The Underdiagnosis Crisis — Ferritin Thresholds, Reference Range Failures, and What Clinicians Are Missing

Freed explains why iron deficiency anemia is being missed at an alarming rate in clinical practice and what the ASH draft guidelines propose to correct this systemic failure.

Iron deficiency is among the most prevalent nutritional deficiencies in the world, yet it is being underdiagnosed at a scale that most clinicians would find surprising. Data from the National Health and Nutrition Examination Survey (NHANES) — a nationally representative cohort of approximately 10,000 healthy adults — illustrate the magnitude of the problem. Using a ferritin cutoff of 15 µg/dL, a threshold applied by many laboratories, 17% of premenopausal women meet criteria for iron deficiency. At a cutoff of 30 µg/dL, that proportion rises to 39%. At 50 µg/dL — a threshold with strong physiologic and clinical justification — 62% of premenopausal women are iron deficient. Most of those women are not receiving iron supplementation, meaning the gap between prevalence and treatment is not marginal but structural. A national cohort analysis of more than 120,000 patients found that among the roughly 12,000 who presented to primary care with a chief complaint of fatigue, only half had a complete blood count (CBC) performed, and only approximately 10% had ferritin testing — despite iron deficiency being one of the most common and correctable causes of that symptom. Of those who did have ferritin measured, two-thirds were found to be iron deficient.

The root cause of this underdiagnosis is not clinical inattention but a structural flaw in how laboratory reference ranges are constructed. Reference ranges are developed locally: each institution validates its own normal range by measuring a parameter in approximately 120 healthy individuals from the surrounding community and defining normal as the central 95th percentile of that distribution. This approach works reasonably well for most laboratory values, but it fails catastrophically for iron and ferritin in populations where iron deficiency is endemic. When two-thirds of premenopausal women are iron deficient, those women constitute the healthy reference population — and iron deficiency becomes enshrined as normal by the mathematics of the method itself. The consequence is visible in hemoglobin reference ranges: many laboratories use a lower limit of approximately 10.7 g/dL for women compared with 13.5 g/dL for men, a gap of nearly 3 g/dL that cannot be explained by the physiologic effects of testosterone alone, which accounts for approximately 1 g/dL of true sex-based difference. The remainder reflects under-recognized, under-treated iron deficiency that has been absorbed into the definition of normal.

In this video discussion on iron deficiency anemia, Jason Freed, MD, an assistant professor of medicine at Harvard Medical School, examines what the ASH draft guidelines — issued for public comment in 2025 — propose to address this problem. The draft recommends a ferritin cutoff of 30 µg/dL for general healthy adults, with consideration for a threshold of 50 µg/dL, representing a meaningful departure from the World Health Organization's longstanding cutoff of 15 µg/dL and from the local reference ranges most clinicians encounter in their electronic health records. Freed notes that the debate between 30 and 50 reflects a legitimate concern about over-medicalizing iron deficiency in asymptomatic individuals, but argues that for any patient presenting with a symptom attributable to iron deficiency — anemia, fatigue, restless legs syndrome — a ferritin below 50 µg/dL is an actionable finding. He emphasizes that correcting reference ranges at the health system level will ultimately require institutional advocacy and committee action, not just individual clinician awareness, and that the finalization of the ASH guidelines is a necessary prerequisite for that broader implementation effort to begin in earnest.