Prevalence of hepatitis B virus infection and treatment eligibility in Lesotho, Southern Africa: a population-based cross-sectional study with case-based follow-up

Background

Globally, about 2 billion people have been acutely infected with hepatitis B virus (HBV), and over 300 million people have a chronic infection.1–3 Currently, there is no cure for chronic HBV infection, and persons with chronic infection have a 15–40% lifetime risk of liver cirrhosis, liver failure or hepatocellular carcinoma.4 5 The WHO estimates that in 2015, 887 000 people died from hepatitis B-related liver disease, with low- and middle-income countries bearing the highest burden.2 5–7 In sub-Saharan Africa, an estimated 82 million people currently live with HBV.3

WHO has set an agenda for the global elimination of viral hepatitis as a public health threat by 2030.8 HBV vaccines are an effective tool to prevent and control HBV infection, offering nearly 100% protection against the virus.9 10 In the early 90s, WHO recommended the inclusion of the hepatitis B three-dose primary series in national immunisation schedules, with varying implementation by countries.2 Available data show that Lesotho introduced the hepatitis B vaccine in 2003.11 Furthermore, antiretroviral agents such as nucleoside/nucleotide reverse transcriptase inhibitors have the potential to alter the natural history of chronic hepatitis B₁₂, and are recommended in patients with a co-infection with HIV and in HIV-negative individuals with HBV infection and signs of liver inflammation.12 13

However, only a few people are aware of their HBV infection status globally. The WHO reports that in 2016, 10.5% of individuals with HBV infection were diagnosed, and 17% of those diagnosed received antiviral treatment.14 Evidence on the number of people infected with HBV who meet treatment eligibility criteria in different global regions is scant.5 6 Published data indicate that in Africa, only 2% of people living with chronic HBV infection are diagnosed, and 0.1% taking treatment.15 In Lesotho, Southern Africa, few studies report on the prevalence of HBV infection, and if so, only among an HIV-co-infected population.16 17

Our objectives were to estimate the prevalence of HBV infection among the general population in Lesotho, assess the determinants for acquiring the infection and assess treatment eligibility among participants identified to have HBV infection according to the 2024 WHO guidelines.

Method

Design, setting and participants

This is a nested cross-sectional study during a large household-based survey among the general population in two districts in Northern Lesotho to assess the prevalence and determinants of several chronic diseases.18 Butha-Buthe and Mokhotlong districts have an estimated population of about 118 000 and 101 000, respectively. There is an adult HIV prevalence of 18.8% and 18.9% in Butha-Buthe and Mokhotlong, respectively, compared with the national prevalence of 22.7%.19 20 Each district has one central mid-size town, Butha-Buthe with ca. 25 000 inhabitants and Mokhotlong with ca. 10 000 inhabitants. The remaining population lives in rural villages scattered over a mountainous area of 5842 km², predominantly subsistence farmers, mine workers, as well as construction or domestic labourers who work in neighbouring South Africa. Butha-Buthe has 10 nurse-led rural health centres, one missionary hospital, and one government hospital. In Mokhotlong, there are nine nurse-led rural health centres and one government hospital. Figure 1 displays a map of the country, including the village clusters involved in this study.

Request permission

Figure 1

Map of Lesotho. The survey was conducted in Butha-Buthe and Mokhotlong districts, in the village clusters marked with black dots. Map produced by TL using QGIS v3.34 (QGIS.org. QGIS Geographic Information System. Open Source Geospatial Foundation Project. 2024; Available from: http://qgis.org).

This study follows the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines.21 A detailed description of the overall survey, where this cross-sectional study was embedded, has been published elsewhere.18 Briefly, a two-stage cluster sampling was used with the population clusters as primary sampling units and household members as secondary sampling units. From a list containing 785 clusters with at least 30 households each, 120 clusters were randomly selected (60 in each district), stratified by settlement (urban vs rural) and accessibility with respect to a catchment health facility (hard-to-reach vs easy-to-reach areas). All households in a sampled cluster were eligible if consent was provided by the head of household. Household members≥10 years old were randomly selected by an algorithm based on age, sex and settlement (rural vs urban), programmed in the Open Data Kit data collection tool.22 Adult participants provided written informed consent, if illiterate, consent documentation was sought by thumbprint and a witness had to provide a written consent. For participants below 18 years of age, a caregiver provided written consent and from the child verbal assent was sought. The sampled individuals were offered testing.

Results

Figure 2 shows the study flow. Among 7412 eligible participants, 6709 (90.5%) were screened for HBsAg and 6705 had a valid result. Of the 7412 eligible participants, 694 (9.4%) participants could not be screened due to logistical reasons. Overall, 78 had a reactive HBsAg test, yielding a point prevalence of 1.2% (95% CI: 0.9 to 1.4). Prevalence was lower among participants less than 18 years of age (0.24% (95% CI: 0.08 to 0.73)) compared with adult participants (1.4% (95% CI: 1.1 to 1.7)). Among the 78 participants screened HBsAg positive 62 attended the hospital for clinical assessment. Thus, DBS results are available from all 78 but detailed clinical evaluation and follow-up information only from 62/78 (79.5%) participants.

