Story
More is not better
Last updated
In one pass Extra nutrients you take in do not simply vanish.
Educational content, not medical advice. See a clinician about your own case.
Story path
Chapter 1
Stored or flushed out
Extra nutrients you take in do not simply vanish. Where they go depends on which exit the body has left open for them.
Flushed out by the kidneys: vitamin C dissolves in water. In an inpatient study of 7 healthy men, blood plasma was fully saturated at about 1000 mg a day; with single doses of 500 mg or more, the share absorbed fell, and most of what was absorbed left in the urine.
Stored in the liver: vitamin A dissolves in fat. Most of the body's supply is kept as retinyl esters in small lipid droplets inside the liver's stellate cells. Any excess piles up there too: sources note that taking roughly 10 times the recommended amount or more over the long term can lead to chronic toxicity after months to years, harming the liver.
Almost no exit: the body loses only about 1–2 mg of iron a day, mostly as cells are shed from surfaces such as the skin and the gut lining, and that loss is not meaningfully adjusted to how much iron the body holds, so the body's iron total can be regulated almost only by controlling absorption at the gut entrance (bleeding and menstruation take some extra). A child who swallows iron-containing products is an emergency: get medical help or call poison control right away.
So when you meet a nutrient, one question is worth asking first: does the body store it, and how does it get out?
Flushed out by the kidneys: vitamin C dissolves in water. In an inpatient study of 7 healthy men, blood plasma was fully saturated at about 1000 mg a day; with single doses of 500 mg or more, the share absorbed fell, and most of what was absorbed left in the urine.
Stored in the liver: vitamin A dissolves in fat. Most of the body's supply is kept as retinyl esters in small lipid droplets inside the liver's stellate cells. Any excess piles up there too: sources note that taking roughly 10 times the recommended amount or more over the long term can lead to chronic toxicity after months to years, harming the liver.
Almost no exit: the body loses only about 1–2 mg of iron a day, mostly as cells are shed from surfaces such as the skin and the gut lining, and that loss is not meaningfully adjusted to how much iron the body holds, so the body's iron total can be regulated almost only by controlling absorption at the gut entrance (bleeding and menstruation take some extra). A child who swallows iron-containing products is an emergency: get medical help or call poison control right away.
So when you meet a nutrient, one question is worth asking first: does the body store it, and how does it get out?
Numbers · The vitamin C saturation curve
The level of vitamin C in the blood does not keep rising with every extra milligram you take. In 1996, Levine and colleagues kept 7 healthy men in hospital for 4 to 6 months, first on a diet with less than 5 mg of vitamin C a day, then raised the dose step by step.Plotted against dose, plasma concentration traced an S-shaped curve: at 200 mg a day it was already past the steepest part of the rise, and at 1000 mg a day plasma was completely saturated. A single 200 mg dose was fully absorbed; at single doses of 500 mg or more, the share absorbed fell, and most of what was absorbed left in the urine.
The authors also recorded that at 1000 mg a day, more oxalate and uric acid came out in the urine, and they judged that more than 400 mg a day showed no evident value.
The study had only 7 people and no control group, and it measured concentrations in blood and urine, not health outcomes. It shows how the kidney exit works; it cannot tell anyone which dose is better for them.
Chapter 2
Vitamin A piles up in the liver
Preformed vitamin A (retinol and its esters) that the body does not need right away is mostly sent to the liver and handed to hepatic stellate cells for safekeeping: they turn it into retinyl esters and pack it into small lipid droplets inside the cell. Sources estimate that about 70% of an adult's total vitamin A sits in the liver, and how many droplets there are follows how much vitamin A the diet supplies.
Case material shows that after long-term intake of roughly 10 times the recommended amount or more, stellate cells swell, become activated and start making collagen, and fibrosis forms around the liver sinusoids; after years of this, some people develop portal hypertension. The trouble is that over the stretch from ample stores to toxic ones, blood retinol barely reflects the change. Agencies in both the US and the EU have set a long-term daily upper limit for preformed vitamin A, which is 3000 µg for adults.
Beta-carotene from plants does not count toward that limit, and the sources have found no evidence that high intakes of it cause vitamin A poisoning. Mouse and cell experiments show that the gut step that turns it into vitamin A tightens once vitamin A is plentiful.
Case material shows that after long-term intake of roughly 10 times the recommended amount or more, stellate cells swell, become activated and start making collagen, and fibrosis forms around the liver sinusoids; after years of this, some people develop portal hypertension. The trouble is that over the stretch from ample stores to toxic ones, blood retinol barely reflects the change. Agencies in both the US and the EU have set a long-term daily upper limit for preformed vitamin A, which is 3000 µg for adults.
Beta-carotene from plants does not count toward that limit, and the sources have found no evidence that high intakes of it cause vitamin A poisoning. Mouse and cell experiments show that the gut step that turns it into vitamin A tightens once vitamin A is plentiful.
