Work

NAFLD/MASLD, including cirrhosis: Burden from 1990 to 2040, GBD decomposition, CDC WONDER analysis, and machine-learning attribution

Year: 2026

Type: article

Abstract: Non-alcoholic fatty liver disease (NAFLD)/metabolic dysfunction-associated steatotic liver disease (MASLD), including cirrhosis, is an increasing global health concern, yet long-term trends, drivers, and subnational disparities remain inadequately characterized. We assessed the global burden and U.S. subnational variations using machine-learning-based attribution. Using Global Burden of Disease 2021 data, we analyzed global incidence, prevalence, disability-adjusted life years (DALYs), temporal trends, decomposition, inequalities, and Bayesian age-period-cohort projections through 2040. For complementary U.S. analyses, we used CDC WONDER data, extreme gradient boosting (XGBoost)–SHapley Additive exPlanations attribution, and autoregressive integrated moving average/Prophet forecasting. Globally, incident cases, prevalence, and DALYs approximately doubled from 1990 to 2021. Age-standardized incidence and prevalence increased, whereas DALY rates remained stable (estimated annual percentage change = 0.04), with population growth and aging identified as the main drivers. Absolute inequality, measured by the slope index of inequality (SII), widened from 20 to 36 per 100,000. High-burden regions included North Africa and the Middle East, Latin America, and Eastern Europe. In the U.S., the South and Midwest had the highest and fastest-growing mortality rates. XGBoost identified diabetes prevalence as the dominant state-level predictor, showing a nonlinear positive association above approximately 10–12%, whereas higher educational attainment showed an inverse association above approximately 90%. A sharp inflection in mortality occurred in 2020. Mortality is projected to increase through 2035, particularly in the Southwest and Appalachia. The absolute burden of NAFLD/MASLD, including cirrhosis, continues to rise despite stable age-standardized DALY rates. Global trends and U.S. machine-learning evidence indicate that prevention should shift toward systemic metabolic risk management, with liver fibrosis screening integrated into routine diabetes and obesity care.

Source: BMC Gastroenterology

DOI: 10.1186/s12876-026-05442-9