Why Ferritin Cutoffs for Iron Deficiency Keep Changing

Ferritin cutoffs for iron deficiency keep changing because a single number cannot serve every population at once. A lower threshold catches fewer true cases but rarely flags healthy people; a higher threshold catches more real deficiency but mislabels some iron-replete patients. Guideline panels weigh that sensitivity-versus-specificity trade differently depending on who is being tested and how much inflammation distorts the result, so the same biomarker ends up with several defensible cutoffs rather than one.

Ferritin is a good test measured against a hard truth

Serum ferritin correlates with total body iron stores better than almost any other routine blood test, which is why it anchors iron-deficiency diagnosis. Its accuracy is usually benchmarked against bone marrow iron staining, the reference standard, because that directly visualizes stored iron. The problem is that the relationship between a ferritin number and marrow iron is a sliding scale, not a switch. Wherever a panel draws the line, some patients on either side are misclassified. The choice of line is therefore a value judgment about which error is worse: missing real deficiency, or treating people who do not need it.

That judgment is quantified by two numbers. Sensitivity is the share of truly iron-deficient people the cutoff correctly flags. Specificity is the share of iron-replete people it correctly clears. Raising the cutoff pushes sensitivity up and specificity down. Lowering it does the reverse. No single value maximizes both, and that tension is the whole reason the thresholds keep moving.

The AGA case for a higher line

The clearest illustration comes from the American Gastroenterological Association's 2020 clinical practice guideline on the gastrointestinal evaluation of iron-deficiency anemia, led by Cynthia Ko and colleagues in Gastroenterology. Its accompanying technical review pooled dozens of diagnostic studies that used bone marrow iron as the reference and reported that the long-standing cutoff of ferritin below 15 ng/mL had a sensitivity of only about 59 percent against a specificity near 99 percent. In plain terms, that threshold almost never falsely labels a replete person, but it misses roughly four in ten people who genuinely lack iron.

Moving the cutoff to below 45 ng/mL changed the balance: sensitivity rose to about 85 percent while specificity fell to about 92 percent. The AGA concluded that in a patient who already has anemia, defining iron deficiency at ferritin below 45 ng/mL better serves the clinical goal, because failing to identify a treatable, sometimes serious cause of anemia carries more weight than the cost of some additional confirmatory workup. The number moved not because the biology changed, but because the panel reweighted the two errors for a specific population: anemic adults being evaluated for a gastrointestinal source.

Menstruation, pregnancy, and the shifting baseline

A cutoff calibrated for anemic adults does not automatically fit a healthy menstruating person whose iron stores turn over with every cycle. Recognizing this, the American Society of Hematology released draft recommendations for diagnosing iron deficiency, open for public comment through late October 2025 and still in draft form as of this writing. For menstruating individuals, the draft proposes a serum ferritin threshold of 20 ng/mL or below, rather than the older 12 ng/mL. For pregnant individuals, it suggests a threshold of about 30 ng/mL and advises against using a threshold as low as 15 ng/mL.

The logic tracks the same trade-off. Menstruating and pregnant people face higher iron demand and a higher pretest probability of true deficiency, so a panel can accept lower specificity to gain the sensitivity that catches early depletion before anemia sets in. Because these are draft proposals under active comment, they are best read as a direction of travel toward population-specific cutoffs, not settled numbers.

Inflammation breaks the assumption

The hardest case is inflammation, and it is where thresholds diverge most. Ferritin is an acute-phase reactant, meaning it rises during infection, chronic inflammatory disease, malignancy, and other stressed states independent of iron stores. The AGA technical review flagged exactly this: a patient with inflammatory bowel disease can have a falsely reassuring ferritin while genuinely iron-deficient. A cutoff that works in a healthy person becomes misleadingly low in an inflamed one.

The World Health Organization addressed this directly in its 2020 guideline on using ferritin to assess iron status. For apparently healthy individuals, WHO retained a threshold below 15 micrograms per liter. But in the presence of infection or inflammation, it advised raising the cutoff that defines deficiency to below 70 micrograms per liter in adults and below 30 micrograms per liter in children, and measuring inflammatory markers alongside ferritin so the value can be interpreted in context. Same analyte, same person, a very different line, chosen so that inflammation does not hide real deficiency.

Why the numbers will keep moving

Three forces keep ferritin cutoffs in motion. The first is population: pretest probability of deficiency differs between an anemic gastroenterology patient, a menstruating adult, and a child, so the optimal balance point differs too. The second is the competing cost of errors: when missed deficiency is cheap to confirm and expensive to overlook, panels favor sensitivity and raise the line. The third is inflammation, which corrupts the biomarker itself and forces context-dependent adjustment rather than a fixed value.

None of this means ferritin is unreliable or that guidelines are inconsistent for no reason. It means a threshold is a policy decision layered on top of a continuous measurement. When you see the AGA use 45 ng/mL, the WHO use 15 or 70 micrograms per liter, and ASH propose 20 or 30 ng/mL, you are watching the same evidence appraised for different people and different purposes. Reading the cutoff alongside the population it was built for, and alongside any inflammatory markers, matters more than memorizing a single number. This article is educational and not a substitute for individual medical advice.