Mechanism · Stellate cells go from storage to fibre making
The liver does not keep its vitamin A in hepatocytes, the cells that do the metabolic work, but in stellate cells, which sit between the hepatocytes and the blood sinusoids. In adult rodents with ample vitamin A, more than 90% of liver vitamin A is held as retinyl esters in the droplets of these cells; that proportion comes from animals, and for humans there is only the estimate that the liver holds about 70% of the body's total. Once a stellate cell is activated, it loses its droplets and the retinyl esters inside them.In excess, a liver biopsy shows enlarged stellate cells crammed with lipid droplets and fibrosis in the surrounding sinusoids; the source stresses that these droplet-filled cells are not hepatocytes. The process it describes runs like this: surplus vitamin A builds up in the stellate cells, and past a certain point the cells activate, enlarge and produce more collagen, ending in fibrosis; the damage grows with dose and can be reproduced in animal models.
The timing needs to be read in two parts. Symptoms and abnormal liver tests from chronic excess can appear months after intake begins; portal hypertension generally takes 1 to 8 years of fairly high doses, and it can appear before clear cirrhosis can be shown. Early changes can reverse, but recovery after stopping is slow, and people who already have chronic liver disease are more vulnerable. These conclusions rest on case reports, case series and animal experiments: they show the harm is real, but they cannot say what share of people will develop it.
Numbers · How the 3000 µg figure was derived
An upper limit (, tolerable upper intake level) is the highest long-term daily total intake from all dietary sources that is not expected to cause adverse health effects; it is not a recommended intake, risk rises once intake goes above it, and briefly exceeding it now and then is not necessarily harmful.The US IOM derivation of 2001 followed two lines, and both land on 3000 µg. For adults other than women of reproductive age, liver toxicity had been reported at supplement doses of 14,000 µg a day, and because that damage is serious and irreversible, the figure was divided by an uncertainty factor of 5. For women of reproductive age, the no-adverse-effect level for birth defects from preformed vitamin A is about 4500 µg a day, divided by 1.5 to allow for differences between individuals. This limit covers only preformed vitamin A taken long term from food, fortified food and supplements, not carotenoids.
The EU's EFSA reviewed it in 2024, named birth defects as the critical effect, and kept the adult limit at 3000 µg retinol equivalents a day for both sexes, including during pregnancy and breastfeeding; other ages are scaled down, from 600 µg for infants aged 4 to 11 months to 2600 µg for ages 15 to 17. These numbers come from risk assessment plus a safety margin, built on case reports and observational studies; no trial has shown that going above 3000 µg is bound to cause harm.
Evidence · High doses in early pregnancy and birth defects
The face, skull and some heart structures of a fetus come from a group of cells called the cranial neural crest, which migrate and specialise in early pregnancy. A 1995 prospective cohort followed 22,748 pregnant women. Counting supplements alone, women taking more than 10,000 a day, compared with those taking no more than 5,000 IU, had a prevalence ratio of 4.8 (95% 2.2 to 10.5) for defects arising from the cranial neural crest, and the excess was concentrated among mothers with high intake before week 7 of pregnancy.This is a single observational study: it shows an association and cannot prove that vitamin A was the cause, and the apparent threshold of about 10,000 IU was read off a smoothed curve.
The WHO guideline of 2011 does not include vitamin A supplementation in routine antenatal care, because it does not lower illness or death in mothers and babies; only in areas of severe deficiency, where night blindness is common, is it given to prevent night blindness in pregnancy. The same guideline says that a single dose above 25,000 IU is undesirable, especially between days 15 and 60 after conception.
rothman-vitamina-1995who-2011-vitamin-a-pregnancy
Mechanism · The gate in the gut
Beta-carotene becomes vitamin A only after it is absorbed into the cells lining the small intestine and cut apart by an enzyme called BCO1 (formerly BCMO1). Experiments in mice and human-derived cell lines show a negative feedback loop: retinoic acid, a product of vitamin A, acts through its receptors to raise the expression of an intestinal transcription factor called ISX, and ISX in turn suppresses SR-BI, which takes up beta-carotene, and BCO1, which cleaves it. The more vitamin A there is, the tighter this gate closes.Whether the pathway works the same way in people has so far only been inferred from cell lines, and the study gives no figures for conversion rates in humans. The agencies' conclusion is a separate line of evidence: the US IOM states that high beta-carotene intake has not been shown to cause hypervitaminosis A, which is why the upper limit covers only preformed vitamin A.
That does not mean beta-carotene supplements can be taken freely. The EU's EFSA set no upper limit for beta-carotene because the data on supplementation and lung cancer risk are not enough to describe how dose relates to risk, not because it has been shown to be safe; the available data also cannot say whether it worsens the toxicity of preformed vitamin A.
Safety · Smokers and beta-carotene supplements
Finland's ATBC trial randomly assigned 29,133 male smokers aged 50 to 69, with some taking 20 mg of beta-carotene a day, and followed them for 5 to 8 years. Those given beta-carotene had an 18% higher incidence of lung cancer (95% 3% to 36%) and 8% higher total mortality (1% to 16%). These are relative increases, the lower bound of the interval is only 3%, and the participants were limited to heavy-smoking men.On this basis the EU's EFSA considers that smokers should avoid supplements containing beta-carotene, and that supplementation in the general population should be limited to meeting vitamin A needs. The caution is about the concentrated doses in supplements, not the beta-carotene in vegetables and fruit.
atbc-1994-nejm
Chapter 3
Too much vitamin D pushes calcium into the blood
When vitamin D intake runs far too high, the first changes show up in the blood and the kidneys: calcium in the blood rises, which is called hypercalcemia, and more calcium leaves in the urine, which is called hypercalciuria.
