About NAFLD Fibrosis Calculator
7 min read
NAFLD Fibrosis Calculator: Predict Advanced Fibrosis Risk in Nonalcoholic Fatty Liver Disease
TL;DR: Enter your age, BMI, diabetes status, AST, ALT, platelet count, and albumin to get your NAFLD Fibrosis Score (NFS). A score below -1.455 rules out advanced fibrosis with 93% negative predictive value; above 0.676 confirms high probability with 90% positive predictive value. Scores between those thresholds fall into an indeterminate zone that typically requires further evaluation such as elastography or biopsy. This calculator runs the validated Angulo et al. (2007) formula in seconds, replacing manual arithmetic with instant risk stratification.
Table of Contents
- What Your Liver Blood Tests Are Not Telling You
- Six Scenarios Where the NFS Changes Clinical Decisions
- The NFS Formula and Scoring Thresholds
- From Lab Report to Fibrosis Score in Seven Steps
- See How the Numbers Play Out: Two Worked Examples
- Where People Go Wrong With the NAFLD Fibrosis Score
- FAQ
- Assumptions and Notes
- Your Next Step
- Further Reading
What Your Liver Blood Tests Are Not Telling You
A routine liver panel showing mildly elevated ALT does not tell you whether scar tissue is forming. Nonalcoholic fatty liver disease affects roughly 25% of the global adult population, and most of those people will never progress beyond simple steatosis. But between 7% and 30% of NAFLD patients develop nonalcoholic steatohepatitis (NASH), and a subset of those progress to advanced fibrosis, cirrhosis, and liver failure. The clinical challenge is separating the majority who face minimal risk from the minority who need aggressive monitoring.
The NAFLD Fibrosis Score, published by Angulo et al. in 2007 in the journal Hepatology, was designed to solve exactly this problem without a liver biopsy. It combines six routine clinical variables into a single number that stratifies patients into low, indeterminate, or high probability of advanced fibrosis (stage F3-F4). The original validation study included 733 patients with biopsy-confirmed NAFLD across multiple centres.
Genetic variation in the PNPLA3 gene (rs738409 variant) significantly affects individual susceptibility to fibrosis progression. Carriers of the G allele develop more severe steatosis and fibrosis at identical BMI and metabolic profiles compared to non-carriers. The NFS does not account for this variant directly, but it captures downstream metabolic effects through BMI, diabetes status, and liver enzyme ratios.
The calculator above runs this formula in about ten seconds.
Six Scenarios Where the NFS Changes Clinical Decisions
-
Your ultrasound shows hepatic steatosis but liver enzymes are only mildly elevated. An ALT of 50 U/L and AST of 40 U/L can exist in both simple fatty liver and early fibrosis. Roughly 20% of patients with normal or near-normal transaminases already have significant fibrosis on biopsy. The NFS adds the missing context that enzyme levels alone cannot provide.
-
You have type 2 diabetes and a BMI above 30. Diabetic patients with obesity carry a fibrosis prevalence of 15 to 20%, compared with 7% in the general NAFLD population. Running the NFS at diagnosis and annually thereafter flags progression before clinical symptoms appear, potentially saving years of undetected scarring.
-
You are a primary care physician deciding whether to refer a patient to hepatology. Liver biopsy costs between $1,500 and $3,000 in most health systems and carries a 0.5% complication rate. A low NFS (below -1.455) can prevent unnecessary referrals in about 75% of NAFLD patients, reserving specialist resources for indeterminate and high-risk scores.
-
You are monitoring metabolic syndrome and want a liver-specific risk marker. Metabolic syndrome affects roughly 35% of US adults, and NAFLD overlaps with it in 42 to 69% of cases. Adding the NFS to your metabolic panel every 12 months provides a concrete fibrosis trajectory that waist circumference and fasting glucose do not capture on their own.
-
You have lost more than 10% of your body weight through lifestyle changes and want to confirm liver benefit. A 10% weight reduction decreases the NFS by an average of 0.3 to 0.8 points in patients with baseline scores in the indeterminate range. Recalculating after weight loss quantifies the hepatic benefit of your intervention rather than relying solely on ultrasound resolution of steatosis.
-
You are a patient with NAFLD who wants to understand your biopsy risk before an appointment. Walking into a gastroenterology consultation with a pre-calculated NFS creates a more productive conversation. If your score is -2.1, you and your clinician can discuss surveillance frequency rather than biopsy scheduling; if it is 0.9, you can prepare for the likelihood of further invasive workup.
The NFS Formula and Scoring Thresholds
The score combines age, body mass index, the presence or absence of impaired fasting glucose or diabetes, the AST-to-ALT ratio, platelet count, and serum albumin into a single continuous value.
