To benchmark African Union progress toward the World Health Organization (WHO) 2030 hepatitis B elimination targets [
1,
2], we mapped country-level burden in 2010 and 2023, projected age-standardised rate (ASR) trajectories to 2030, and quantified COVID-19-associated departures from pre-2020 trends [
3,
4], we further quantified projection uncertainty using prediction intervals and evaluated robustness using alternative pre-COVID training windows.
We analysed publicly available estimates from the Global Burden of Disease Study 2023 (GBD 2023) for 52 African countries [
5]. The analysed condition was “Total burden related to hepatitis B” as defined in GBD 2023. We summarised ASRs (per 100,000; both sexes unless specified) for hepatitis B-related incidence, prevalence, deaths, and disabilityadjusted life years (DALYs) in 2010 and 2023, and visualised country-level heterogeneity using choropleth maps. Regional and sex-stratified trends (1990–2023) were examined for Africa overall and the five African regions. We projected 2030 ASRs using Nordpred-informed age-specific Poisson regression models with period terms and population offsets. Uncertainty was quantified as 95% prediction intervals (PIs) via 1,000 parametric bootstrap simulations. To address potential COVID-19 data contamination, sensitivity analyses were performed using alternative pre-pandemic training windows (2010–2019 and 2010–2018) alongside the main 2010–2023 model [
6]. To evaluate the COVID-19 impact, we performed interrupted time-series (ITS) segmented regression on annual ASRs (2010–2023) with 2020 as the intervention point [
7-
9]. These analyses were conducted both Africa-wide and stratified by the five African regions to capture heterogeneous disruption and recovery trajectories.
We additionally estimated excess burden for 2020–2023 as the difference between observed and counterfactual ASRs extrapolated from the 2010–2019 pre-pandemic trend [
10].
Marked geographic heterogeneity persisted across Africa in both 2010 and 2023, with comparatively lower ASRs in several North African settings and higher burdens clustering in parts of Central/West/East Africa (
Fig. 1A, 1B).
Across 52 countries, the median incidence ASR declined from 1,920.0 (interquartile range [IQR], 1,693.5–2,556.9) in 2010 to 1,520.5 (IQR, 1,261.3–2,028.3) in 2023 (range, 477.3–3,778.1 vs. 326.7–2,823.9), with declines in all countries. The median prevalence ASR similarly decreased from 7,552.0 (IQR, 5,633.9–11,053.0) to 5675.1 (IQR, 4,287.2– 7,599.3) (range, 1,565.2–15,200.9 vs. 1,086.8–12,070.0), again decreasing in all countries. In contrast, improvements were less uniform for severe outcomes: the median death ASR fell from 18.3 (IQR, 13.4–23.6) to 16.4 (IQR, 12.0–19.7) (range, 6.9–53.3 vs. 4.7–34.9), with 47/52 countries declining and five (Gambia, Mozambique, South Sudan, Burundi, and Lesotho) showing modest increases. The median DALY ASR decreased from 583.9 (IQR, 449.7–797.9) to 554.6 (IQR, 395.3–661.8) (range, 163.2–1,357.7 vs. 118.5–1,212.0), with increases in five countries (Gambia, Côte d’Ivoire, South Sudan, Burundi, and Mozambique) (
Fig. 1A, 1B,
Supplementary Table 1).
The highest-burden countries differed by metric and year: Zimbabwe remained highest for incidence in both 2010 and 2023, whereas prevalence shifted from Mauritania (2010) to Chad (2023). For mortality, Egypt ranked highest in 2010, while Chad ranked highest in 2023; Chad also had the highest DALY ASR in both years. Conversely, Tunisia repeatedly appeared among the lowest-burden countries across multiple metrics in 2010 and 2023(
Fig. 1A, 1B,
Supplementary Table 1).