Why is calcium pushed into the blood? Vitamin D's day job is managing calcium: it makes the gut absorb more of it and, when needed, lets bone release it. Normally the kidneys make only a small amount of the active form, on demand, so this switch is tightly controlled. In an overdose, in the blood rises far above normal. A review devoted to vitamin D toxicity sets out several explanations, all pointing to vitamin D signalling inside cells being switched on too hard; which one is right has not been settled.
One case report describes two adults whose supplements had both manufacturing and labeling errors. They were taking more than 1000 times the recommended daily dose of vitamin , and their blood calcium rose severely. In these two people, how high the calcium went tracked the level of 25(OH)D in the blood (the circulating form the liver makes from vitamin D), not the active form 1,25(OH)2D. This kind of excess clears slowly: 25(OH)D took about a year to return to normal.
Why is calcium pushed into the blood? Vitamin D's day job is managing calcium: it makes the gut absorb more of it and, when needed, lets bone release it. Normally the kidneys make only a small amount of the active form, on demand, so this switch is tightly controlled. In an overdose, in the blood rises far above normal. A review devoted to vitamin D toxicity sets out several explanations, all pointing to vitamin D signalling inside cells being switched on too hard; which one is right has not been settled.
One case report describes two adults whose supplements had both manufacturing and labeling errors. They were taking more than 1000 times the recommended daily dose of vitamin , and their blood calcium rose severely. In these two people, how high the calcium went tracked the level of 25(OH)D in the blood (the circulating form the liver makes from vitamin D), not the active form 1,25(OH)2D. This kind of excess clears slowly: 25(OH)D took about a year to return to normal.
Numbers · Urine calcium climbs with the dose
A three-year (RCT) enrolled 311 healthy adults aged 55–70 who were not short of vitamin D to begin with. They took 400, 4000 or 10000 a day. The share of people with high urine calcium was 17%, 22% and 33% respectively, rising with the dose. The share with high blood calcium was 0, 4% and 9%; all cases were mild and had resolved on retesting.This matches the choice EFSA made: it treated persistent hypercalciuria as the key effect for setting the upper limit, because that signal may appear earlier than persistent hypercalcemia.
Evidence · Taking more did not turn out better
Neither of two found a benefit from taking more, and each turned up a different unfavourable signal.The first enrolled 2256 community-dwelling women over 70 at high risk of fracture. They took a single oral dose of 500,000 of vitamin each autumn or winter for 3–5 years. Compared with placebo, the incidence rate ratio (IRR) was 1.15 for falls and 1.26 for fractures. For fractures, the lower bound of the sat right at 1.00, so the result only just reached statistical significance. A post hoc analysis found that fall risk was highest in the first 3 months after each dose. This applies only to a once-a-year large dose in these older women; it cannot be carried over to ordinary daily or monthly doses.
The second is the three-year trial described above. In both the 4000 IU and the 10000 IU groups, at the radius (the outer bone of the forearm) was lower than in the 400 IU group; at the tibia, only the 10000 IU group was lower. Bone strength, however, did not differ significantly between groups, and the trial measured bone density rather than fractures, so it cannot be read as bones becoming brittle or breaking more easily. The result ran against what the researchers expected, and the authors themselves said it only generates a hypothesis that other studies need to confirm.
Clinical · Cases of accidental massive overdose
Both case reports below come from product accidents: the real dose was far higher than the label said, not an ordinary dose taken a little too often. Case reports show that this harm really happens, but they cannot tell us how common it is.Seven children aged 0.7–4.2 took the same Fish Oil supplement. Testing found about 4000 times the labeled concentration of vitamin , for an estimated intake of 266,000 to 800,000 a day. All seven developed hypercalcemia. With hospital treatment, blood calcium returned to normal within about 3 days and vitamin D levels within 2–3 months, and at one-year follow-up none had nephrocalcinosis (calcium deposits in kidney tissue).
The two adults mentioned at the start had normal blood calcium and no symptoms once their fell below 400 ng/mL, and they were left with no long-term harm.
Numbers · How the upper limit is derived
EFSA in 2023 took the lowest dose that produced hypercalciuria in two human (the LOAEL): 250 µg a day. Because no dose could be identified as clearly free of problems (a NOAEL), it divided by an uncertainty factor of 2.5, giving 100 µg a day for adults. The same figure applies to pregnant and breastfeeding women and to adolescents aged 11–17; children aged 1–10 get 50 µg a day because of their smaller body size. The unit is µg of vitamin D equivalents.The US IOM in 2011 set the for everyone aged 9 and over at 4000 a day, which at 40 IU per µg is also 100 µg. That figure is a committee's expert judgment that takes uncertainty into account, not a number read directly off a dose-toxicity trial.
So it is neither a poisoning threshold nor a target; above this line, the existing evidence simply cannot assure safety. A UL concerns long-term daily intake, and going over it briefly now and then does not necessarily cause harm. It is derived to protect the general population, including its more sensitive members, and it does not apply to people taking the nutrient under medical supervision.