NFS = -1.675
+ 0.037 x Age (years)
+ 0.094 x BMI (kg/m²)
+ 1.13 x IFG_or_Diabetes (0 = No, 1 = Yes)
+ 0.99 x AST/ALT ratio
- 0.013 x Platelets (x10⁹/L)
- 0.66 x Albumin (g/dL)
NFS Interpretation Thresholds
| Score Range | Classification | Clinical Meaning | Predictive Value |
|---|---|---|---|
| Below -1.455 | Low probability | Advanced fibrosis unlikely | NPV 93% |
| -1.455 to 0.676 | Indeterminate | Cannot rule in or out | Requires further testing |
| Above 0.676 | High probability | Advanced fibrosis likely | PPV 90% |
How Each Variable Pushes the Score
| Variable | Direction | Reasoning |
|---|---|---|
| Age | Higher age raises score | Fibrosis accumulates over time with chronic liver injury |
| BMI | Higher BMI raises score | Obesity drives steatohepatitis and fibrotic progression |
| Diabetes/IFG | Present raises score by 1.13 | Insulin resistance accelerates hepatic stellate cell activation |
| AST/ALT ratio | Higher ratio raises score | Reversal of the normal ALT-dominant pattern suggests cirrhotic remodelling |
| Platelets | Higher count lowers score | Thrombocytopenia signals portal hypertension from advanced fibrosis |
| Albumin | Higher level lowers score | Low albumin reflects impaired hepatic synthetic function |
Limitations: The NFS was validated in adult populations (age 18+). It has reduced accuracy in patients under 35 and in those with morbid obesity (BMI above 45), where the BMI coefficient can disproportionately inflate the score. Patients on antiplatelet therapy or with haematological conditions may have platelet counts that do not reflect hepatic portal pressure.
From Lab Report to Fibrosis Score in Seven Steps
-
Collect your lab values from a recent blood panel. You need AST (U/L), ALT (U/L), platelet count (x10^9/L), and albumin (g/dL). All four appear on a standard metabolic panel and complete blood count. Results should be from the same draw or within 2 weeks of each other.
-
Record your age in years. Use your current age, not age at diagnosis. The formula weights each year at 0.037 points, so a 10-year age difference shifts the score by 0.37.
-
Measure or confirm your BMI. Divide your weight in kilograms by the square of your height in metres. A BMI of 30 adds 2.82 to the formula; a BMI of 25 adds 2.35. The difference between these two (0.47 points) can shift a borderline score across a threshold.
-
Determine your diabetes or impaired fasting glucose status. If you have a diagnosis of type 2 diabetes or a fasting glucose above 100 mg/dL on two occasions, enter 1. Otherwise enter 0. This single binary variable adds 1.13 points when present.
-
Calculate your AST-to-ALT ratio. Divide AST by ALT. In healthy livers, ALT typically exceeds AST, producing a ratio below 1.0. As fibrosis progresses, AST rises relative to ALT, pushing the ratio toward and above 1.0.
-
Enter all values into the calculator. The formula applies positive and negative coefficients simultaneously. You do not need to memorise which variables add and which subtract.
-
Compare your result against the three threshold zones. A score below -1.455 provides strong reassurance. A score above 0.676 warrants prompt specialist evaluation. The indeterminate zone between them typically leads to transient elastography (FibroScan) as the next diagnostic step.
Non-obvious insight: The AST/ALT ratio carries nearly as much weight as the diabetes variable (0.99 coefficient vs. 1.13). A patient whose AST/ALT ratio shifts from 0.7 to 1.1 over 12 months sees a 0.40-point score increase from that single change, which can move an indeterminate result into the high-probability zone.
See How the Numbers Play Out: Two Worked Examples
Example 1: Retired Teacher With Metabolic Syndrome, Age 62
Margaret is 62, has a BMI of 32, was diagnosed with type 2 diabetes 4 years ago, and her latest labs show AST 42 U/L, ALT 38 U/L, platelets 180 x10^9/L, and albumin 3.8 g/dL.
AST/ALT ratio = 42 / 38 = 1.105
NFS = -1.675
+ 0.037 x 62 = +2.294
+ 0.094 x 32 = +3.008
+ 1.13 x 1 = +1.130
+ 0.99 x 1.105 = +1.094
- 0.013 x 180 = -2.340
- 0.66 x 3.8 = -2.508
NFS = -1.675 + 2.294 + 3.008 + 1.130 + 1.094 - 2.340 - 2.508
NFS = 1.003
| Variable | Value | Contribution |
|---|---|---|
| Constant | — | -1.675 |
| Age (62) | +2.294 | Raises score |
| BMI (32) | +3.008 | Raises score |
| Diabetes | +1.130 | Raises score |
| AST/ALT (1.105) | +1.094 | Raises score |
| Platelets (180) | -2.340 | Lowers score |
| Albumin (3.8) | -2.508 | Lowers score |
| NFS | 1.003 | High probability |
Margaret's score of 1.003 exceeds the 0.676 threshold, placing her in the high-probability category. Her clinician should arrange transient elastography or consider referral for liver biopsy. The dominant contributors are her age, BMI, and diabetes status. If she achieves a 10% weight loss (BMI dropping to roughly 28.8), her BMI contribution alone would decrease by 0.30 points.
