TOPLINE

Of nine risk scores tested in veterans with steatotic liver disease (SLD), the Steatosis-Associated Fibrosis Estimator (SAFE) score was the most useful for predicting cirrhosis within 10 years, though its edge was modest. No score reliably identified which patients without cirrhosis should be screened for hepatocellular carcinoma (HCC).

METHODOLOGY

• The American Association for the Study of Liver Diseases recommends the Fibrosis-4 Index (FIB-4) to screen for cirrhosis in SLD, but it can miss at-risk patients and flag many low-risk ones for extra testing, so researchers compared it with eight other risk scores for guiding repeat cirrhosis checks and HCC screening.
• Drawing on data from a nationwide US veterans health system, they analyzed 853,131 patients (median age, 61 years; 92.7% men) with imaging-confirmed SLD between 2008 and 2020; those with viral hepatitis, other primary liver disease, or cirrhosis, hepatic decompensation, or HCC diagnosed before or shortly after steatosis were excluded. 
• The following nine risk scores were calculated from clinical data around the time of the first imaging showing steatosis: FIB-4; aspartate aminotransferase-to-platelet ratio (APRI); albumin-bilirubin; age-male-albumin-bilirubin-platelet; BMI, age, aspartate aminotransferase-to-alanine aminotransferase ratio, and diabetes; Nonalcoholic Fatty Liver Disease Fibrosis Score; SAFE; Sinn; and Tate (the last two being liver cancer scores).
• Primary outcomes were first diagnoses of cirrhosis or HCC, identified through diagnostic codes in the healthcare records, plus a cancer registry for HCC. 
• Patients were followed for a median duration of 8.5 years.

TAKEAWAY

• Over 10 years, 33,794 patients (3.96%) developed cirrhosis and 2978 (0.35%) developed HCC.
• FIB-4, APRI, and SAFE scores performed about equally well at distinguishing patients who developed cirrhosis over 10 years from those who did not. SAFE, Tate, and FIB-4 performed best for HCC.
• SAFE had the largest net benefit for cirrhosis. At a score of 29.5, which translated to a 2.5% 10-year risk, it identified one additional case for every 53 patients flagged as at risk, a small edge over reassessing everyone with SLD. 
• For HCC, the gain was minimal. In patients without cirrhosis, a SAFE score of 43.8 translated to a 0.25% 10-year risk and yielded about 1.6 additional cases per 1000 patients flagged as at risk.

IN PRACTICE

"[D]istinguishing patients at high vs low risk of incident cirrhosis may inform the intensity of risk monitoring and reduce both invasive testing and premature referral to hepatology. Patients at low risk for incident cirrhosis may need less frequent monitoring, which can be managed by their primary care physician, than current recommendations and may not need an early hepatology referral," Melinda Wang, MD, MHS, wrote in an accompanying invited commentary.

SOURCE

The study was led by Catherine Mezzacappa, MD, PhD, MPH of the Department of Internal Medicine at Yale School of Medicine in New Haven, Connecticut. It was published online on September 21, 2026, in JAMA Internal Medicine.

LIMITATIONS

Missing data precluded assessment of six published risk scores. The study may have overestimated the predictive performance of the Tate score because it was developed using data from the same health system. Reliance on imaging report text to identify steatosis may have introduced selection bias.

DISCLOSURES

The study was supported by grants from the US National Institutes of Health, including the National Institute of Diabetes and Digestive and Kidney Diseases, National Cancer Institute, and National Institute on Alcohol Abuse and Alcoholism, with in-kind support from the US Department of Veterans Affairs. The authors reported no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.