Between late 1992 and early 1993, five toddlers in Los Angeles County died after swallowing iron tablets. They were 11 to 18 months old. The tablets were prenatal iron, 325 milligrams of ferrous sulfate each, carrying 60 milligrams of elemental iron. Where anyone could count, the children had taken 30 to 40 of them.
The Centers for Disease Control and Prevention published those cases in Morbidity and Mortality Weekly Report on 19 February 1993, and stated in the same report that iron was then the most common cause of pediatric poisoning deaths reported to US poison control centers.
Four years later a federal rule put a warning inside a thin printed box on iron supplement labels. That sentence is probably the most-read piece of supplement regulation in America, and almost everything people assume about it is a little off.
It is required text, not a published finding. It reaches fewer products than it appears to. And the deaths it describes became far rarer after it was required. All three are true at once, and none of them makes iron safer for a small child.
The sentence is required text, not a finding
The rule is 21 CFR 101.17(e)(1). It does not describe a warning. It dictates one, word for word, and the first sentence a label must carry reads: “WARNING: Accidental overdose of iron-containing products is a leading cause of fatal poisoning in children under 6.”
Two more required sentences follow. One instructs the reader to keep the product out of the reach of children. The other instructs that in case of accidental overdose, a doctor or poison control center should be called immediately.
Now look at what the regulation does not contain. No data. No study. No citation. No date attached to the claim, and no statement anywhere that the claim is current. The words are the requirement. A manufacturer complies by printing them, not by verifying them.
The typography is doing work the text does not do. Paragraph (e)(5) requires the statement to be set off in a box using hairlines, which is why it reads like a finding lifted from a surveillance report. The rest of paragraph (e) is placement: the information panel of the immediate container, unit-dose labeling that stays intact until all units are used, the retail package, and any other labeling that carries warnings.
What the rule reaches
The scope clause is narrower than the sentence it requires. The rule applies to “any dietary supplement in solid oral dosage form (e.g., tablets or capsules) that contains iron or iron salts for use as an iron source.”
The sentence in the box says “iron-containing products.” The rule that forces it onto the label covers solid oral dosage dietary supplements. Liquids and powders are not in scope, and paragraph (e) contains no exemption clause letting them out. They were never in.
Two other rules fill part of that gap. FDA’s compliance guide points to 21 CFR 310.518(a) for iron-containing drugs. And the Consumer Product Safety Commission, at 16 CFR 1700.14(a)(12) and (a)(13), requires child-resistant packaging for iron-containing drugs and supplements holding 250 milligrams or more of elemental iron per package, reaching liquids as well as solids.
So a liquid iron supplement without the printed box did not slip past a rule. The labeling rule never covered it, while another agency’s packaging rule may still apply.
What happened after the warning went on labels
According to FDA’s Small Entity Compliance Guide on this subject, the agency published the final rule on 15 January 1997 at 62 FR 2218, effective that July. It had two parts: the label warning, and unit-dose packaging for products containing 30 milligrams or more of iron per dosage unit.
The packaging half did not survive. In Nutritional Health Alliance v. FDA, the court concluded that the statute gave FDA no authority to regulate such packaging for poison prevention, and FDA removed those provisions on 17 October 2003 at 68 FR 59714. FDA’s guide states that the ruling affected only the unit-dose requirements. The label warning was untouched and is still in force.
Then the numbers. The NIH Office of Dietary Supplements reports that between 1983 and 2000, at least 43 US children died from ingesting supplements containing high doses of iron, and that between 1998 and 2002, only one such death was reported.
One reported death across five years, against at least 43 over the preceding seventeen.
Two cautions belong with that. The windows overlap, the counts come from different reporting systems, and labeling was not the only thing that changed. The defensible claim is that the reported toll fell steeply as packaging and labeling changed, not that the printed box did it alone.
Notice too what the 1993 CDC sentence, the closest federal epidemiological statement behind the warning’s wording, does not say. It names poison control center reports and contains no reference to children under 6. That phrase belongs to the regulation, not to the epidemiology. The sentence in the box is a 1990s sentence about a 1990s hazard, still legally required, and quoting it as a description of America in 2026 misstates both the regulation and the data.
The hazard itself has not gone anywhere
The reason for all of this sits in per-kilogram arithmetic that has not changed since 1993.