Chapter 4
Iron has an entrance but no exit
Once iron is in the body, it has almost no way out. Each day the body loses only about 1 to 2 mg of iron in cells shed from the gut lining and other surfaces (bleeding adds a little more), and this amount does not rise when the body holds more iron; there is also no active route for excreting it. So how much iron the whole body holds depends almost entirely on how much the gut absorbs.
The gatekeeper is a small hormone made by the liver, hepcidin. On the side of the gut cell that faces the blood sits the channel iron uses to leave, called ferroportin. Cell experiments show that when hepcidin binds it, the channel is pulled into the cell and broken down, and iron can no longer get out. The fuller the iron stores, the higher hepcidin rises and the tighter the door closes; EFSA regards this as the most important safeguard against iron overload in the body.
The door also responds to a single large dose of iron. In young women who were iron-depleted but not anemic, a single dose of 60 mg or more left hepcidin still raised 24 hours later, and the share of the next dose that was absorbed was 35% to 45% lower. Swallow twice as much, and far less than twice as much gets in.
The gatekeeper is a small hormone made by the liver, hepcidin. On the side of the gut cell that faces the blood sits the channel iron uses to leave, called ferroportin. Cell experiments show that when hepcidin binds it, the channel is pulled into the cell and broken down, and iron can no longer get out. The fuller the iron stores, the higher hepcidin rises and the tighter the door closes; EFSA regards this as the most important safeguard against iron overload in the body.
The door also responds to a single large dose of iron. In young women who were iron-depleted but not anemic, a single dose of 60 mg or more left hepcidin still raised 24 hours later, and the share of the next dose that was absorbed was 35% to 45% lower. Swallow twice as much, and far less than twice as much gets in.
Evidence · What the alternate-day trial actually compared
If one dose of iron keeps hepcidin raised for a whole day, would taking it every other day be absorbed better? A small randomized trial tested exactly that. Forty women who were iron-depleted but not anemic each took 60 mg per dose, 14 doses in all: one group took a dose every day for 14 days, the other took a dose every other day across 28 days. Measured with isotope-labelled iron, the alternate-day group absorbed 21.8% of the total in the end and the daily group 16.3%; the cumulative amount absorbed was about 175 mg versus 131 mg. The daily group also had higher hepcidin.Four limits apply when reading this result. First, both groups took the same number of doses and the same total, so the alternate-day group needed twice as many days; over the same 28 days a daily taker swallows twice as much iron, and this study did not compare who absorbs more within the same span of time. Second, it measured how much was absorbed, not whether hemoglobin or fatigue improved. Third, the women were not anemic, so whether anemic people respond the same way still needs separate testing. Fourth, the sample was small and the trial was not blinded. That raised hepcidin suppresses the next day's absorption was observed alongside the result; it is a reasonable explanation, not a proven cause.
The same team's 2015 study adds one more number: raising the dose sixfold only about tripled the total amount absorbed. The bigger the dose, the smaller the share absorbed.
Numbers · Why agencies set different limits for iron
Major agencies give different numbers for the long-term upper limit of iron because they are watching different endpoints.First, what an upper limit is. In EFSA's definition, the tolerable upper intake level () is the highest level of long-term daily intake from all sources at which no adverse health effects are expected; it is not a recommended intake, and briefly going over it now and then is not the same as harm.
In 2001 the US Institute of Medicine (IOM) set the adult UL for iron at 45 mg a day, based on gastrointestinal distress seen with iron-salt supplements. That is an acute reaction that clears on its own, not iron overload; the limit also does not apply to people taking iron under a doctor's supervision to treat deficiency.
When EFSA reassessed iron in 2024, it could not set a UL. It accepts that iron overload throughout the body damages organs, but the only indicator for which a dose-response could be established was black stools, and black stools are not an adverse effect in themselves. EFSA therefore gave only a safe level of intake of 40 mg a day for adults (including during pregnancy and breastfeeding): no black stools at 20 to 25 mg of supplemental iron, plus about 15 mg from the diet. This figure is not a threshold above which harm begins, nor a recommended dose; the values for children are lower.
China's 2017 standard for dietary reference intakes sets the iron UL at 42 mg a day for adults aged 18 and over (including pregnant and breastfeeding women). It is not the same concept as EFSA's 40 mg, so the two should not be lined up side by side as national upper limits.
Clinical · People whose iron keeps building up
Since the body has no exit for iron, any extra that gets in can only stay. Some people are more likely to end up there: carriers of a genotype linked to hereditary hemochromatosis, two copies of the C282Y variant of the HFE gene. In a North American primary-care screening of nearly 100,000 people, about 0.44% of non-Hispanic white participants had two copies, with far lower rates among Native Americans, Hispanics, Black participants and Asians.The problem lies at that door. Too little hepcidin leads to iron overload, and hereditary hemochromatosis is the typical case: the door fails to close when it should, the gut keeps absorbing more, and iron slowly builds up. So these people accumulate iron not because they eat more of it than others.