From 2010 to 2023, ASRs declined across all five African regions for incidence, prevalence, deaths, and DALYs. Relative reductions ranged from -20.9% to -32.7% for incidence, -23.3% to -27.8% for prevalence, -15.5% to -29.7% for deaths, and -14.8% to -31.0% for DALYs. The largest declines were observed in Southern Africa for incidence (-32.7%) and prevalence (-27.8%), and in Northern Africa for deaths (-29.7%) and DALYs (-31.0%). Despite overall improvements, substantial heterogeneity persisted in 2023: incidence ASR was highest in Western Africa (2,110.7; 95% uncertainty interval [UI], 1595.6–2558.0) and lowest in Southern Africa (842.8; 95% UI, 657.1–1056.3); prevalence ASR was highest in Western Africa (8,921.8; 95% UI, 8,161.6–9,727.7) and lowest in Northern Africa (3,249.0; 95% UI, 3,034.6–3,506.3). For severe outcomes, death ASR peaked in Central Africa (20.7; 95% UI, 14.7–27.2) and was lowest in Southern Africa (6.44; 95% UI, 5.42–7.94), while DALY ASR was highest in Central Africa (708.7; 95% UI, 494.0–927.3) and lowest in Southern Africa (212.2; 95% UI, 177.5–262.6) (
Fig. 1C).
The persisting heterogeneity likely reflects differences in vaccination (including timely birth dose and coverage), diagnosis and linkage-to-care, access to long-term antiviral therapy, and competing liver disease risks (e.g., HDV/HCV co-infection and metabolic liver disease), as well as fragility to service disruptions during the COVID-19 period. These contextual factors may help explain why some countries achieved substantial improvements whereas others plateaued or worsened, underscoring the need for locally tailored implementation strategies.
Projections to 2030 suggested continued reductions in incidence and prevalence, but heterogeneous trajectories for mortality and DALYs. For incidence, the median ASR was projected to fall from 1,520.5 (IQR, 1,261.3–2,028.2) in 2023 to 1,232.8 (IQR, 1,041.4–1,699.4) in 2030, decreasing in 51/52 countries (median change, -18.9%); the only increase was projected for South Sudan (+4.47%). In 2030, the highest projected incidence ASRs were Somalia (2,594.7), Mauritania (2,310.6), and Zimbabwe (2,168.5), while Tunisia remained lowest (304.2). For prevalence, the median ASR was projected to decrease from 5,675.0 (IQR, 4,287.2–7,599.2) to 4,726.7 (IQR, 3,436.1–5,893.5), declining in all countries (median change, -16.7%). In 2030, the highest prevalence ASRs were projected for Somalia (11,373.8), Central African Republic (10,161.0), and Chad (10,146.9), with Tunisia lowest (1,047.3). In contrast, the median death ASR was projected to rise from 16.4 (IQR, 11.9– 19.7) to 18.0 (IQR, 12.3–22.5) (median change, +9.6%), with increases in 38/52 countries; the highest projected death ASRs were Chad (38.2), Egypt (34.8), and Sierra Leone (34.7). DALY projections were mixed: the median DALY ASR decreased modestly from 554.6 (IQR, 395.3–661.8) to 537.3 (IQR, 392.2–725.2) (median change, -3.1%), with 28 countries increasing and 24 decreasing. Chad remained highest for DALYs in 2023 and projected 2030 (1,224.7), while Tunisia remained lowest (83.9) (
Fig. 1D,
Supplementary Table 2). Projection patterns were robust across training-window sensitivity analyses (
Supplementary Fig. 1,
Supplementary Table 5). Across all windows, incidence and prevalence ASRs consistently declined in most countries, whereas mortality projections more frequently increased, and DALY projections remained heterogeneous. In the main window (2010–2023), incidence declined in 51/52 countries and prevalence declined in 52/52, while deaths increased in 38/52 and DALYs increased in 28/52; corresponding median relative changes were -18.3%, -20.6%, +8.2%, and +0.9%, respectively. Country-level 2030 estimates with 95% PIs are provided to support uncertainty-aware interpretation.
COVID-19-associated deviations (region-stratified ITS and excess burden). COVID-19–associated deviations (region-stratified ITS and excess burden). To address heterogeneity in pandemic timing and health-system disruption, we repeated ITS segmented regression at both Africa-wide and five-region levels (interruption year=2020) and quantified excess burden as observed ASRs minus counterfactual values extrapolated from 2010–2019 trends. Africa-wide ITS confirmed significant pre-2020 declines for all metrics (incidence, -32.17 per 100,000/year; prevalence, -162.13; deaths, -0.189; DALYs, -2.52; all P<0.05). At 2020, level changes were not significant for incidence and prevalence, whereas DALYs decreased (-28.96; P=0.021) and deaths showed a borderline decrease (-0.505; P=0.051). Post-2020 slope changes indicated attenuation of improvements for prevalence (+55.40; P=0.020), deaths (+0.248; P=0.009), and DALYs (+9.07; P=0.029), while the incidence slope change was not significant (+7.37; P=0.409).
