Abstract
Background
As the third leading cause of cancer-related mortality globally, liver cancer exhibits a rising metabolic risk profile, with high BMI emerging as a critical etiological driver.
Objective
To systematically quantify the global burden of liver cancer disease attributable to high BMI from 1990 to 2021 and project its epidemiological trajectory through 2036.
Methods
Using the Global Burden of Disease data for 2021, we analyzed mortality rates, disability-adjusted life years (DALYs), and age-standardized rates categorized by sex, age, time cohort, and sociodemographic index (SDI) strata. We employed spatiotemporal clustering to identify regional patterns of burden, and utilized autoregressive integrated moving average models to predict future trends.
Results
In 2021, high BMI accounted for 46 200 (95% uncertainty interval: 42 800–49 700) liver cancer deaths and 1.24 million (95% uncertainty interval: 1.15–1.33) DALYs globally, representing 3.5 and 3.2% increases from 1990, respectively. The burden of deaths and DALYs demonstrated significant sex differences (male to female ratio of 1.6 : 1 and 1.9 : 1, respectively), age-dependent progression (peak mortality at 90–94 years for men: 6.54 per 100 000; peak mortality at 95+ for women: 6.53 per 100 000), and SDI-associated increases (R2 = 0.039, P < 0.0001). Projections indicate that the age-standardized death rate will rise to 17.63 per 100 000 (95% uncertainty interval: 16.83–18.44) by 2036, representing a 120% increase from 2021 levels.
Conclusion
High BMI-associated liver cancer burden has intensified globally since 1990, disproportionately affecting males, aging populations, and high-SDI regions. Accelerated epidemiological transitions and persistent obesity trends suggest continued burden escalation without targeted metabolic intervention strategies.
Keywords: disability-adjusted life year, Global Burden of Disease 2021, high BMI, liver cancer, mortality
Introduction
Global cancer statistics for the year 2022 indicate that liver cancer ranks sixth in incidence, with 865 269 new cases, and third in mortality, with 757 948 deaths among malignant tumors [1]. Based on joint modeling conducted by the Global Burden of Disease (GBD) study, the Collaborative International Cancer Registry Project (CI5plus), and the WHO, it is projected that the number of new liver cancer cases worldwide will reach 1 392 474 by 2040. This represents an increase of 53.8%, corresponding to an average annual growth rate of 2.2%, compared with the baseline level of 905 347 cases in 2020 [2]. Notably, although areas endemic to hepatitis B virus (HBV) and hepatitis C virus (HCV) (e.g. China and Southeast Asian countries) have demonstrated a decreasing trend in the incidence of virus-associated liver cancer because of the widespread adoption of vaccination and antiviral therapy, the proportion of liver cancer attributed to metabolic risk factors (e.g. obesity and diabetes) has increased [3]. Furthermore, incidence rates in traditionally low-risk regions such as Europe, America, Australasia, and South America continue to rise [1], suggesting a structural shift in the etiological spectrum. The dominance of traditional liver cancer risk factors (HBV/HCV infection and alcoholic liver disease) is increasingly being challenged by metabolic risk factors. Global epidemiological data indicate that the attributable proportion of liver cancer associated with nonalcoholic fatty liver disease (NAFLD) rises to 42.6% in patients without viral hepatitis (HBV/HCV) and alcoholic liver disease [4]. Notably, a prospective study from the USA revealed that the prevalence of NAFLD was as high as 38% among the obese population with a BMI greater than or equal to 30 kg/m² [5]. Given that the prevalence of obesity continues to rise globally (with WHO data predicting that nearly 50% of adults will be overweight or obese by 2030) [6], the disease burden of NAFLD-associated liver cancer is projected to become a priority area for public health intervention in the coming decade.
Trends in the evolution of the GBD indicate a significant correlation between obesity-related metabolic disorders and mortality from liver cancer. Evidence-based research suggests that BMI and type 2 diabetes mellitus synergistically promote hepatocellular carcinogenesis through the insulin resistance-inflammation axis [7]. The underlying mechanism involves the dysregulation of adipokines, which activates hepatic stellate cells and disrupts the gut flora-hepatic metabolism axis [8]. Notably, obesity-related metabolic syndrome has emerged as the third most important preventable carcinogenic factor, following smoking and infections. A systematic assessment by the International Agency for Research on Cancer confirmed a dose–response relationship between high BMI with 13 malignancies (including liver cancer and postmenopausal breast cancer) [9]. Regional epidemiological analyses reveal significant heterogeneity: Asian populations exhibit a steeper BMI-liver cancer risk curve because of genetic susceptibility (e.g. the PNPLA3 rs738409 polymorphism) and changing dietary patterns (Chinese cohort study showed an average annual growth rate of 6.7%) [3]. Furthermore, with the rising prevalence of metabolic dysfunction–associated fatty liver disease in Western countries, obesity is projected to surpass viral hepatitis as the leading risk factor for liver cancer by 2030 [10]. Given the modifiable nature of obesity and the evolving landscape of its social determinants, the development of multidimensional disease burden prediction models has emerged as a critical priority for optimizing prevention and control strategies. Utilizing data from GBD 2021, this study aims to: (a) quantify the evolving trajectories of liver cancer burden [mortality and disability-adjusted life years (DALYs)] attributable to high BMI from 1990 to 2036; (b) analyze its interaction effects with sociodemographic index (SDI), gender stratification, and age-specific population structures; and (c) identify high-risk subgroups to provide evidence-based decision support for establishing risk-stratified surveillance systems and precision prevention strategies.