CDC’s 1993 report gives the poison-center figures for 1991: 5,144 reported ingestions of iron supplements, 11 of them fatal. Children under 6 accounted for 3,578 of those ingestions, 69.6 percent, and nine of the eleven deaths.
The same report puts a toxic dose of elemental iron at 30 milligrams per kilogram of body weight and a fatal dose at typically more than 250 milligrams per kilogram, while noting deaths as low as 60 milligrams per kilogram. CDC put the scale this way: as few as five or six tablets of a high-potency preparation could be fatal for a child weighing 10 kilograms, about 22 pounds.
The Office of Dietary Supplements, in its iron fact sheet updated 4 September 2025, gives the adult figure: acute intakes above 20 milligrams per kilogram of body weight, roughly 1,365 milligrams for someone weighing 150 pounds, can cause corrosive necrosis of the intestine, and overdoses can lead to multisystem organ failure, coma, convulsions and death.
The gap between those two numbers is the whole problem. A bottle sized for a 150-pound adult sits in a house with a 22-pound person who cannot read it.
That makes the last sentence in the regulatory box the operative one. A suspected iron ingestion by a small child is a call to a doctor or a poison control center straight away. It is not a situation for waiting to see whether symptoms appear, and the regulation uses the word immediately for a reason.
The iron numbers, and what the limit counts
These are reference values set by the Food and Nutrition Board at the National Academies and republished by the Office of Dietary Supplements in that September 2025 fact sheet. They describe populations, not people.
Recommended Dietary Allowances, in milligrams of iron per day: 0.27 from birth to 6 months, which is an Adequate Intake rather than an RDA; 11 at 7 to 12 months; 7 at ages 1 to 3; 10 at 4 to 8; 8 at 9 to 13. At 14 to 18 the figures split by sex, 11 for boys and 15 for girls. From 19 to 50 the split widens to 8 for men and 18 for women. At 51 and over both return to 8. Pregnancy is 27 at every age band, lactation 10 at 14 to 18 and 9 at 19 to 50.
The adult gap is the striking one, and the reason is menstrual blood loss. A woman of 30 has a figure more than twice a man’s; the same woman at 55 has the figure he does. Pregnancy and other life stages are covered in article 335.
The Tolerable Upper Intake Level is 40 milligrams a day from birth through 13 years, and 45 milligrams a day from age 14 upward, including during pregnancy and lactation.
What that ceiling counts is worth getting right. The Office of Dietary Supplements states that the Food and Nutrition Board established the iron upper limits for iron from food and supplements, based on the amounts associated with gastrointestinal effects following supplemental intakes of iron salts, and that they apply to healthy infants, children and adults. The National Academies table of upper limits for elements attaches no footnote to iron, so its general rule governs: unless a nutrient says otherwise, its upper limit covers total intake from food, water and supplements. Magnesium carries such a footnote. Iron does not.
A ceiling is not a cliff, and not a cap on medical treatment. The same fact sheet notes that physicians sometimes prescribe intakes above the upper limit, for instance when someone with iron deficiency anemia needs higher doses to rebuild stores, and reports that supplemental iron at 45 milligrams a day or more may cause side effects such as nausea and constipation. Upper limits are covered in article 323.
One more label trap: those figures are elemental iron, and a tablet is named for the whole compound. The Office of Dietary Supplements gives ferrous sulfate as 20 percent elemental iron by weight, ferrous fumarate as 33 percent and ferrous gluconate as 12 percent.
Deficiency and anemia are not the same condition
They are routinely treated as one thing, and the order of events separates them. Iron stores go first. The Office of Dietary Supplements describes a progression in which stores deplete and transferrin saturation falls while hemoglobin is usually still inside the normal range. Iron deficiency anemia arrives at the end, once stores are exhausted, as low hemoglobin with decreases in hematocrit.
The consequence for anyone reading their own lab results is concrete. A normal hemoglobin, and a normal complete blood count, do not rule out iron deficiency.
The professional consensus moved in 2026. The American Society of Hematology published guidelines for the diagnosis of iron deficiency in Blood Advances, stating in its abstract that existing guidance on diagnostic criteria had been limited and not evidence based. Reviewing evidence through July 2025, the panel set serum ferritin thresholds with the stated goal of diagnosing iron deficiency irrespective of the presence of anemia. That last phrase is the point of the document: catch the shortage before the anemia.
