LiverRisk score (LRS) predicts long-term liver-related outcomes in a cohort of Veterans without evidence of liver disease

BACKGROUND AND AIMS: Metabolic dysfunction-associated steatotic liver disease (MASLD) is increasing and disproportionately affects Hispanics. The LiverRisk score (LRS), a non-invasive stratification tool, could enhance early identification of patients at risk for liver disease; however, it has not been validated in a large U.S. Veteran cohort with sufficient liver-related outcomes. APPROACH: We conducted a retrospective cohort study of Veterans Health Administration (VHA) patients from 1999 to 2023. Individuals with chronic liver disease, viral hepatitis, or missing LRS values were excluded using laboratory and ICD-10 data. Patients were classified into minimal (LRS <6), low (6 to <10), medium (10 to <15), and high-risk (≥15) categories. Outcomes included liver-related events, mortality, hepatocellular carcinoma (HCC), and C-statistics. Analyses used parametric/non-parametric methods and Cox proportional hazard models with competing risks, adjusting for missing data. RESULTS: Among 170,998 Veterans (93% male; median age 60; 63% White), 71%, 26%, 2%, and 0.5% were classified as minimal-risk, low-risk, medium-risk, and high-risk, respectively. Over 13.5 years median follow-up, 15,463 (9%) developed a liver-related event, 6219 (3.6%) died of liver causes, and 640 (0.5%) developed HCC. C-statistics were 0.61 for events, 0.70 for mortality, and 0.71 for HCC. Accuracy was highest among Hispanic/Latino Veterans (mortality 0.77, 95% CI 0.73-0.81; HCC 0.82, 95% CI 0.75-0.90) versus non-Hispanic Whites (mortality 0.62, 95% CI 0.61-0.62). CONCLUSIONS: LRS effectively predicted liver-related events, mortality, and HCC in Veterans. Its integration into clinical alerts and lab reports could support proactive patient care and liver disease prevention.