Abstract
IMPORTANCE: Metabolic dysfunction-associated steatotic liver disease (MASLD) is an increasingly significant cause of premature liver mortality (PLM) in the US; however, the proportion that is preventable and the upstream factors remain poorly understood. OBJECTIVE: To quantify the burden of preventable MASLD-PLM and identify county-level factors associated with its variation. DESIGN, SETTING, AND PARTICIPANTS: Ecological, cross-sectional study of publicly available mortality and county-level data from January 2011 to December 2020 for US counties with complete covariate information. MASLD-PLM was defined as liver-related deaths where MASLD was recorded as an underlying or contributing cause, among decedents aged 75 years or younger for males and 80 years or younger for females. Statistical analysis was performed from February 2024 to August 2025. MAIN OUTCOMES AND MEASURES: The primary outcome was the proportion of MASLD-PLM considered preventable, calculated by comparing county rates with the mean rate in the 3 states with the lowest mortality; to assess robustness, an alternative benchmark using the 10th percentile of the county-level MASLD-PLM distribution was also applied. Structural equation modeling estimated total, direct, and indirect associations of proportion of preventable MASLD-PLM with county-level factors. Associations are expressed as standardized β coefficients, representing the change in proportion of preventable MASLD-PLM (in SD units) per 1-SD change in each factor. RESULTS: In total, 2704 of 3143 counties with complete covariate information were included (county level composition: mean [SD] age, 41.17 [5.20] years; mean [SD] percentage female, 50.1% [2.3%]). Between 2011 and 2020, the mean annual MASLD-PLM was 18 345 deaths (8.22 per 100 000 population), with 8095 deaths (44.1%) to 10 636 deaths (58.0%) estimated as preventable. The proportion of MASLD-PLM distribution was highly left-skewed, with 2182 counties (80.7%) having a preventable proportion of 70% or greater. Higher proportion of preventable MASLD-PLM occurred more often in rural counties and was associated with higher rates of physical inactivity (β = 0.31; 95% CI, 0.28-0.35), obesity (β = 0.23; 95% CI, 0.18-0.27), preventable hospitalizations (β = 0.15; 95% CI, 0.11-0.18), and diabetes (β = 0.09; 95% CI, 0.04-0.14). These factors were associated with upstream conditions including lower access to exercise-friendly environments, unfavorable food environments, and higher uninsurance rates. Findings were consistent when using an alternative benchmark based on the 10th percentile of county-level mortality rates. CONCLUSION: In this ecological cross-sectional study of US counties, most MASLD-PLM was preventable, with the highest burdens in rural, socioeconomically disadvantaged, and medically underserved areas. Multiple upstream community-level factors were associated with proportion of preventable MASLD-PLM and may warrant further evaluation in future studies.