Region-stratified ITS demonstrated non-uniform inflections, particularly for severe outcomes: significant post-2020 slope changes for deaths were observed in Central, Northern, and Southern Africa (all
P<0.05), and for DALYs in Central, Northern, Southern, and Western Africa (all
P<0.05). Consistently, in 2023 observed ASRs exceeded counterfactual expectations for prevalence (+4.41%), deaths (+4.63%), and DALYs (+2.78%); incidence showed a modest excess (+0.66%). The largest 2023 relative excess varied by metric (incidence: Central Africa; prevalence: Southern Africa; deaths and DALYs: Southern Africa), whereas cumulative excess over 2020–2023 was greatest in Central Africa for incidence, deaths, and DALYs, and in Western Africa for prevalence (
Fig. 1E, 1F,
Supplementary Tables 3,
4).
As GBD estimates are modeled, our projections—despite PIs and pre-COVID window sensitivity analyses—should be interpreted as scenario-based trajectories rather than precise forecasts.
Between 2010 and 2023, hepatitis B burden declined across Africa, but progress remains uneven, with persistent country- and region-level disparities. Projections to 2030 indicate that incidence and prevalence are likely to continue falling in most settings, whereas mortality and DALY trajectories are less favourable and heterogeneous, implying that prevention gains are not yet translating into commensurate reductions in severe outcomes. Region-stratified ITS and excess-burden analyses further suggest that the COVID-19 period was associated with attenuated improvements— most notably for deaths and DALYs in several regions— highlighting system vulnerability to service disruption [
11-
14]. To meet the WHO 2030 targets, strategies should prioritise sustaining infant and birth-dose vaccination, scaling testing and linkage-to-care, and protecting long-term antiviral treatment and monitoring. Given the modeled nature of GBD inputs, projections should be interpreted as scenario-based trajectories rather than precise forecasts [
15].
FOOTNOTES
-
Data Availability Statement
All data used in this study are publicly available from the Global Burden of Disease (GBD) 2023 Results tool (https://vizhub.healthdata.org/gbd-results/). The analytical datasets generated from these sources and the code used for data processing and analyses are available from the corresponding author upon reasonable request
-
Authors’ Contribution
Kui Wang: Conceptualization; Study design; Data curation; Statistical analysis; Visualization; Interpretation of results; Drafting of the manuscript; Critical revision for important intellectual content. Ruchen Zhou: Study design; Methodology; Data validation; Interpretation of results; Drafting of the manuscript; Critical revision for important intellectual content. Both authors approved the final version of the manuscript and agree to be accountable for all aspects of the work
-
Conflicts of Interest
We declare there is no any conflict of interest.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 1.
Sensitivity analyses of country-level projections of hepatitis B burden in Africa to 2030 using alternative training windows and 95% prediction intervals. For each of 52 African countries (grouped by the five African regions), age-standardised rates (ASRs; per 100,000) for incidence, prevalence, deaths, and disability-adjusted life years (DALYs) are shown as 2023 observed values (circles) and 2030 projected values (triangles), connected by a line. Horizontal bars denote 95% prediction intervals (PIs) for the 2030 projections. Blue indicates a projected decline from 2023 to 2030, and red indicates a projected increase. Panels correspond to the main training window (2010–2023) and two pre- COVID sensitivity windows (2010–2019; 2010–2018). COVID, coronavirus disease.
Figure 1.(A-F) Hepatitis B burden in Africa (Global Burden of Disease Study [GBD] 2023), projections to 2030, and coronavirus disease 2019 (COVID-19)-associated deviations in trends. (A, B) Country-level choropleth maps of age-standardised rates (ASRs; per 100,000; both sexes) for hepatitis B-related incidence, prevalence, deaths, and disability-adjusted life years (DALYs) in 2010 (A) and 2023 (B). (C) Regional and sex-stratified temporal trends (1990–2023) in ASRs for incidence, prevalence, deaths, and DALYs across Africa and the five African regions (both sexes, males, females). (D) Country-specific Nordpred projections comparing 2023 observed versus 2030 projected ASRs (per 100,000) for incidence, prevalence, deaths, and DALYs; each point represents one country, displayed to visualise cross-country heterogeneity in projected change. (E) Interrupted time-series (ITS) analysis of Africa-wide annual ASRs (2010–2023) with 2020 as the intervention point (COVID-19 period). Points denote observed values; solid lines denote fitted trends; dashed lines denote counterfactual post-2020 trajectories extrapolated from pre-2020 trends. Excess-burden analyses (2020–2023) showing annual differences between observed and counterfactual ASRs (Observed − Expected) for incidence, prevalence, deaths, and DALYs; positive values indicate excess burden and negative values indicate deficits relative to expectation.