Two federal prevalence figures get merged constantly and measure different things. NCHS Data Brief 519, published 4 December 2024, measured anemia by hemoglobin in NHANES data from August 2021 through August 2023: 9.3 percent of people aged 2 and over, 13.0 percent of females against 5.5 percent of males, and 17.4 percent of adolescent girls aged 12 to 19. It measures anemia, whatever its cause.
The iron factsheet from CDC’s 2026 National Nutrition Report, released in June 2026, measures iron deficiency by serum ferritin adjusted for inflammation, in females aged 12 to 49. It rose from about 13 percent in 2003-2004 to about 18 percent in August 2021 to August 2023, and among Mexican-American women from about 15 percent to about 28 percent, with no change among non-Hispanic Black or White women. Different condition, different age band, different denominator.
What ferritin measures, and what inflammation does to it
The World Health Organization’s 2020 guideline describes ferritin as an iron-storage protein present in all cells. Measuring it in serum estimates how much iron is banked rather than how much is circulating.
The Office of Dietary Supplements gives the usual reading: a serum ferritin below 30 micrograms per liter suggests iron deficiency, and below 10 suggests iron deficiency anemia. One nanogram per milliliter equals one microgram per liter, so a lab reporting ng/mL and a guideline reporting mcg/L are using the same number.
Then the complication that makes ferritin a conversation rather than a verdict. The Office of Dietary Supplements notes that inflammation, for example in people with infectious disease, can raise serum ferritin, and WHO lists liver disease, obesity, inflammation and malignancy among the things that elevate it. WHO builds an adjustment into its own recommendations: where infection or inflammation is present, a conditional recommendation graded low certainty allows a ferritin below 30 micrograms per liter in children and below 70 in adults to indicate iron deficiency. The adult threshold more than doubles.
Clinical literature often labels ferritin an acute phase reactant. That is not the Office of Dietary Supplements’ term, and WHO applies the acute-phase response protein label to C-reactive protein and alpha-1-acid glycoprotein instead. The behavior is what matters: ferritin rises when the body is inflamed, whether or not stored iron rose with it. Deficiency measurement in general is covered in article 321.
Hemochromatosis, and the people for whom more iron is the wrong direction
The National Institute of Diabetes and Digestive and Kidney Diseases, on a page last reviewed in January 2020, describes hemochromatosis as a disorder in which extra iron builds up to harmful levels, damaging the liver, heart, pancreas, endocrine glands and joints. Gene mutations cause the most common type.
It is not rare. The 2011 practice guideline of the American Association for the Study of Liver Diseases calls hereditary hemochromatosis the most common identified genetic disorder in Caucasians, with a prevalence of roughly 1 per 220 to 250 individuals in populations of northern European origin. The Office of Dietary Supplements gives the American figures the same way: about 1 in 10 White Americans carries the C282Y mutation, and 4.4 per 1,000 are homozygous and have hemochromatosis.
Carrying the gene is not having the disease. AASLD reports phenotypic expression in approximately 70 percent of C282Y homozygotes, and severe iron overload with organ damage in fewer than 10 percent, part of why it concluded that general population screening is not indicated.
Treatment is removal rather than restriction. AASLD’s regimen is therapeutic phlebotomy, which NIDDK describes as drawing about a pint of blood at a time on a schedule, and the guideline quantifies why diet is not the lever: a low-iron diet changes absorption by 2 to 4 milligrams a day, against roughly 250 milligrams a week mobilized by phlebotomy.
Supplements are treated differently from food. AASLD’s treatment table says to avoid vitamin C supplements, because pharmacological doses accelerate the mobilization of iron to a level that can saturate circulating transferrin and raise pro-oxidant activity. NIDDK’s patient page says a doctor may recommend avoiding iron and vitamin C supplements, and limiting alcohol.
One AASLD recommendation deserves a general reader’s attention: patients with abnormal iron studies should be evaluated as hemochromatosis patients even without symptoms. The laboratory finding comes first.
Why clinicians want a number before anyone adds iron
Four things set iron apart from most nutrients sold in a bottle.
Low iron is a finding with a cause, and the cause is usually the clinically important part. In 2020 the American Gastroenterological Association published a clinical practice guideline in Gastroenterology devoted entirely to the gastrointestinal evaluation of iron deficiency anemia. A specialty society wrote a whole guideline about where to look.
The two most familiar checks do not settle the question alone. A normal hemoglobin does not rule out depleted stores, and a ferritin result can be pushed upward by inflammation. Each error runs in a different direction.
