The NAFLD fibrosis score turns six routine clinical and laboratory values into a single number that estimates the probability of advanced liver fibrosis in patients with nonalcoholic fatty liver disease. Developed by Angulo and colleagues in 2007, it remains one of the most widely used non-invasive tools for deciding which NAFLD patients need a specialist referral, elastography, or biopsy — and which do not.
How the score works
The NAFLD fibrosis score combines age, BMI, impaired fasting glucose or diabetes status, the AST/ALT ratio, platelet count, and albumin into a weighted regression: NFS = -1.675 + 0.037×age + 0.094×BMI + 1.13×(IFG/diabetes) + 0.99×(AST/ALT) − 0.013×platelets − 0.66×albumin.
The direction of each coefficient reflects biology: older age, higher BMI, diabetes, and a rising AST/ALT ratio all push the score up (toward more advanced fibrosis), while higher platelet count and higher albumin — both markers of preserved liver synthetic function and the absence of portal hypertension — pull the score down.
Interpreting the two cutoffs
Two cutoffs, both derived and validated in the original cohort, split the score into three zones. A score below -1.455 predicts the absence of advanced (F3-F4) fibrosis with a negative predictive value near 93% — most of these patients can avoid a biopsy. A score above 0.676 predicts the presence of advanced fibrosis with a positive predictive value near 90%, generally prompting further evaluation.
Scores that fall between the two cutoffs are indeterminate — roughly a quarter to a third of patients in the validation cohort landed here. An indeterminate result is not a calculator failure; it means the six variables alone cannot resolve the question, and a second-line test such as transient elastography (FibroScan) or liver biopsy is the appropriate next step.
Limits and how it is used
The NAFLD fibrosis score was derived and validated specifically in patients with biopsy-confirmed NAFLD — it is not intended for other causes of liver disease (viral hepatitis, alcohol-related liver disease, autoimmune hepatitis) and performance may be less reliable outside that population, in younger patients, or in those with normal BMI. It estimates a probability of advanced fibrosis, not a fibrosis stage, and does not replace a full clinical assessment, elastography, or biopsy when those are indicated. Like the score itself, this calculator is a screening aid, not a diagnostic test.