Methods
Data resource
This study utilized population-level data from the GBDs, Injuries, and Risk Factors Study 2021 (GBD 2021), a comprehensive epidemiological repository spanning 204 countries and territories from 1990 to 2021. Mortality and DALYs attributable to liver cancer caused by high BMI were extracted through the GBD Results Tool. This study utilized the database to analyze the global burden of liver cancer attributable to high BMI. All data were extracted through the GBD Results Tool GBD 2021 (https://ghdx.healthdata.org/gbd-2021/sourceshttps://ghdx.healthdata.org/gbd-2021/sources). All datasets were deidentified and publicly accessible in compliance with GBD data use policies, exempting ethical approval and informed consent requirements. Following GBD risk factor methodologies [10], high BMI was defined as a BMI greater than 25 kg/m² for individuals aged 20 years and older [11].
Sociodemographic index
The SDI, a composite development metric endorsed by the Global Burden of Disease Collaborative Network, quantifies national development status through three principal determinants: (a) total fertility rate among women aged less than 25 years, (b) mean years of educational attainment for individuals greater than or equal to 15 years, and (c) lag-distributed income per capita. This index generates a standardized composite score ranging from 0 (lowest development) to 1 (highest development), derived through principal component analysis of normalized component indicators. Based on this metric, 204 countries and territories were stratified into quintile developmental tiers: low SDI (<0.45), medium-low SDI (0.45–0.61), medium SDI (0.61–0.69), medium-high SDI (0.69–0.80), and high SDI (≥0.80) [12]. This categorization aligns with GBD analytical protocols to ensure cross-study comparability in evaluating disease burden-development gradients.
Statistical analysis
The primary metrics comprised mortality and DALYs for liver cancer attributable to high BMI, quantified through age-standardized mortality rate (ASMR) and age-standardized DALY rate (ASDR). All estimates were standardized per 100 000 population with 95% uncertainty intervals derived via 1000 Monte Carlo simulations. To delineate spatiotemporal burden patterns across GBD regions, we performed agglomerative hierarchical clustering using Ward’s minimum variance method, with estimated annual percentage change (EAPC) as the dissimilarity metric. EAPC was computed as: EAPC = [exp(β) − 1] × 100%, where β represents the slope coefficient from log-linear regression models. Temporal trends were classified as: significant increase: EAPC greater than 0 with 95% uncertainty interval lower bound greater than 0; significant decrease: EAPC less than 0 with 95% uncertainty interval upper bound less than 0; stable: 95% uncertainty interval encompassing 0. Disease burden projections through 2036 were generated using autoregressive integrated moving average (ARIMA) models, with optimal (p, d, q) parameters determined by minimizing Akaike information criterion. All analyses were implemented in R v4.3.1 (R Foundation), utilizing the gbd and forecast packages for GBD data harmonization and time-series modeling.
Results
The global burden of high BMI-attributable liver cancer
From 1990 to 2021, global liver cancer deaths attributable to high BMI surged from 10 300 (95% uncertainty interval: 4200–16 700) to 46 200 (95% uncertainty interval: 18 600–78 000), with ASMR increasing by 103.8% (0.26–0.53 per 100 000). Concurrently, DALYs rose from 292 700 (95% uncertainty interval: 119 600–476 000) to 1 237 300 (95% uncertainty interval: 504 200–2 102 000), accompanied by a 103.2% increase in ASDR (6.97–14.16 per 100 000) (Table 1). Persistent gender disparities were observed: males exhibited 1.6-fold higher mortality (28 500 deaths; 95% uncertainty interval: 11 700–49 300) and 1.9-fold greater DALYs (803 500; 95% uncertainty interval: 330 500–1 391 500) compared with females in 2021. Age standardization amplified this gap – male ASMR (0.70/100 000; 95% uncertainty interval: 0.29–1.21) and ASDR (19.05/100 000; 95% uncertainty interval: 7.83–32.99) exceeded female rates by 1.8- and 2.0-fold, respectively (Fig. 1).
Table 1.