Having this genotype does not mean the disease will follow. In the same study, among two-copy carriers who had not been diagnosed before, about 88% of men had above 300 µg/L and about 57% of women had ferritin above 200 µg/L; and raised ferritin does not mean organs are already damaged. Once iron overload sets in throughout the body, though, it damages organs.
Red flag · When a child swallows a handful of iron pills
Swallowing a large amount of iron at once is a different danger from iron building up slowly. In 1997 the US FDA required iron-containing supplements and drugs to carry a warning, because accidental overdose of iron-containing products was then a leading cause of fatal poisoning in children under 6 in the United States.In a home with young children, keep iron-containing products out of their reach; if a child swallows one, get medical help or call poison control right away.
Chapter 5
Water-soluble B6 can still harm nerves
Many people assume that water-soluble vitamins are harmless in excess, because the extra just gets flushed out. B6 is a counterexample, and the damage lands on the sensory nerves: the long nerves that carry touch and position sense from the hands and feet back to the brain. In a 1983 case report, 7 adults who took large daily doses of pyridoxine, one form of B6, developed severe sensory problems and unsteady walking, which doctors call sensory ataxia. Their muscle strength was not affected; 4 of them were severely disabled for a time, and all 7 improved after they stopped.
Why would too much look like too little? The form of B6 that actually does the work in the body is the active form, PLP (pyridoxal 5'-phosphate), which serves as a helper for many enzymes. A laboratory study in cells and enzymes found that large amounts of unactivated pyridoxine inhibited two PLP-dependent enzymes. The authors propose that it competes with PLP for its place, which is why the symptoms of taking too much resemble those of deficiency. This is an explanation put forward in the lab; it has not been measured directly in the human body.
Why would too much look like too little? The form of B6 that actually does the work in the body is the active form, PLP (pyridoxal 5'-phosphate), which serves as a helper for many enzymes. A laboratory study in cells and enzymes found that large amounts of unactivated pyridoxine inhibited two PLP-dependent enzymes. The authors propose that it competes with PLP for its place, which is why the symptoms of taking too much resemble those of deficiency. This is an explanation put forward in the lab; it has not been measured directly in the human body.
Mechanism · Only the pyridoxine form caused trouble
B6 in supplements comes in more than one chemical form. That laboratory study used two kinds of cells, a cell line often used to model nerve cells (SH-SY5Y) and a gut cell line (CaCo-2), and exposed them to five forms of B6 for 24 hours each. Only pyridoxine made the cells die, and more of it killed more cells; the other four forms did not have this effect.In the same study, pyridoxine also inhibited two enzymes that need PLP to work. From this, the authors propose that unactivated pyridoxine takes up the place that belongs to PLP, so the enzymes cannot get going, and the body behaves as if it were short of B6 when it has taken too much.
The limits of this study need saying: these are results from dishes and isolated enzymes. It gives no human dose and cannot tell how much intake causes harm. What it offers is one possible explanation for the nerve symptoms seen in the case reports.
Numbers · Agencies draw the line in different places
Agencies draw very different lines for B6. In 1998 the US Institute of Medicine (IOM) set the for adults in the United States and Canada at 100 mg a day, with sensory neuropathy as the key effect to guard against. After a new assessment in 2023, the European Food Safety Authority (EFSA) set the adult UL at 12 mg a day, replacing the 25 mg set in 2000.Australia's Therapeutic Goods Administration (TGA) took a different route. Among the adverse-reaction reports it received, some people developed peripheral neuropathy at under 50 mg a day, and some had been taking several B6-containing products at once. It therefore required products with a daily dose above 10 mg to carry a warning, and lowered the daily maximum for a single adult product to 100 mg.
These numbers are not the same kind of quantity. The IOM and EFSA ULs cover one person's total intake over the whole day, long term; the TGA's 10 mg is the threshold for printing a warning, and its 100 mg applies to a single product. When you see a B6 upper limit, first check whose it is and what it covers.
efsa-2023-b6-ul
Numbers · Who crosses the line
In this survey, the part over the line came entirely from supplements. In data from the US National Health and Nutrition Examination Survey (NHANES) 2003–2006, among adults aged 19 and over who used supplements, about 3.5% had a total B6 intake from food plus supplements above the IOM's 100 mg a day. In the same group, counting food alone, no one was above it.This number needs to be read correctly. The denominator is supplement users, not all adults. Being above the only means risk may rise; it does not mean these people developed nerve damage, and the survey did not measure any health outcome. It is a cross-sectional survey from the United States about 20 years ago and cannot be taken as the current picture anywhere.
Evidence · How solid this evidence is
Every piece of evidence in this story has a clear weakness. The 1983 report has only 7 cases and no control group. It shows that this nerve damage really exists, what it looks like, and that it can improve after stopping, but it cannot say at what dose risk begins, and it gives no rate of occurrence.The IOM's 100 mg rests on a small number of studies. The TGA's low-dose cases come from pooled adverse-reaction reports; reports of this kind cannot say how many people will be harmed, and on their own they cannot confirm cause and effect. The competing-for-its-place mechanism has only been seen in cells and isolated enzymes.
Put together, the conclusion that holds up is a modest one: high doses of pyridoxine taken over a long time can damage the sensory nerves, and being water-soluble does not guarantee safety. Where exactly the safe line should be drawn is a question the agencies have not yet answered in the same way.