Request permission

Table 1 displays the participants’ characteristics stratified by HBsAg status. The overall median age of participants was 33 years (IQR: 20–53), 1266/6705 (18.9%) were less than 18 years old, and 3509/6705 (52.3%) were female. The most recently published demographic health survey for Lesotho (2014) reports that the proportion of women in Lesotho was 53%, and 23.6% of the population was between 10 and 19 years old.30 This suggests that age and sex distribution were similar to the most recent demographic health survey in Lesotho. Among our participants, 838/6705 (12.5%) reported to be living with HIV. The median age among participants who tested positive for HBsAg was 45 years (IQR: 34–57), and 32 years (IQR: 20–53) for those who tested negative. Among participants who tested positive for HBsAg, 75/78 (96.1%) were≥18 years old, 49/78 (62.8%) were males, 58/78 (74.4%) lived in urban areas, and 26/78 (33.3%) were living with HIV.

Results of univariate and multivariable logistic regression models for HBsAg positivity are given in table 2. Being≥18 years old (adjusted OR (aOR), 4.29; 95% CI: 1.27 to 14.43; p=0.019), male (aOR, 1.88; 95% CI: 1.11 to 3.19; p=0.018), living in urban areas (aOR, 2.54; 95% CI: 1.44 to 4.48; p=0.001), living with HIV (aOR, 3.09; 95% CI: 1.88 to 5.08; p<0.001) and consuming tobacco (aOR, 2.17; 95% CI: 1.23 to 3.82; p=0.007) were associated with increasing odds of HBsAg positivity. Compared with participants within the first (lowest) wealth index quintile, participants in the third, fourth and fifth quintiles had significantly higher odds of HBsAg diagnosis.

For 56/78 (71.8%) participants, the HBV viral load was less than 2000 IU/mL, and for 22/78 (28.2%) participants, the viral load was more than 2000 IU/mL. Online supplemental figure S1 shows HBV viral load by sex, age and HIV status, and tenofovir/lamivudine-containing ART among those living with HIV. Those receiving ART had a significantly lower HBV viral load compared with those not receiving ART: of 23 participants on ART, 2/23 (8.7%) had HBV DNA>2000 IU/mL vs 16/39 (41.0%) among those not taking ART (p=0.007).

Online supplemental table S1 summarises the characteristics of the 62 participants who tested HBsAg-positive who were linked to care and thus underwent a clinical and laboratory assessment. Among them, 35/62 (56.5%) fulfilled the WHO 2024 eligibility criteria for antiviral treatment. 25 (25/35, 71.4%) were eligible due to HIV co-infection (figure 3). 15 participants (15/62; 24.2%), had an APRI score>0.5. 18 participants (18/62, 29.0%) participants had HBV viral load levels above 2000 IU/mL, 29/62 (46.8%) participants had ALT levels above 30 U/L for males/above 19 IU/L for females. Among the 37 mono-infected participants, 10 (27%) were eligible for treatment.

Request permission

In comparison, using the 2015 WHO guidelines,31 valid at the time of the study, to determine treatment eligibility (APRI score>2; or ALT levels above 30 IU/L for men/above 19 IU/L for women and HBV DNA levels>20 000 IU/mL; or presence of HIV co-infection), a total of 33/62 (53.2%) were eligible, 25/33 (75.8%) due to HIV co-infection. Three (3/62, 4.8%) had APRI score >2 and 13/62 (21.0%) had HBV viral load levels above 20 000 IU/mL.

Lowering the treatment threshold of the APRI score from 2 to 0.5 and HBV viral load from 20 000 to 2000 IU/mL in the 2024 WHO guidelines increases the number of eligible participants. However, in our sample, most participants were eligible due to co-infection with HIV and thus, overall, the total number of eligible participants only increased from 33/62 (53.2%; 2015 guidelines) to 35/62 (56.5%; 2024 guidelines) participants (figure 3).

Discussion

This study determined the prevalence of HBV infection among the general population in two districts in Lesotho, Southern Africa, assessed the determinants of HBV infection, and treatment eligibility according to the 2024 WHO guidelines. HBV infection among Basotho was low, but substantially more prevalent among adults compared with younger participants. Male sex, urban settlement, presence of HIV and increasing household wealth were positive determinants of HBV infection. A high proportion of participants linked to care were eligible for antiviral treatment, based on the WHO treatment eligibility criteria. Participants on ART had significantly lower HBV DNA levels compared with participants not on ART. These findings offer insights into factors associated with the risk of HBV infection in Lesotho, suggest a testing and treatment gap especially among people not living with HIV, and emphasise the importance of ART to lower HBV DNA.