The number of deaths and disability-adjusted life years, age-standardized rate of death and disability-adjusted life years, and its annual growth rate with the 95% uncertainty interval (lower, upper) of high BMI-attributable liver cancer by genders, sociodemographic index groups, and Global Burden of Disease regions
| GBD 2021 super regions | Liver cancer–related death caused by high BMI | Liver cancer–related DALYs caused by high BMI | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| All-ages number | Age-standardized rate | Annual change of number (%) | Annual change of rate (%) | All ages number | Age-standardized rate | Annual change of number (%) | Annual change of rate (%) | |||||
| (×1000, 95% UI) | (/100 000, 95% UI) | (×1,000, 95%UI) | (/100,000, 95% UI) | |||||||||
| 1990 | 2021 | 1990 | 2021 | 1990 | 2021 | 1990 | 2021 | |||||
| Global | 10.3 (4.2–16.7) | 46.2 (18.6–78.0) | 0.26 (0.11–0.42) | 0.53 (0.21–0.90) | 11.2 | 3.3 | 292.7 (119.1–476.0) | 1237.3 (504.2–2102.0) | 6.97 (11.34,2.84) | 14.16 (5.77,24.06) | 10.4 | 3.3 |
| Gender | ||||||||||||
| Male | 5.9 (2.5–9.7) | 28.5 (11.7–49.3) | 0.31 (0.13–0.52) | 0.70 (0.29–1.21) | 12.4 | 4.1 | 178.6 (74.6–292.2) | 803.5 (330.5–1391.5) | 8.75 (3.66,14.34) | 19.05 (7.83,32.99) | 11.2 | 3.8 |
| Female | 4.4 (1.7–7.1) | 17.7 (7.2–29.5) | 0.21 (0.08–0.33) | 0.38 (0.15–0.64) | 9.7 | 2.6 | 114.1 (44.5–185.2) | 433.8 (178.3–726.3) | 5.26 (2.05–8.53) | 9.52 (3.91–15.9) | 9 | 2.6 |
| SDI grouping levels | ||||||||||||
| High SDI | 3.4 (1.4–5.8) | 14.1 (5.8–23.4) | 0.31 (0.13–0.53) | 0.69 (0.28–1.13) | 10.2 | 4 | 88.1 (37.1–149.1) | 323.0 (133.9–530.4) | 8.29 (3.49–13.99) | 17.19 (7.16–28.19) | 8.6 | 3.4 |
| High-middle SDI | 3.1 (1.3–5.2) | 11.5 (4.6–20.2) | 0.31 (0.13–0.52) | 0.58 (0.23–1.03) | 8.7 | 2.8 | 87.7 (35.8–146.7) | 308.4 (123.6–549.6) | 8.47 (3.46–14.16) | 16.05 (6.44–28.59) | 8.1 | 2.9 |
| Middle SDI | 2.0 (0.9–3.3) | 12.8 (5.2–21.9) | 0.19 (0.08–0.31) | 0.47 (0.19–0.80) | 17.4 | 4.8 | 65.9 (27.5–105.3) | 371.9 (152.2–643.0) | 5.48 (2.28–8.80) | 13.09 (5.35–22.58) | 15 | 4.5 |
| Low-middle SDI | 1.3 (0.5–2.2) | 6.38 (2.7–10.3) | 0.21 (0.08–0.37) | 0.43 (0.18–0.71) | 12.6 | 3.4 | 37.3 (14.4–64.5) | 183.2 (78.6–298.1) | 5.51 (2.12–9.63) | 11.68 (5.02–19.05) | 12.6 | 3.6 |
| Low SDI | 0.4 (0.1–0.8) | 1.5 (0.6–2.7) | 0.18 (0.07–0.32) | 0.28 (0.11–0.49) | 8.9 | 1.8 | 13.2 (5.0–23.7) | 50.0 (18.5–87.3) | 5.02 (1.89–9.00) | 8.17 (3.03–14.20) | 9 | 2.0 |
| High income | ||||||||||||
| High-income Asia Pacific | 0.8 (0.4–1.4) | 2.0 (0.8–3.3) | 0.41 (0.18–0.66) | 0.44 (0.18–0.74) | 4.8 | 0.2 | 24.1 (10.5–39.7) | 41.3 (16.7–68.3) | 11.48 (5.01–18.84) | 10.72 (4.33–17.89) | 2.3 | -0.2 |
| Western Europe | 2.0 (0.8–3.5) | 6.2 (2.5–11.0) | 0.35 (0.14–0.6) | 0.67 (0.27–1.18) | 6.8 | 2.9 | 46.5 (18.6–82.0) | 130.2 (52.2–230.0) | 8.39 (3.36–14.83) | 15.68 (6.28–27.59) | 5.8 | 2.8 |
| Australasia | 0.05 (0.02–0.09) | 0.5 (0.2–0.8) | 0.21 (0.08–0.38) | 0.91 (0.38–1.54) | 29 | 10.8 | 1.3 (0.5–2.4) | 11.3 (4.7–18.8) | 5.79 (2.25–10.35) | 23.45 (9.83–38.98) | 24.8 | 9.8 |
| High-income North America | 0.9 (0.4–1.7) | 5.7 (2.5–9.4) | 0.27 (0.11–0.48) | 0.88 (0.38–1.44) | 17.2 | 7.3 | 23.9 (9.6–42.1) | 138.6 (60.4–224.9) | 7.28 (2.94–12.76) | 10.72 (4.33–17.89) | 15.4 | 1.5 |
| Southern Latin America | 0.05 (0.02–0.09) | 0.3 (0.1–0.5) | 0.1 (0.04–0.18) | 0.34 (0.14–0.57) | 16.1 | 7.8 | 1.3 (0.5–2.3) | 7.1 (2.9–11.9) | 2.71 (1.06–4.8) | 8.34 (3.41–13.99) | 14.3 | 6.7 |