Chapter 6
How the upper limit is set, and how stacking crosses it
How much of a nutrient, arriving each day from food, fortified foods and supplements, can the body take in over the long run before harm is expected? The Tolerable Upper Intake Level () answers that question. It counts all sources added together, and it is set to protect the general population, including its more sensitive members. It is not a target and not a recommended intake: above it, risk starts to rise. It concerns long-term daily intake, so going over it briefly now and then does not by itself mean harm.
The line is worked out backward. First comes the critical adverse effect, the problem that shows up first as the dose climbs. Next comes the highest dose at which that effect was not seen, the no-observed-adverse-effect level (NOAEL); if there is none, the step back is to the lowest dose at which the effect did appear, the lowest-observed-adverse-effect level (LOAEL). Last, that dose is divided by an uncertainty factor (UF). The less certain the evidence, the larger the factor and the lower the limit.
The limit matters because some nutrients stay in the body and have no easy way out: vitamin A builds up in the liver, and iron has almost no exit.
The line is worked out backward. First comes the critical adverse effect, the problem that shows up first as the dose climbs. Next comes the highest dose at which that effect was not seen, the no-observed-adverse-effect level (NOAEL); if there is none, the step back is to the lowest dose at which the effect did appear, the lowest-observed-adverse-effect level (LOAEL). Last, that dose is divided by an uncertainty factor (UF). The less certain the evidence, the larger the factor and the lower the limit.
The limit matters because some nutrients stay in the body and have no easy way out: vitamin A builds up in the liver, and iron has almost no exit.
Mechanism · The four steps behind an upper limit
The 2024 guidance from the European Food Safety Authority (EFSA) and the 1998 model from the US Institute of Medicine (IOM) follow the same risk assessment. Step one identifies the hazard: what goes wrong with too much. Step two describes how the effect depends on dose, picking the critical effect and its NOAEL or LOAEL. Step three estimates how much people actually take in. Step four puts these together to say who is at risk, and at what intake.The limit itself comes out of step two: the NOAEL (or LOAEL) divided by one factor that bundles all the relevant uncertainties. When only a LOAEL is available, extra room is left for extrapolating down from a dose that already caused an effect, so the factor is usually larger. An IOM report from 2000 listed the factors used up to then, from 1 (great confidence in the NOAEL) to 36. The 36 was applied to adult vitamin E, because it required extrapolating from animal data to humans and from a LOAEL to a NOAEL. That was only the range at the time; the factor follows the evidence and is not a fixed rule.
iom-2000-dri-applications
Mechanism · Limits that count only some sources
Not every limit counts every source. The vitamin A limit counts only preformed vitamin A (retinol and its esters). The beta-carotene in carrots is left out, because large amounts of beta-carotene have not been shown to cause vitamin A excess.The folate limit counts only synthetic folic acid from fortified foods and supplements, not the folate that occurs naturally in food. The reasoning is that in people short of vitamin B12, large amounts of folic acid may hide the anemia, delay diagnosis, and possibly trigger or worsen nerve damage. The IOM itself calls this evidence limited and mainly drawn from case reports.
The magnesium limit counts only supplements and magnesium-containing medicines, not magnesium from food and water. It guards against the diarrhea that these sources can cause.
iom-1997-mg
Evidence · Most people over the limit take supplements
In US National Health and Nutrition Examination Survey (NHANES) data from 2003 to 2006, among adults who took supplements, counting food and supplements together, about 7.0% went over the limit for synthetic folic acid, about 3.5% for vitamin B6, about 3.0% for preformed vitamin A, and about 1.6% for vitamin C. Counting food alone, none of the same people went over. Minerals show the same pattern. People using mineral-containing supplements went over the limits for calcium, zinc, iron and magnesium more often, while fewer than 0.5% went over from food alone. In the same data, even with supplements, more than 14% of adults still fell short of the average requirement for calcium and magnesium. Looking at whole age groups in 2009 to 2012, supplements raised the share above the limit for several nutrients, mostly to under 5%. The exception was calcium in the 51 to 70 age group, at 9.0%; this analysis was funded by supplement companies.These are all cross-sectional surveys, with intake based on what people recalled. Going over a limit means risk may be higher; it does not show that anyone was harmed.
bailey-2011-mineral-supplement-ulblumberg-2017-supplements-adequacy-age
Evidence · Children show the same pattern
Among 7250 children aged 2 to 18 in NHANES 2003 to 2006, taking supplements made intake above the limit more likely for iron, zinc, copper, selenium, synthetic folic acid, and vitamins A and C. The other half of the same study's findings: children over 8 filled gaps in magnesium, phosphorus and vitamins A, C and E with supplements, and even with supplements more than a third of children did not get enough calcium and vitamin D.A Danish survey from 2011 to 2013 covered 499 children aged 4 to 10; 64% took supplements, mostly multivitamin and mineral products. Among the children taking supplements, 21% to 73% went over the limits for preformed vitamin A, zinc and iodine, and 6% to 45% went over a Danish provisional guidance level for iron. Yet from diet alone, 12% to 30% of children were already over the limit for vitamin A, zinc or iodine. The authors state that the long-term health effects of this excess in children are unknown, and that excess zinc did not appear to be linked with harm.