A meaningful number of people have the opposite problem. On the Office of Dietary Supplements’ figures, 4.4 White Americans per 1,000 have hemochromatosis, and nothing about how they feel will tell them so.
The evidence for routine supplementation is thinner than its popularity, even where it is most standard. On 20 August 2024 the US Preventive Services Task Force gave a grade of I to both screening and routine iron supplementation in asymptomatic pregnant adolescents and adults: the evidence was insufficient to weigh benefits against harms. A grade of I is not a recommendation against, and insufficient evidence is not evidence of no benefit. The Task Force judged the harms of routine supplementation to be no greater than small, and noted gastrointestinal effects.
What a blood test is deciding
- Are iron stores low, or is any anemia coming from something other than iron?
- If stores are low, where is the iron going?
- Is inflammation holding the ferritin figure above what the stores justify?
- Is this the opposite problem, where iron is accumulating rather than running out?
Those four questions are the work, none of them can be answered from a label, and the answers point in different directions. Timing against medicines is covered in article 677: iron interacts with levothyroxine, levodopa and proton pump inhibitors, among others.
The short version
The sentence in the hairline box is a labeling requirement, in force since 1997. It carries no study, no citation and no date, and it reaches only solid oral dosage dietary supplements though it speaks about iron-containing products generally. The reported child deaths behind it fell steeply after packaging and labeling changed.
None of that softens the hazard, because the arithmetic underneath is unchanged. A dose scaled to a 150-pound adult is enormous for a 22-pound toddler, and CDC’s 1993 thresholds still describe what happens. The last sentence in that box is the one to act on, and acting on it means calling immediately.
For adults the useful fact is different. Iron deficiency and iron deficiency anemia are not the same condition, ferritin is the measurement that separates them, and inflammation moves ferritin. That is why the question belongs with a clinician holding a number, and why the professional guidelines here are about diagnosis, not bottles.
Sources
- Cornell Law School, Legal Information Institute. “21 CFR 101.17 – Food labeling warning, notice, and safe handling statements.” law.cornell.edu
- Cornell Law School, Legal Information Institute. “16 CFR 1700.14 – Substances requiring special packaging.” law.cornell.edu
- U.S. Food and Drug Administration. “Small Entity Compliance Guide: Label Warning Statements for Iron-Containing Supplements and Drugs.” fda.gov
- NIH Office of Dietary Supplements. “Iron: Fact Sheet for Health Professionals.” ods.od.nih.gov
- Centers for Disease Control and Prevention. “Toddler Deaths Resulting from Ingestion of Iron Supplements – Los Angeles, 1992-1993. MMWR, 19 February 1993.” cdc.gov
- CDC National Center for Health Statistics. “Anemia Prevalence: United States, August 2021-August 2023. NCHS Data Brief No. 519.” cdc.gov
- Centers for Disease Control and Prevention. “2026 CDC Nutrition Report: Iron factsheet.” cdc.gov
- National Institute of Diabetes and Digestive and Kidney Diseases. “Hemochromatosis.” niddk.nih.gov
- U.S. Preventive Services Task Force. “Iron Deficiency Anemia in Pregnant Persons: Screening and Supplementation.” uspreventiveservicestaskforce.org
- World Health Organization (via NCBI Bookshelf). “WHO guideline on use of ferritin concentrations to assess iron status in individuals and populations, 2020.” ncbi.nlm.nih.gov
- National Academies of Sciences, Engineering, and Medicine, Food and Nutrition Board. “Dietary Reference Intakes (DRIs): Tolerable Upper Intake Levels, Elements.” nationalacademies.org
- PubMed. “Powers JM, Lim MY, Achebe MO, et al. "American Society of Hematology 2026 guidelines for diagnosis of iron deficiency." Blood Advances, 2026. PubMed. DOI: 10.1182/bloodadvances.2025015950.” doi.org
- PubMed. “Ko CW, Siddique SM, Patel A, et al. "AGA Clinical Practice Guidelines on the Gastrointestinal Evaluation of Iron Deficiency Anemia." Gastroenterology, 2020. PubMed. DOI: 10.1053/j.gastro.2020.06.046.” doi.org
- PubMed. “Bacon BR, Adams PC, Kowdley KV, Powell LW, Tavill AS. "Diagnosis and management of hemochromatosis: 2011 practice guideline by the American Association for the Study of Liver Diseases." Hepatology, 2011. PubMed. DOI: 10.1002/hep.24330.” doi.org