| Central Europe, Eastern Europe, and Central Asia | ||||||||||||
| Central Europe | 0.7 (0.3–1.1) | 1.2 (0.5–2.0) | 0.44 (0.18–0.74) | 0.52 (0.21–0.91) | 2.3 | 0.6 | 17.2 (6.9–29.3) | 26.9 (10.9–47.2) | 11.25 (4.5–19.11) | 12.87 (5.22–22.57) | 1.8 | 0.5 |
| Eastern Europe | 0.6 (0.3–1.0) | 1.5 (0.6–2.5) | 0.22 (0.09–0.36) | 0.42 (0.17–0.72) | 4.8 | 2.9 | 16.7 (6.7–27.5) | 36.7 (14.9–61.8) | 5.85 (2.35–9.6) | 10.85 (4.4–18.23) | 3.9 | 2.8 |
| Central Asia | 0.4 (0.2–0.7) | 0.9 (0.4–1.6) | 0.89 (0.38–1.46) | 1.05 (0.44–1.88) | 4 | 0.6 | 12.2 (5.1–20.4) | 24.9 (10.3–44.7) | 24.68 (10.35–40.88) | 27.65 (11.45–49.51) | 3.3 | 0.4 |
| Latin America and Caribbean | ||||||||||||
| Tropical Latin America | 0.1 (0.06–0.3) | 0.8 (0.3–1.3) | 0.16 (0.06–0.28) | 0.3 (0.12–0.51) | 22.6 | 2.8 | 4.3 (1.7–7.7) | 20.2 (8.0–34.5) | 4.25 (1.72–7.59) | 7.68 (3.06–13.1) | 11.9 | 2.6 |
| Central Latin America | 0.3 (0.1–0.5) | 1.3 (0.6–2.2) | 0.33 (0.14–0.59) | 0.52 (0.22–0.89) | 10.8 | 1.9 | 7.5 (3.1–13.2) | 32.7 (13.9–55.1) | 8.51 (3.45–14.95) | 12.76 (5.42–21.51) | 10.8 | 1.6 |
| Andean Latin America | 0.05 (0.02–0.08) | 0.2 (0.09–0.4) | 0.22 (0.08–0.4) | 0.38 (0.16–0.68) | 9.7 | 2.3 | 1.4 (0.5–2.6) | 6.0 (2.5–10.6) | 6.11 (2.4–11.54) | 9.87 (4.12–17.42) | 10.6 | 1.9 |
| Caribbean | 0.04 (0.02–0.07) | 0.1 (0.06–0.2) | 0.17 (0.07–0.27) | 0.26 (0.11–0.44) | 4.8 | 1.7 | 1.2 (0.5–2.1) | 3.7 (1.5–6.4) | 4.58 (1.8–7.65) | 6.93 (2.87–11.87) | 6.7 | 1.7 |
| Southeast Asia, East Asia, and Oceania | ||||||||||||
| East Asia | 1.9 (0.8–3.1) | 13.2 (5.3–23.5) | 0.2 (0.08–0.33) | 0.61 (0.24–1.08) | 19.2 | 6.6 | 63.6 (26.3–103.7) | 390.6 (153.6–716.5) | 6.24 (2.59–10.2) | 18.16 (7.14–33.32) | 16.6 | 6.2 |
| Southeast Asia | 0.3 (0.1–0.6) | 2.1 (0.8–3.6) | 0.12 (0.05–0.19) | 0.3 (0.12–0.52) | 19.4 | 4.8 | 11.6 (4.9–18.9) | 65.1 (25.6–111.0) | 3.77 (1.59–6.12) | 8.78 (3.46–15.1) | 14.9 | 4.2 |
| Oceania | 0.01 (0.004–0.03) | 0.04 (0.01–0.07) | 0.38 (0.13–0.74) | 0.45 (0.18–0.8) | 9.7 | 0.6 | 0.4 (0.2–0.9) | 1.3 (0.5–2.3) | 11.61 (4.16–22.88) | 13.25 (5.21–24.16) | 7.2 | 0.5 |
| North Africa and Middle East | ||||||||||||
| North Africa and Middle East | 1.0 (0.4–2.0) | 5.1 (2.2–8.6) | 0.62 (0.22–1.16) | 1.12 (0.48–1.88) | 13.2 | 2.6 | 30.5 (11.1–56.0) | 146.1 (62.2–241.5) | 16.37 (5.89–30.39) | 29.06 (12.36–48.32) | 12.2 | 2.5 |
| South Asia | ||||||||||||
| South Asia | 0.2 (0.07–0.3) | 2.0 (0.8–3.3) | 0.03 (0.01–0.04) | 0.13 (0.05–0.21) | 29 | 10.8 | 5.6 (2.2–8.9) | 60.5 (23.9–101.1) | 0.83 (0.33–1.32) | 3.71 (1.46–6.18) | 31.6 | 11.2 |
| Sub-Saharan Africa | ||||||||||||
| Southern sub-Saharan Africa | 0.1 (0.05–0.3) | 0.7 (0.3–1.2) | 0.51 (0.19–0.9) | 1.26 (0.53–2.11) | 19.4 | 4.7 | 4.8 (1.7–8.2) | 22.4 (9.3–37.4) | 14.98 (5.46–25.84) | 34.16 (14.22–57.28) | 11.8 | 4.1 |
| Eastern sub-Saharan Africa | 0.1 (0.04–0.2) | 0.5 (0.2–1.0) | 0.13 (0.05–0.22) | 0.3 (0.1–0.53) | 12.9 | 4.2 | 3.3 (1.3–5.5) | 17.3 (6.0–30.2) | 3.79 (1.54–6.19) | 8.4 (2.87–14.81) | 13.7 | 3.9 |
| Central sub-Saharan Africa | 0.05 (0.01–0.1) | 1.3 (0.6–2.2) | 0.19 (0.06–0.45) | 0.41 (0.12–1.01) | 80.6 | 3.7 | 1.4 (0.4–3.3) | 7.4 (2.2–18.1) | 5.41 (1.57–12.38) | 11.16 (3.25–27.09) | 13.8 | 3.4 |
| Western sub-Saharan Africa | 0.4 (0.2–0.8) | 1.5 (0.6–2.5) | 0.48 (0.17–0.9) | 0.73 (0.29–1.22) | 8.9 | 1.7 | 13.6 (4.9–24.5) | 47.0 (18.4–80.4) | 13.57 (4.84–24.71) | 19.64 (7.72–32.91) | 7.9 | 1.4 |
DALY, disability-adjusted life years; GBD, Global Burden of Disease; SDI, sociodemographic index; UI, uncertainty interval.