bailey-2012-supplements-children-sufficiencychristensen-2023-danish-children-supplements-ul
In practice · How stacking pushes intake past the limit
The limit counts the total from all sources, and in the US adult survey, nearly everyone over the limit for these vitamins was taking supplements. So one nutrient can arrive from several places at once: a multivitamin, a single-nutrient product, a few fortified foods. Each looks modest on its own, but together they can cross the line. Among the adverse event reports the TGA received were people who developed peripheral neuropathy while taking several products that each contained vitamin B6.The number on the label is not always what is in the bottle, either. A case report describes 7 young children who took a wrongly manufactured fish oil supplement containing about 4000 times the labeled vitamin . All developed vitamin D toxicity with high blood calcium, and all recovered after treatment. Accidents like this are rare. They show a failure of quality control, not that taking a little more than a normal dose leads to poisoning.
So when you read an upper limit, two questions matter more than the number itself: does this nutrient stay in the body, and how does it leave; and what is the total from every source added together.
Safety · People the upper limit does not cover
Upper limits are set for the general population. They do not apply to people taking a nutrient under medical supervision, and some limits leave out particular groups: the vitamin E limit, for example, does not apply to people on anticoagulant or antiplatelet drugs.References · 29
- Levine, M., et al. (1996). Vitamin C pharmacokinetics in healthy volunteers: evidence for a recommended dietary allowance. PNAS, 93(8), 3704–3709. 10.1073/pnas.93.8.3704
- Blaner, W. S., O'Byrne, S. M., Wongsiriroj, N., Kluwe, J., D'Ambrosio, D. M., Jiang, H., Schwabe, R. F., Hillman, E. M. C., Piantedosi, R., & Libien, J. (2009). Hepatic stellate cell lipid droplets: A specialized lipid droplet for retinoid storage. Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids, 1791(6), 467-473. 10.1016/j.bbalip.2008.11.001
- National Institute of Diabetes and Digestive and Kidney Diseases. (2020). Vitamin A. In LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. Bethesda, MD: NIDDK. Bookshelf ID NBK548165. www.ncbi.nlm.nih.gov/books/NBK548165
- Ganz, T. (2013). Systemic iron homeostasis. Physiological Reviews, 93(4), 1721–1741. 10.1152/physrev.00008.2013
- EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA), Turck, D., Bohn, T., Castenmiller, J., de Henauw, S., Hirsch-Ernst, K.-I., Knutsen, H. K., Maciuk, A., Mangelsdorf, I., McArdle, H. J., Pentieva, K., Siani, A., Thies, F., Tsabouri, S., Vinceti, M., Aggett, P., Fairweather-Tait, S., de Sesmaisons Lecarré, A., Fabiani, L., … Naska, A. (2024). Scientific opinion on the tolerable upper intake level for iron. EFSA Journal, 22(6), e8819. 10.2903/j.efsa.2024.8819
- Penniston, K. L., & Tanumihardjo, S. A. (2006). The acute and chronic toxic effects of vitamin A. The American Journal of Clinical Nutrition, 83(2), 191–201. https://doi.org/10.1093/ajcn/83.2.191 10.1093/ajcn/83.2.191
- Institute of Medicine. (2001). Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academies Press. www.ncbi.nlm.nih.gov/books/NBK222310
- EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA). (2024). Scientific opinion on the tolerable upper intake level for preformed vitamin A and β-carotene. EFSA Journal, 22(6), e8814. 10.2903/j.efsa.2024.8814
- Lobo, G. P., Hessel, S., Eichinger, A., Noy, N., Moise, A. R., Wyss, A., Palczewski, K., & von Lintig, J. (2010). ISX is a retinoic acid-sensitive gatekeeper that controls intestinal β,β-carotene absorption and vitamin A production. The FASEB Journal, 24(6), 1656–1666. 10.1096/fj.09-150995
- Araki, T., Holick, M. F., Alfonso, B. D., Charlap, E., Romero, C. M., Rizk, D., & Newman, L. G. (2011). Vitamin D intoxication with severe hypercalcemia due to manufacturing and labeling errors of two dietary supplements made in the United States. The Journal of Clinical Endocrinology & Metabolism, 96(12), 3603–3608. https://doi.org/10.1210/jc.2011-1443 10.1210/jc.2011-1443
- EFSA Panel on Nutrition, Novel Foods and Food Allergens. (2023). Scientific opinion on the tolerable upper intake level for vitamin D, including the derivation of a conversion factor for calcidiol monohydrate. EFSA Journal, 21(8), e08145. 10.2903/j.efsa.2023.8145
- EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA), Turck, D., Bohn, T., et al. (2024). Guidance for establishing and applying tolerable upper intake levels for vitamins and essential minerals. EFSA Journal, 22(11), e9052. 10.2903/j.efsa.2024.9052
