Calculate your NAFLD Fibrosis Score
Enter the values from your lab report, using the exact units shown.
Adults 18–110. The score was validated in adults.
BMI is computed as kg/m².
Aspartate aminotransferase (IU/L).
Alanine aminotransferase (IU/L).
Same as K/µL or ×10³/µL on most reports.
Pick the unit your lab uses ↓
44 g/L = 4.4 g/dL.
0.00
—
How we got there
Total NFS = —
Inputs used: —
Interpreting the score
Cutoffs from the original Angulo et al. (2007) study, validated in 733 biopsy-confirmed NAFLD patients.
< −1.455
Low probability of advanced fibrosis (F0–F2). Sensitivity 90%; negative predictive value 93% (derivation) / 88% (validation).
−1.455 to 0.675
Indeterminate. 20–58% of patients land here — the score cannot decide, so second-line assessment (e.g. elastography) is the honest next step.
> 0.675
High probability of advanced fibrosis (F3–F4). Specificity 97%, sensitivity 67%; positive predictive value 90% (derivation) / 82% (validation).
The formula
NFS = −1.675 + 0.037 × age (y) + 0.094 × BMI (kg/m²)
+ 1.13 × IFG/diabetes (yes=1, no=0) + 0.99 × (AST/ALT)
− 0.013 × platelets (×10⁹/L) − 0.66 × albumin (g/dL)
Worked examples
Tap one to load it into the calculator. All scores recomputed with this tool's engine.
Low — −1.62
Age 50, 85 kg / 174 cm (BMI 28.1), no diabetes, AST 45, ALT 55, platelets 200, albumin 40 g/L (= 4.0 g/dL) → −1.62 → low probability.
High — 3.46
Age 65, BMI 35, diabetes, AST 60, ALT 30, platelets 120, albumin 3.2 g/dL → 3.46 → high probability.
Indeterminate — 0.39
Age 55, BMI 28, diabetes, AST 45, ALT 45, platelets 160, albumin 4.0 g/dL → 0.39 → indeterminate.
Scope boundary: NFS vs FIB-4 vs APRI
This tool calculates the NAFLD Fibrosis Score (Angulo) only. The three scores are genuinely different calculations.
| NFS — Angulo (this tool) | FIB-4 — Sterling | APRI — Wai | |
|---|---|---|---|
| Inputs | Age, BMI, IFG/diabetes, AST/ALT ratio, platelets, albumin | Age, AST, ALT, platelets | AST, AST ULN, platelets |
| Formula | −1.675 + 0.037·age + 0.094·BMI + 1.13·IFG + 0.99·(AST/ALT) − 0.013·platelets − 0.66·albumin | (age × AST) / (platelets × √ALT) | ((AST/ULN) × 100) / platelets |
| Cutoffs | −1.455 / 0.675 | 1.30 / 2.67 (2.0 for age 65+) | 0.5 / 1.5 |
| Developed in | NAFLD (biopsy-confirmed) | HIV/HCV coinfection; widely validated incl. NAFLD | Chronic hepatitis C |
| Use together? | Yes — they are complementary. Discordant results (one low, one high) usually mean further testing is needed. | ||
Related calculators
- FIB-4 Calculator — the Sterling score: age, AST, ALT and platelets.
- APRI Calculator — the AST-to-platelet ratio index, a simpler two-variable screen.
Frequently asked questions
What is the NAFLD Fibrosis Score?
The NAFLD Fibrosis Score (NFS) is a noninvasive clinical score developed by Angulo et al. (Hepatology, 2007) to estimate the likelihood of advanced liver fibrosis in people with nonalcoholic fatty liver disease (NAFLD). It was derived and validated in 733 patients with biopsy-confirmed NAFLD, using six routine variables: age, BMI, impaired fasting glucose or diabetes, AST/ALT ratio, platelet count, and serum albumin. It is a screening aid, not a diagnosis.
What is the exact NFS formula, and which units does it need?
NFS = -1.675 + 0.037 x age (years) + 0.094 x BMI (kg/m^2) + 1.13 x impaired fasting glucose/diabetes (yes = 1, no = 0) + 0.99 x AST/ALT ratio - 0.013 x platelet count (x10^9/L) - 0.66 x albumin (g/dL). Units matter: platelets must be in x10^9/L (same as K/uL or x10^3/uL), and albumin must be in g/dL — if your lab reports albumin in g/L, divide by 10 before entering it (this calculator has a unit toggle that does it for you).