Fig. 1.
Temporal trends in the Global Burden of Disease for high BMI-attributable liver cancer (1990–2021). Error bars indicate the 95% uncertainty interval for numbers. Shading indicates the 95% uncertainty interval for rates. DALY, disability-adjusted life years.
Burden of high BMI-attributable liver cancer stratification by sociodemographic index
In the country and territory analysis, the SDI demonstrated a nonlinear relationship with the ASMR and ASDR for high-BMI-attributable liver cancer burden. Analysis at the regional level indicated that, with the exception of the high-income Asia-Pacific region, the burden of disease significantly increased with rising SDI in all regions (P < 0.05). In 2021, among the 21 GBD regions: Southern sub-Saharan Africa exhibited the highest ASMR [1.26 (95% uncertainty interval: 0.53–2.11)] and ASDR [34.16 (95% uncertainty interval: 14.22–57.28)]; South Asia showed the lowest ASMR [0.13 (95% uncertainty interval: 0.05–0.21)] and ASDR [3.71 (95% uncertainty interval: 1.46–6.18)]. Country-level analysis revealed that the top three countries with the heaviest burden were Mongolia [ASMR: 9.54 (95% uncertainty interval: 3.81–16.52); ASDR: 243.77 (95% uncertainty interval: 95.89–422.27)], Tonga [ASMR: 5.53 (95% uncertainty interval: 2.37–9.48); ASDR: 170.45 (95% uncertainty interval: 71.45–298.29)], and Egypt [ASMR: 4.72 (95% uncertainty interval: 2.05–8.05); ASDR: 115.79 (95% uncertainty interval: 49.87–196.75)]; the three countries with the lightest burden were Timor-Leste [ASMR: 0.04 (95% uncertainty interval: 0.01–0.09); ASDR: 1.34 (95% uncertainty interval: 0.42–2.96)], Morocco [ASMR: 0.07 (95% uncertainty interval: 0.02–0.12); ASDR: 1.82 (95% uncertainty interval: 0.66–3.27)], and Bangladesh [ASMR: 0.06 (95% uncertainty interval: 0.02–0.12); ASDR: 1.96 (95% uncertainty interval: 0.70–3.51)]. Pearson correlation analysis confirmed a significant covarying relationship between SDI and ASDR (R2 = 0.28, P < 0.0001) (Fig. 2, Table 1, and Supplementary Figures S1, S2, and Table S1, Supplemental digital content 1, https://links.lww.com/EJGH/B208).
Fig. 2.
Standardized rates of mortality and disability-adjusted life years for high BMI-attributable liver cancer in different regions (a, b) and countries (c, d) by SDI, 1990–2021. The black line is the adaptive association based on all data points fitted with adaptive loess regression. Different symbols and colors in the figure indicate different countries and regions. ASDR, age-standardized DALYs rate; ASMR, age-standardized mortality rate; DALY, disability-adjusted life years; SDI, sociodemographic index.
Hierarchical clustering of the 21 GBD regions based on EAPC in ASMR and ASDR revealed distinct epidemiological trajectories. Between 1990 and 2021, the high-BMI-attributable liver cancer burden showed significant acceleration in ASMR and ASDR across four regions: Central Europe, Oceania, Central Asia, and the high-income Asia-Pacific region. Conversely, significant declines in ASMR and ASDR were observed in Central Latin America, Andean Latin America, Western sub-Saharan Africa, and the Caribbean (Fig. 3).
Fig. 3.
Temporal-spatial clustering patterns of EAPC in ASMR and ASDR for liver cancer attributable to high BMI, 1990–2021. ASDR, age-standardized disability-adjusted life years; ASMR, age-standardized mortality rate; EAPC, estimated annual percentage change.
Age patterns of high BMI-attributable liver cancer
Globally, mortality rates for high BMI-attributable liver cancer demonstrated progressive age-dependent escalation, peaking at 90–94 years in males and greater than or equal to 95 years in females. Regional stratification revealed high SDI regions exhibited elevated mortality across all age groups versus other development tiers. While male mortality trajectories mirrored global patterns in most SDI strata, low-SDI regions displayed unique sex divergence: male rates fell below female counterparts in post-55–59 age strata. Female mortality peaks demonstrated SDI-dependent transitions – high SDI regions aligned with global trajectories, whereas medium-high/medium SDI regions peaked at 90–94 years, followed by moderate declines, and medium-low/low-SDI regions peaked earlier at 75–79 years. Parallel analysis of DALY rates confirmed these sex-specific patterns, though maximum burden shifted to 65–69 years for both genders (Fig. 4 and Supplementary Table S2, Supplemental digital content 1, https://links.lww.com/EJGH/B208).
Fig. 4.
Age-stratified and sex-specific disparities in ASMR (a) and ASDR (b) of liver cancer attributable to high BMI across SDI regions. ASDR, age-standardized disability-adjusted life year rate; ASMR, age-standardized mortality rate; DALY, disability-adjusted life years; SDI, sociodemographic index.