- Institute of Medicine. (2011). Dietary Reference Intakes for Calcium and Vitamin D. National Academies Press. www.ncbi.nlm.nih.gov/books/NBK56070
- Burt, L. A., Billington, E. O., Rose, M. S., Raymond, D. A., Hanley, D. A., & Boyd, S. K. (2019). Effect of high-dose vitamin D supplementation on volumetric bone density and bone strength: A randomized clinical trial. JAMA, 322(8), 736–745. 10.1001/jama.2019.11889
- Sanders, K. M., Stuart, A. L., Williamson, E. J., Simpson, J. A., Kotowicz, M. A., Young, D., et al. (2010). Annual High-Dose Oral Vitamin D and Falls and Fractures in Older Women. JAMA, 303(18), 1815. 10.1001/jama.2010.594
- Kara, C., Gunindi, F., Ustyol, A., & Aydin, M. (2014). Vitamin D intoxication due to an erroneously manufactured dietary supplement in seven children. Pediatrics, 133(1), e240–e244. https://doi.org/10.1542/peds.2013-0711 10.1542/peds.2013-0711
- Jones, G. (2008). Pharmacokinetics of vitamin D toxicity. American Journal of Clinical Nutrition, 88(2), 582S-586S. 10.1093/ajcn/88.2.582S
- Nemeth, E., Tuttle, M. S., Powelson, J., Vaughn, M. B., Donovan, A., Ward, D. M., Ganz, T., & Kaplan, J. (2004). Hepcidin regulates cellular iron efflux by binding to ferroportin and inducing its internalization. Science, 306(5704), 2090–2093. 10.1126/science.1104742
- Moretti, D., Goede, J. S., Zeder, C., et al. (2015). Oral iron supplements increase hepcidin and decrease iron absorption from daily or twice-daily doses in iron-depleted young women. Blood, 126(17), 1981–1989. 10.1182/blood-2015-05-642223
- Stoffel, N. U., Cercamondi, C. I., Brittenham, G., Zeder, C., Geurts-Moespot, A. J., Swinkels, D. W., et al. (2017). Iron absorption from oral iron supplements given on consecutive versus alternate days and as single morning doses versus twice-daily split dosing in iron-depleted women: two open-label, randomised controlled trials. The Lancet Haematology, 4(11), e524-e533. 10.1016/S2352-3026(17)30182-5
- National Health and Family Planning Commission of the People's Republic of China. (2017). Chinese dietary reference intakes - Part 3: Trace elements (WS/T 578.3-2017). Health industry standard, issued 2017-09-14, effective 2018-04-01; drafted by the Chinese Nutrition Society with the National Institute for Nutrition and Health, China CDC. www.nhc.gov.cn/ewebeditor/uploadfile/2017/10/20171017153105952.pdf
- Adams, P. C., Reboussin, D. M., Barton, J. C., McLaren, C. E., Eckfeldt, J. H., McLaren, G. D., Dawkins, F. W., Acton, R. T., Harris, E. L., Gordeuk, V. R., Leiendecker-Foster, C., Speechley, M., Snively, B. M., Holup, J. L., Thomson, E., & Sholinsky, P. (2005). Hemochromatosis and iron-overload screening in a racially diverse population. The New England Journal of Medicine, 352(17), 1769–1778. 10.1056/NEJMoa041534
- U.S. Food and Drug Administration. (1997). Iron-containing supplements and drugs: Label warning statements and unit-dose packaging requirements (Final rule; Docket Nos. 91P-0186 and 93P-0306). Federal Register, 62(10), 2218–2250. www.govinfo.gov/content/pkg/FR-1997-01-15/html/97-947.htm
- Schaumburg, H., Kaplan, J., Windebank, A., et al. (1983). Sensory neuropathy from pyridoxine abuse: a new megavitamin syndrome. The New England Journal of Medicine, 309(8), 445-448. 10.1056/NEJM198308253090801
- Vrolijk, M. F., Opperhuizen, A., Jansen, E. H. J. M., Hageman, G. J., Bast, A., & Haenen, G. R. M. M. (2017). The vitamin B6 paradox: supplementation with high concentrations of pyridoxine leads to decreased vitamin B6 function. Toxicology in Vitro, 44, 206-212. 10.1016/j.tiv.2017.07.009
- Institute of Medicine. (1998). Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National Academies Press. www.ncbi.nlm.nih.gov/books/NBK114310
- Therapeutic Goods Administration. (2022, November 10). Health supplements containing vitamin B6 can cause peripheral neuropathy [Safety alert]. Australian Government Department of Health and Aged Care. web.archive.org/web/20250803071635/https://www.tga.gov.au/news/safety-alerts/health-supplements-containing-vitamin-b6-can-cause-peripheral-neuropathy
- Bailey, R. L., Fulgoni, V. L., III, Keast, D. R., & Dwyer, J. T. (2012). Examination of vitamin intakes among US adults by dietary supplement use. Journal of the Academy of Nutrition and Dietetics, 112(5), 657–663.e4. https://doi.org/10.1016/j.jand.2012.01.026 10.1016/j.jand.2012.01.026
- Institute of Medicine. (1998). Dietary Reference Intakes: A Risk Assessment Model for Establishing Upper Intake Levels for Nutrients. National Academies Press. 10.17226/6432
Written from the sources listed above. No physician reviews this site: it explains how the body works, and it is not medical advice. How we make content
Spotted an error, or something unclear? Email us. We read every message, check it against the sources, and fix what's wrong.