What do the NFS cutoffs mean?
A score below -1.455 means low probability of advanced fibrosis (stages F0-F2). A score from -1.455 to 0.675 is indeterminate — it neither rules advanced fibrosis in nor out. A score above 0.675 means high probability of advanced fibrosis (stages F3-F4). The boundary values -1.455 and 0.675 themselves fall in the indeterminate zone.
How accurate is the NAFLD Fibrosis Score?
In Angulo et al. (2007), the score had an area under the ROC curve of 0.88 in the derivation group and 0.82 in the validation group. The low cutoff (-1.455) excluded advanced fibrosis with 90% sensitivity and a negative predictive value of 93% (derivation) and 88% (validation). The high cutoff (0.675) detected advanced fibrosis with 97% specificity, 67% sensitivity, and a positive predictive value of 90% (derivation) and 82% (validation). Applying the model would have avoided liver biopsy in 549 of 733 patients (75%), with correct prediction in 496 of those (90%).
My result is indeterminate — what should I do?
An indeterminate result is common and honest: studies report that 20% to 58% of tested patients fall between the two cutoffs. It means the blood-based score cannot decide, so the usual next step is a second-line assessment such as liver stiffness measurement (elastography) plus referral to a hepatologist or liver clinic. Do not try to nudge the inputs to get a 'better' answer — discuss the result with your clinician.
How is the NAFLD Fibrosis Score different from FIB-4?
They are genuinely different scores, not the same calculation. NFS (Angulo) uses six variables — age, BMI, IFG/diabetes, AST/ALT ratio, platelets, albumin — with cutoffs -1.455 and 0.675. FIB-4 (Sterling) uses four variables — age, AST, ALT, platelets — computed as (age x AST) / (platelets x sqrt(ALT)), with classic cutoffs 1.30 and 2.67 (often 2.0 for age 65+). Both screen for advanced fibrosis and can be used together; discordant results usually mean further testing is needed.
How is it different from APRI?
APRI (AST to Platelet Ratio Index) is a much simpler score: ((AST / AST upper limit of normal) x 100) / platelet count, with classic cutoffs 0.5 and 1.5. It was developed mainly for chronic hepatitis C and is not NAFLD-specific. NFS was developed specifically in NAFLD patients and incorporates metabolic factors (BMI, glucose/diabetes, albumin) that APRI ignores.
My lab reports albumin in g/L. What do I enter?
Divide the g/L value by 10 to get g/dL — for example, 44 g/L equals 4.4 g/dL. This is the most common unit mistake with this score, and it changes the result by a factor of 10 on the albumin term (0.66 x albumin), which can easily flip the category. This calculator includes a g/dL / g/L toggle so you can enter the value exactly as your lab reports it.
Does diabetes really change the score?
Yes — substantially. Impaired fasting glucose or diabetes contributes a full 1.13 points when present (yes = 1) versus 0 when absent, making it one of the strongest single terms in the formula. That reflects the underlying biology: insulin resistance and diabetes are strong independent predictors of advanced fibrosis in NAFLD, which is why the original study selected the variable.
Can I use this score without a NAFLD diagnosis, and who should not rely on it?
The score was developed and validated in people with NAFLD confirmed by liver biopsy, so it is most meaningful in that context; in unselected populations the predictive values differ. It was validated in adults, is not designed for other liver diseases (viral hepatitis, alcohol-related disease, autoimmune liver disease), and never replaces clinical judgment. It is a screening aid — not medical advice. Discuss your result with your doctor or a hepatologist, and seek prompt care if you have symptoms of liver disease.
Can you show a worked example of the calculation?
Take a 50-year-old with BMI 28, no diabetes, AST 45 U/L, ALT 55 U/L, platelets 200 x10^9/L, albumin 4.0 g/dL. NFS = -1.675 + (0.037 x 50) + (0.094 x 28) + (1.13 x 0) + (0.99 x 45/55) - (0.013 x 200) - (0.66 x 4.0) = -1.675 + 1.850 + 2.632 + 0 + 0.810 - 2.600 - 2.640 = -1.62, which is below -1.455, so low probability of advanced fibrosis.