Period patterns of high BMI-attributable liver cancer
From 1990 to 2021, ASMR and ASDR for high BMI-attributable liver cancer demonstrated sustained global increases across all GBD regions. High and medium-high SDI regions bear a greater burden of ASMR and ASDR compared with global averages. Strikingly, medium SDI regions demonstrated the most rapid acceleration of ASMR and ASDR, even surpassing those in high SDI regions. In contrast, low-SDI regions maintained the lowest burdens and exhibited the slowest growth trajectories (Fig. 5 and Supplementary Table S3, Supplemental digital content 1, https://links.lww.com/EJGH/B208).
Fig. 5.
ASMR (a) and ASDR (b) for liver cancer attributable to high BMI in different SDI regions, 1990–2019. ASDR, age-standardized disability-adjusted life year rate; ASMR, age-standardized mortality rate; SDI, sociodemographic index.
Future prediction of high BMI-attributable liver cancer death burden in different sociodemographic index
Projections indicate a persistent escalation in high BMI-attributable liver cancer burden, with global ASDR projected to reach 17.63/100 000 (95% uncertainty interval:16.83–18.44) by 2036 – a 24.5% increase from 2021 levels (annualized growth rate: 1.6%). Striking socioeconomic gradients persist: High SDI regions are projected to exhibit a peak ASDR of 19.81 per 100 000 (95% uncertainty interval: 16.81–22.81), while low-SDI regions demonstrate substantially lower burdens with an ASDR of 10.52 per 100 000 (95% uncertainty interval: 9.42–11.63). Pronounced gender disparities persisted, with males demonstrating a 2.1-fold higher ASDR burden than females (Fig. 6 and Supplementary Table S4, Supplemental digital content 1, https://links.lww.com/EJGH/B208).
Fig. 6.
Burden projections of DALYs for liver cancer attributable to high BMI across SDI strata, by sex: total (a), male (b), and female (c), 2021–2036. DALY, disability-adjusted life years; SDI, sociodemographic index.
Discussion
This study presents the first systematic evaluation of the global spatiotemporal evolution patterns in high BMI-related liver cancer burden from 1990 to 2036. Data reveals that in 2021, global DALYs attributable to high BMI-induced liver cancer increased by 944 600 cases compared with the 1990 baseline, with an ASDR rise of 7.19 per 100 000. The persistent male predominance aligns with gender-specific obesity trends: epidemiological surveillance indicates that by 2021, the age-standardized overweight/obesity rate had increased by 208.3% (95% uncertainty interval: 195.6–221.7) in males versus 113.9% (106.4–121.8) in females [13]. ARIMA models predict the ASDR will further rise to 17.63/100 000 (16.83–18.44) by 2036. This disease burden demonstrates a marked socioeconomic gradient, exhibiting multidimensional disparities across gender, age, period, and SDI levels.
The SDI-stratified burden of liver cancer mirrors global obesity dynamics, revealing critical prevention priorities. High-SDI regions exhibit the heaviest current and projected burden (2036年 ASDR: 12.3 per 100 000), driven by metabolic comorbidities compounded by industrial pollutants and pharmaceutical exposures [14–16]. Middle-SDI areas face dual challenges – rapid 4.7% annual ASMR growth alongside nutritional transition to ultraprocessed diets [17].
SDI-stratified burden reflects global obesity dynamics: High-SDI regions currently bear and are projected to sustain the heaviest burden, while middle-SDI areas exhibit rapid growth despite elevated baseline rates. Although low-SDI regions demonstrate lower baseline burdens, their persistently rising ASDR and ASMR warrant urgent attention. This distribution pattern closely aligns with the global obesity pandemic, as BMI escalation persists across all SDI quintiles, with low-middle and middle-high SDI regions projected to experience 28–35% increases from 2010 to 2036 [6,18]. This distribution is closely related to nutritional transition – westernized diets and sedentary lifestyles increase metabolic risk in high- and medium-SDI areas. Low-SDI regions face pressure from both urbanized dietary changes and inadequate chronic disease management. Notably, the decline in ASDR in high-income Asia-Pacific regions may be attributed to the preservation of traditional dietary patterns alongside active transportation systems [19]. Environmental coexposures further exacerbate this risk: traditional heavy industries are relocating to low- and medium-SDI regions, while emerging electronics and pharmaceutical industries are concentrated in high-SDI regions. This duality synergistically elevates liver cancer risk through genotoxic mechanisms, as well as through the interaction of metabolic disorders caused by industrial pollutants in high-SDI regions and occupational toxicants in transitioning economies. Therefore, it is essential to develop differentiated prevention and control strategies based on SDI stratification. In high SDI areas, the emphasis should be on the prevention and management of metabolic diseases: incorporating NAFLD into the basic public health service package and promoting noninvasive screening for liver fibrosis (e.g. fibrosis-4 index, elastography). In intermediate SDI areas, the focus should shift to risk transition interventions, such as implementing interventions to curb the proliferation of ‘Westernized’ dietary patterns (e.g. restricting the addition of trans-fatty acids and promoting the consumption of fiber-rich foods). In low-SDI areas, primary prevention should take precedence by expanding hepatitis B vaccination coverage to over 90% and promoting affordable and portable tools for self-measuring BMI, such as waist circumference measuring tapes.