Our study addresses an important knowledge gap on the epidemiology of HBV infection in the general population in Lesotho, adding to the relevant health data aggregation, and informing policy and planning in the country. Until now, available data in the setting have focused on hospitalised patients and key populations such as people living with HIV.16 17 Recent Global Burden of Disease hepatitis B estimates for Southern Africa report a prevalence of 4.5% overall, 6.9% for Lesotho, 3.5% and 4.9% for neighbouring South Africa and Eswatini, respectively.2 Compared with these numbers, our study found a lower prevalence of HBV infection. A likely explanation for the disparity is that the Global Burden of Disease estimates are based on the only two publicly available studies on the prevalence of HBV infection in the country—conducted among people living with HIV, with a reported prevalence of HBV/HIV co-infection of 10.5%17 and 4.3%.16 Available data indicate that the Ministry of Health of Lesotho first rolled out the programme on hepatitis B vaccination at birth in 200311 (18 years prior to this study). According to the 2023 WHO and UNICEF estimates for immunisation coverage for Lesotho, the percentage of infants who received the third dose of hepatitis B vaccine has ranged between 87% and 96% over the past decade.32 This certainly has contributed to the low HBV prevalence in the younger individuals included in this study.33 34 A recent study in Malawi reported an HBV infection prevalence of 0.3% among participants born after vaccine introduction35—similar to the 0.24% prevalence observed in our study population aged less than 18 years old.

Beyond age, the associations between male sex, urban settlement, presence of HIV and HBV infection have been documented in other studies.36 37 It has been suggested that men are more susceptible to HBV infection due to behavioural risk factors and the effects of androgen hormones.36 38 39 Interestingly, we found that the odds of HBV infection increased in higher wealth groups compared with the lowest wealth group. Studies conducted within and outside of Africa show that increasing wealth is associated with higher hepatitis B vaccine coverage, and lower HBV infection.35 40–42 However, it has been reported that belonging to the richest income households is associated with having multiple sexual partners which may explain our findings.43–45

The reduced HBV DNA in people living with HIV and taking ART, compared with participants not taking ART shows the effect of tenofovir-based ARTs to suppress HBV replication.46–49 One participant who was not suppressed despite being on tenofovir-based ART could be attributed to non-adherence. In addition, we found a high proportion of HIV-negative individuals who were eligible for treatment (21.6%) but were untreated. From the literature, treatment eligibility of HIV-negative HBsAg-positive individuals varies considerably, ranging from 2% to 22% in the African region.35 50 51 In the absence of routine HBV viral load testing and FibroScan to assess liver architecture, WHO treatment eligibility guidelines are difficult to apply in a setting such as Lesotho. The general focus in these settings is on HIV, while mono-infection with HBV is a neglected health topic. The new 2024 WHO guidelines lower the treatment eligibility thresholds for APRI score and HBV DNA levels and therefore allow more people to be treated, however, the requirement to measure HBV viral load remains. In a setting such as Lesotho with high HIV prevalence, the number of participants eligible for treatment increases only marginally when applying the 2024 guidelines versus the 2015 guidelines. Our study shows that there is a substantial number of people living only with HBV and who are treatment eligible—mostly determined by HBV viral load. More research is needed to identify mono-infected participants and to simplify the treatment eligibility criteria for these individuals in order to reduce the treatment gap. Point-of-care HBV viral load and widespread rollout of HBV viral load testing have been suggested as a solution.52 53

Key strengths of our study are the large sample size, a systematic sampling of the study population and the fact that our sample was a general population rather than a specific key population subgroup. Our study has several limitations. First, because of the nature of the study, the chronicity of HBV infection could not be determined since tests could only be performed once. Second, some participants with HBV infection were lost to follow-up with the potential to introduce a selection bias in our evaluation for treatment eligibility. Third, a FibroScan would have been desirable as an additional treatment eligibility criterion but is not routinely available in Lesotho and was not feasible to introduce during the study period. Finally, for participants with HIV, data on HIV viral load nor duration of ART were not available to provide further insight into potential factors contributing to HBV treatment eligibility, efficacy or resistance.

In summary, we conducted a large-scale population-based study to determine the prevalence of HBV infection in a general population sample in two districts in Lesotho, and assess the determinants of HBV infection and treatment eligibility according to the 2024 WHO guidelines. HBV infection in the population was low, but higher among adults than younger participants, and positively associated with male sex, urban settlement, presence of HIV and increasing household wealth. Being on ART was associated with a significantly lower HBV DNA level. Our study found a significant testing and treatment gap among individuals not living with HIV, who need antiviral treatment. Increased efforts and research are needed to identify and treat people with hepatitis B but without HIV, and simplified treatment eligibility criteria for settings such as Lesotho would be desired.