. HuK, AW, and AK participated in the data presentation, analysis, and discussion. HuK wrote the final manuscript draft. PO, DN, HK, CK, EW, HS, JE and DK reviewed the manuscript draft. All authors contributed to the article and approved the submitted version.

This work was funded by the Islamic University in Uganda (IUIU) through the Islamic Development Bank (IsDB) Grant (Grant number: IsDB-IUIUI-001) awarded to HuK by the Research Publications and Innovations (RPI) Department of the Islamic University in Uganda (IUIU) for his doctoral studies. We also received funding from Makerere University, Research and Innovation Fund (Mak-RIF) awarded to HS.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.