Our analysis reveals persistent gender disparities in the global burden of high BMI-associated liver cancer from 1990 to 2036, with males demonstrating significantly higher ASMRs (0.89/100 000, 95% uncertainty interval: 0.84–0.93) and DALYs (24.0/100 000, 95% uncertainty interval: 22.7–25.2) – 1.9- and 2.1-fold greater than females, respectively. This trend is consistent with the distribution of the global burden of liver cancer observed over the past decade. Although there is no significant gender difference in the incidence of NAFLD-associated hepatocellular carcinoma [17], men remain the primary sufferers of the disease. This may be linked to their basal metabolic profiles: men have a significantly higher tendency to accumulate visceral fat compared with women, whose subcutaneous fat distribution in premenopausal stages is metabolically protective. The role of visceral fat, a key driver of metabolic disorders, is directly associated with insulin resistance and the progression of NAFLD [20–22]. In addition, men in the middle and high SDI regions are further affected by chronic exposure to occupational hazards, alcohol abuse, and low compliance with health screenings, all of which exacerbate the risk of liver cancer. Notably, gender differences in low-SDI regions reveal a specific pattern: liver cancer DALYs and mortality rates associated with high BMI are significantly higher in older women (≥55 years) compared with men. Mechanistic studies have demonstrated that the decline in estrogen levels following menopause results in a weakened metabolic compensatory capacity, which accelerates the progression from non alcoholic steatorrhoeic hepatitis to liver cancer. This issue is compounded by a lack of healthcare resources, including inadequate chronic disease management and insufficient monitoring of liver disease, as well as an imbalanced nutritional structure characterized by a high-calorie diet and decreased physical activity because of urbanization. This creates a ‘biological vulnerability–social environment risk’. The double whammy of biological vulnerability and socioenvironmental risk ultimately leads to a significant deterioration in the prognosis for this population [23,24]. Therefore, in low-SDI areas, it is essential to establish a metabolic syndrome screening system for postmenopausal women, including blood glucose and liver function monitoring, strengthen hepatitis B vaccination coverage, and improve dietary structure through community nutritional interventions. In contrast, in medium and high SDI areas, the focus should be on promoting the early detection of liver cancer in high-risk male groups through abdominal ultrasound and alpha-fetoprotein testing, while simultaneously implementing occupational health protections and addressing addictive behaviors, such as tobacco control and alcohol restriction.
Our findings suggest that the burden of liver cancer associated with high BMI increases with age. This conclusion is supported by previous studies that analyze this phenomenon as a long-term cumulative result of metabolic damage [25], chronic inflammation [4], immune senescence, and the synergistic effects of multiple risk factors [26]. In addition, long-term exposure to lifestyle and environmental factors, along with insufficient screening and treatment, may contribute to an increased risk among elderly individuals with obesity. Therefore, it is essential to enhance the management of metabolic syndrome (e.g. through NAFLD screening), optimize antiviral therapy for patients infected with HBV or HCV, and implement lifestyle interventions to reduce the risk of obesity-associated liver cancer in the elderly population.
Analyses based on GBD data deepen the understanding of BMI-driven liver cancer epidemiology, but limitations need to be noted: (a) observational data heterogeneity may lead to selection bias, especially in low-SDI areas with imperfect vital registration systems [6]; (b) diagnostic misclassification (e.g. cirrhosis versus liver cancer) and exposure misestimation in resource-constrained areas may affect the results; and (c) Bayesian interpolation assumptions (e.g. the unifying risk-exposure relationship) may lead to ecological fallacies that underestimate race-specific metabolic susceptibility; (d) although acknowledging the confounding effect of comorbidities (e.g. overlap between high BMI and diabetes/metabolic syndrome), the ecological nature of GBD data precludes comorbidity adjustment at the individual level, potentially compromising the precision of disease burden attribution. Future studies should integrate clinical cohort data and apply causal inference methods to isolate the independent effects of such comorbidities.
Conclusion
The disease burden of liver cancer associated with high BMI has exhibited a significant upward trend globally over the past three decades, and this trend is expected to continue. The data reveal substantial regional and demographic disparities in disease burden, particularly in high-income countries. The findings of this study underscore the necessity for optimizing the secondary prevention system for obesity-related liver disease and the urgent need to develop targeted intervention strategies based on robust evidence.
Acknowledgements
X.M.: Conceptualization, methodology, writing – original draft, writing – review and editing, visualization. T.P.: Conceptualization, methodology, writing – original draft, writing review and editing, visualization. N.G.: Software, formal analysis, writing – review and editing. S.Y.: Formal analysis, visualization. X.M.: Formal analysis and visualization. D.P.: Writing – review and editing. P.L.: Writing –review and editing.
Conflicts of interest
There are no conflicts of interest.
Supplementary Material
Footnotes
Xiaohua Ma and Ting Pan contributed equally to the writing of this article.
Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website, www.eurojgh.com.