Example 2: Shift Worker With Incidental Steatosis on Ultrasound, Age 38
Darren is 38, has a BMI of 27, no diabetes or impaired fasting glucose, and his labs show AST 30 U/L, ALT 52 U/L, platelets 240 x10^9/L, and albumin 4.3 g/dL. His ultrasound flagged mild fatty liver during an unrelated abdominal scan.
AST/ALT ratio = 30 / 52 = 0.577
NFS = -1.675
+ 0.037 x 38 = +1.406
+ 0.094 x 27 = +2.538
+ 1.13 x 0 = +0.000
+ 0.99 x 0.577 = +0.571
- 0.013 x 240 = -3.120
- 0.66 x 4.3 = -2.838
NFS = -1.675 + 1.406 + 2.538 + 0.000 + 0.571 - 3.120 - 2.838
NFS = -3.118
| Variable | Value | Contribution |
|---|---|---|
| Constant | — | -1.675 |
| Age (38) | +1.406 | Raises score |
| BMI (27) | +2.538 | Raises score |
| Diabetes | +0.000 | No effect |
| AST/ALT (0.577) | +0.571 | Mild raise |
| Platelets (240) | -3.120 | Lowers score |
| Albumin (4.3) | -2.838 | Lowers score |
| NFS | -3.118 | Low probability |
Darren's score of -3.118 sits well below the -1.455 low-probability threshold. Advanced fibrosis is unlikely (93% NPV). His clinician can recommend lifestyle modification for the steatosis, including dietary changes targeting a 5 to 7% weight reduction, and recheck the score annually rather than pursuing invasive workup. His healthy platelet count and albumin level are the strongest protective factors in the formula.
Where People Go Wrong With the NAFLD Fibrosis Score
Using non-fasting blood draws for AST and ALT. While transaminases are not as meal-sensitive as glucose or triglycerides, vigorous exercise within 48 hours of the draw can raise AST by 20 to 50% because AST is also released from damaged muscle tissue. Schedule blood draws after at least 48 hours without intense physical activity. A falsely raised AST inflates the AST/ALT ratio and pushes the score upward.
Entering albumin in g/L instead of g/dL. Many European and Australian labs report albumin in g/L (normal range 35 to 50), while the NFS formula requires g/dL (normal range 3.5 to 5.0). Entering 40 instead of 4.0 produces an albumin contribution of -26.4 instead of -2.64, generating a wildly negative and meaningless score. Check the unit on your lab report before entering the value.
Ignoring the indeterminate zone and treating it as "probably fine." Roughly 30% of patients score between -1.455 and 0.676. This zone does not mean low risk. It means the formula cannot classify you, and further testing (typically FibroScan or the FIB-4 index as a second-line score) is needed. Treating indeterminate as reassuring delays diagnosis in a meaningful fraction of patients who do have advanced fibrosis.
Forgetting to update diabetes status after a new diagnosis. The diabetes/IFG variable adds 1.13 points, the second-largest single contributor after BMI. A patient who was non-diabetic at first calculation but receives a diabetes diagnosis 2 years later needs to recalculate. The previous "low probability" result may no longer be accurate with the added 1.13 points.
Recalculating too frequently with minor lab fluctuations. AST and ALT vary by 10 to 15% between draws even when liver status has not changed, due to diurnal variation, hydration, and assay variability. Recalculating the NFS monthly and reacting to small score changes creates false trend signals. Recalculate every 6 to 12 months, or after a significant clinical change such as major weight loss or new diabetes diagnosis.
Applying the score to patients with other liver diseases. The NFS was validated only for nonalcoholic fatty liver disease. Patients with hepatitis B or C, autoimmune hepatitis, alcoholic liver disease, or hemochromatosis have different fibrosis mechanisms. Using the NFS in these populations produces unreliable classifications because the relationship between the six input variables and fibrosis stage differs from the NAFLD-specific model.
Assumptions and Notes
- Margin of error: The NFS thresholds (-1.455 and 0.676) were derived from the Angulo et al. (2007) cohort of 733 biopsy-confirmed NAFLD patients. Lab values carry inherent variability of 10 to 15% between draws, which propagates into score fluctuations of approximately 0.2 to 0.4 points. Scores within 0.3 points of either threshold should be interpreted cautiously and confirmed with repeat testing or complementary tools such as FIB-4 or elastography. BMI accuracy depends on correct height and weight measurement; a 2 cm height error changes BMI by approximately 0.5 points.
- Professional disclaimer: The NAFLD Fibrosis Score from this calculator is for informational and educational purposes only and does not constitute medical advice or a clinical diagnosis. Fibrosis assessment requires interpretation by a qualified healthcare provider in the context of complete clinical history, imaging, and potentially liver biopsy. Do not make treatment decisions based solely on this score. Patients with known liver disease of any aetiology should consult their hepatologist or gastroenterologist before acting on calculator results.
Your Next Step
Margaret's score told her that the steatosis on her ultrasound was not the whole story. Darren's told him the opposite: that the incidental finding on his scan did not warrant alarm. Both answers came from the same six lab values and the same 30-year-old formula.
Enter your numbers into the calculator above and bring the result to your next clinic visit.