Abbreviations
disability-adjusted life year
Global Burden of Disease Study
World Health Organization
REFERENCES
- 1. Lin CL, Kao JH. Global strategies and actions to eliminate hepatitis B virus infection. Clin Mol Hepatol 2025;31:1197-212.
- 2. Pondé RAA, Amorim GSP. Elimination of the hepatitis B virus: a goal, a challenge. Med Res Rev 2024;44:2015-2034.
- 3. Lee K, Park J, Lee J, Lee M, Kim HJ, Son Y, et al. Long-term gastrointestinal and hepatobiliary outcomes of COVID-19: a multinational population-based cohort study from South Korea, Japan, and the UK. Clin Mol Hepatol 2024;30:943-958.
- 4. Cho JY, Lee YS, Kim SS, Song DS, Lee JH, Kim JH. Update on liver disease management during the pandemic of coronavirus disease 2019 (COVID-19): 2021 KASL guideline. Clin Mol Hepatol 2021;27:515-523.
- 5. GBD 2023 Causes of Death Collaborators. Global burden of 292 causes of death in 204 countries and territories and 660 subnational locations, 1990-2023: a systematic analysis for the Global Burden of Disease Study 2023. Lancet 2025;406:1811-1872.
- 6. Howell J, Seaman C, Wallace J, Xiao Y, Scott N, Davies J, et al. Pathway to global elimination of hepatitis B: HBV cure is just the first step. Hepatology 2023;78:976-990.
- 7. Hardie I, Marryat L, Murray A, King J, Okelo K, Fenton L, et al. COVID-19 public health and social measures (PHSM) and early childhood developmental concerns in Scotland: an interrupted time series analysis. Lancet Reg Health Eur 2026;60:101525.
- 8. Lenglart L, Özmen I, Aguilera-Alonso D, Blazquez-Gamero D, Boeddha NP, Buddingh EP, et al. Paediatric invasive group A streptococcal infections and associations with viral infections in 15 European countries after lifting non-pharmaceutical interventions against SARS-CoV-2: an interrupted time-series analysis. Lancet Reg Health Eur 2025;59:101497.
- 9. Choi K, Jang M, Kim S, Park SJ, Suh HS. Impact of COVID-19 on disease-specific mortality, healthcare resource utilization, and disease burden across a population over 1 billion in 31 countries: an interrupted time-series analysis. EClinicalMedicine 2025;85:103315.
- 10. Nørgaard SK, Nielsen J, Schjørring CB, Kalnæs AS, Richter L, Chalupka A, et al. Excess mortality in Europe estimated by EuroMOMO during the COVID-19 pandemic and previous influenza seasons. Nat Commun 2026;17:1356.
- 11. Jeng WJ, Papatheodoridis GV, Lok ASF. Hepatitis B. Lancet 2023;401:1039-1052.
- 12. Seaman CP, Luong P, Xiao Y, Abeysuriya R, Howell J, Hellard M, et al. A global investment case for hepatitis B elimination: a modelling study. Lancet Gastroenterol Hepatol 2023;8:932-942.
- 13. Jin YJ, Kim HY, Suh YJ, Lee CH, Yu JH, Kim MN, et al. Risk assessment of hepatitis B virus-related hepatocellular carcinoma development using vibration-controlled transient elastography: Systematic review and meta-analysis. Clin Mol Hepatol 2024;30(Suppl):S159-S171.
- 14. Lee CH, Choi GH, Choi HY, Han S, Jang ES, Chon YE, et al. Core indicators related to the elimination of hepatitis B and C virus infection in South Korea: a nationwide study. Clin Mol Hepatol 2023;29:779-793.
- 15. Wu D, Kao JH, Piratvisuth T, Wang X, Kennedy PTF, Otsuka M, et al. Update on the treatment navigation for functional cure of chronic hepatitis B: expert consensus 2.0. Clin Mol Hepatol 2025;31(Suppl):S134-S164.