References
- 1.Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024; 74:229–263. [DOI] [PubMed] [Google Scholar]
- 2.Guo Q, Zhu X, Beeraka NM, Zhao R, Li S, Li F, et al. Projected epidemiological trends and burden of liver cancer by 2040 based on GBD, CI5plus, and WHO data. Sci Rep 2024; 14:28131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Yang S, Deng Y, Zheng Y, Zhang J, He D, Dai Z, Guo C. Burden, trends, and predictions of liver cancer in China, Japan, and South Korea: analysis based on the Global Burden of Disease Study 2021. Hepatol Int 2025; 19:441–459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Huang DQ, Singal AG, Kono Y, Tan DJH, El-Serag HB, Loomba R. Changing global epidemiology of liver cancer from 2010 to 2019: NASH is the fastest growing cause of liver cancer. Cell Metab 2022; 34:969–977.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Harrison SA, Gawrieh S, Roberts K, Lisanti CJ, Schwope RB, Cebe KM, et al. Prospective evaluation of the prevalence of non-alcoholic fatty liver disease and steatohepatitis in a large middle-aged US cohort. J Hepatol 2021; 75:284–291. [DOI] [PubMed] [Google Scholar]
- 6.GBD 2021 Risk Factors Collaborators. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2024; 403:2162–2203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Campbell PT, Newton CC, Freedman ND, Koshiol J, Alavanja MC, Beane Freeman LE, et al. Body mass index, waist circumference, diabetes, and risk of liver cancer for U.S. adults. Cancer Res 2016; 76:6076–6083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dai H, Alsalhe TA, Chalghaf N, Riccò M, Bragazzi NL, Wu J. The global burden of disease attributable to high body mass index in 195 countries and territories, 1990-2017: an analysis of the Global Burden of Disease Study. PLoS Med 2020; 17:e1003198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tsilidis KK, Kasimis JC, Lopez DS, Ntzani EE, Ioannidis JP. Type 2 diabetes and cancer: umbrella review of meta-analyses of observational studies. BMJ (Clin Res ed). 2015; 350:g7607. [Google Scholar]
- 10.Li Y, Ou Z, Yu D, He H, Zheng L, Chen J, et al. The trends in death of primary liver cancer caused by specific etiologies worldwide: results from the Global Burden of Disease Study 2019 and implications for liver cancer management. BMC Cancer. 2023; 23:598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhou XD, Chen QF, Yang W, Zuluaga M, Targher G, Byrne CD, et al. Burden of disease attributable to high body mass index: an analysis of data from the Global Burden of Disease Study 2021. EClinicalMedicine 2024; 76:102848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Afshin A, Forouzanfar MH, Reitsma MB, Sur P, Estep K, Lee A, et al. ; GBD 2015 Obesity Collaborators. Health effects of overweight and obesity in 195 countries over 25 years. N Engl J Med 2017; 377:13–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.GBD 2021 Adult BMI Collaborators. Global, regional, and national prevalence of adult overweight and obesity, 1990-2021, with forecasts to 2050: a forecasting study for the Global Burden of Disease Study 2021. Lancet 2025; 405:813–838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Liu C, Zhu S, Zhang J, Wu P, Wang X, Du S, et al. Global, regional, and national burden of liver cancer due to non-alcoholic steatohepatitis, 1990-2019: a decomposition and age-period-cohort analysis. J Gastroenterol 2023; 58:1222–1236. [DOI] [PubMed] [Google Scholar]
- 15.Yang JD, Hainaut P, Gores GJ, Amadou A, Plymoth A, Roberts LR. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol 2019; 16:589–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.VoPham T, Jones RR. State of the science on outdoor air pollution exposure and liver cancer risk. Environ Adv 2023; 11:100354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tan DJH, Setiawan VW, Ng CH, Lim WH, Muthiah MD, Tan EX, et al. Global burden of liver cancer in males and females: changing etiological basis and the growing contribution of NASH. Hepatology 2023; 77:1150–1163. [DOI] [PubMed] [Google Scholar]
- 18.Swinburn BA, Sacks G, Hall KD, McPherson K, Finegood DT, Moodie ML, Gortmaker SL. The global obesity pandemic: shaped by global drivers and local environments. Lancet 2011; 378:804–814. [DOI] [PubMed] [Google Scholar]
- 19.Sung H, Siegel RL, Torre LA, Pearson-Stuttard J, Islami F, Fedewa SA, et al. Global patterns in excess body weight and the associated cancer burden. CA Cancer J Clin 2019; 69:88–112. [DOI] [PubMed] [Google Scholar]
- 20.Lemieux S, Prud’homme D, Bouchard C, Tremblay A, Després JP. Sex differences in the relation of visceral adipose tissue accumulation to total body fatness. Am J Clin Nutr 1993; 58:463–467. [DOI] [PubMed] [Google Scholar]
- 21.Tchernof A, Després JP. Pathophysiology of human visceral obesity: an update. Physiol Rev 2013; 93:359–404. [DOI] [PubMed] [Google Scholar]
- 22.Tchoukalova YD, Koutsari C, Karpyak MV, Votruba SB, Wendland E, Jensen MD. Subcutaneous adipocyte size and body fat distribution. Am J Clin Nutr 2008; 87:56–63. [DOI] [PubMed] [Google Scholar]
- 23.Xing QQ, Li JM, Chen ZJ, Lin X-Y, You Y-Y, Hong M-Z, et al. Global burden of common cancers attributable to metabolic risks from 1990 to 2019. Med 2023; 4:168–181.e3. [DOI] [PubMed] [Google Scholar]
- 24.Pang C, Li JM, Wang Z, Luo Y-C, Cheng Z-G, Han Z-Y, et al. Age-dependent female survival advantage in hepatocellular carcinoma: a multicenter cohort study. Clin Gastroenterol Hepatol 2024; 22:305–314. [DOI] [PubMed] [Google Scholar]
- 25.Jun BG, Kim M, Shin HS, Yi JJ, Yi SW. Impact of overweight and obesity on the risk of hepatocellular carcinoma: a prospective cohort study in 14.3 million Koreans. Br J Cancer 2022; 127:109–115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.GBD 2019 Cancer Risk Factors Collaborators. The global burden of cancer attributable to risk factors, 2010-19: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 2022; 400:563–591. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.