How this instrument works
FIB-4 multiplies age by AST, then divides by the platelet count times the square root of ALT. Every input is a routine blood test result, which is the entire appeal: no biopsy, no specialized scan, no cost beyond tests most people with liver disease already have on file. The score was built and validated in patients co-infected with HIV and hepatitis C, a population where deciding who truly needed an invasive biopsy mattered a great deal, and it has since spread into far broader everyday use.
Two thresholds do the real work of interpreting a result. Below 1.45, the score carries a negative predictive value around 90 percent for ruling out advanced scarring — a comfortably low-risk reading. Above 3.25, positive predictive value sits closer to 65 percent with specificity around 97 percent, flagging a meaningfully higher chance that advanced fibrosis is actually present. Everything between those two numbers is deliberately inconclusive: the original validation study built that middle band in on purpose, a zone where the blood numbers alone can't settle the question and further testing — elastography or biopsy — is what actually resolves it.
The honest caveat is where this score came from versus where it gets used today. It was derived and validated in HIV/hepatitis-C co-infected adults, not the general population, yet it's now applied widely to screen for scarring in fatty liver disease — NAFLD and its newer name, MASLD — where major liver-society guidance recommends it as a reasonable first pass despite that mismatch in original study population. That gap between original validation and current everyday use is worth knowing rather than glossing over; a tool built for one group can behave somewhat differently in another, even while remaining broadly useful.
- Gather four values from a recent blood panel: age, AST, platelet count, and ALT.
- Enter Age, AST (U/L), Platelet count (×10⁹/L), and ALT (U/L).
- Read FIB-4 index and compare it against the two thresholds: below 1.45 is lower risk, above 3.25 is higher risk, and the space between is indeterminate.
- Bring an indeterminate or elevated score to a clinician rather than acting on it alone — it's a screening flag, not a stand-alone diagnosis.
Worked example — age 35, AST 30, platelets 250, ALT 30
A 35-year-old with an AST of 30, a platelet count of 250, and an ALT of 30. The square root of 30 is about 5.48. Multiply that by the platelet count: 250 × 5.48 ≈ 1369.3. The numerator is age times AST: 35 × 30 = 1050. Divide the two: 1050 ÷ 1369.3 ≈ 0.77 — comfortably under the 1.45 threshold, consistent with a low probability of advanced scarring.
Change the inputs to a 60-year-old with a higher AST of 80, a lower platelet count of 120, and an ALT of 40, and the same steps give roughly 6.32 — well above the 3.25 mark. A middle case, a 50-year-old with an AST of 45, platelets of 180, and an ALT of 35, lands at about 2.11: inside the indeterminate band, where the score alone can't rule scarring in or out and further testing is what actually decides the question.
Questions
What do the FIB-4 cutoffs of 1.45 and 3.25 actually mean?
They split results into three zones. Below 1.45, the original validation study found a negative predictive value around 90 percent for excluding advanced fibrosis — a reassuring low-risk reading. Above 3.25, positive predictive value runs closer to 65 percent with specificity near 97 percent, meaning advanced scarring is meaningfully more likely, though still not certain. The space in between, 1.45 to 3.25, is intentionally indeterminate.
What should happen if my score falls in the indeterminate zone?
That middle band, roughly 1.45 to 3.25, means the blood numbers alone can't confirm or rule out advanced fibrosis. Clinicians typically follow an indeterminate score with a more direct test — transient elastography, a liver stiffness scan, or less often now a biopsy — rather than treating the number itself as an answer.
Was FIB-4 designed for the population I'm being screened in?
Possibly not, and that's worth knowing. It was developed and validated in patients with HIV and hepatitis C co-infection, a specific group with its own patterns of liver damage. It has since become a common first-line screen in fatty liver disease (NAFLD/MASLD) and other chronic liver conditions, and major liver-society guidance now supports that broader use — but performance validated in one population doesn't automatically transfer perfectly to every other group.
Why does age push the score up so much?
Because age sits directly in the numerator, multiplying straight into the result — an older patient with identical enzyme and platelet numbers to a younger one will always score higher. That's intentional: fibrosis risk genuinely climbs with age even at similar enzyme levels, and the formula was built to reflect that rather than treat every age the same.
Can a low platelet count alone drive an abnormal FIB-4?
Yes — platelets sit in the denominator, so a lower count pushes the whole score up even if both liver enzymes look unremarkable. A falling platelet count is itself a recognized sign of advancing liver disease, through a mechanism called portal hypertension, which is part of why the formula leans on it so heavily rather than using enzymes alone.
Is FIB-4 a diagnosis of liver fibrosis?
No — it's a screening calculation built from routine numbers, meant to sort people into lower- and higher-risk groups so further, more direct testing can be aimed where it matters most. A high score raises probability, it doesn't confirm scarring, and a low score lowers probability without fully ruling it out; a clinician interpreting it alongside the full clinical picture is what turns the number into an actual assessment.
References
- Sterling RK et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. 2006;43(6):1317-1325. PMID 16729309
- AASLD Practice Guidance — Noninvasive assessment of hepatic fibrosis and steatosis in NAFLD/MASLD
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.