Mind & Body · Sunday, 30 August 2026

01 Briefing what happened

Your body has no way to throw iron out. It can only decide what to let in.

Mind & Body 1 min 55 sources

Iron carries the oxygen in your blood, and loose iron damages cells. There is no tap. About one to two milligrams leak away each day whatever happens, so the only control your body has is at the gut wall.

3-4 g

iron in an adult body

most of it inside red blood cells [5][4]

1-2 mg

leaves the body per day

sweat, shed cells and blood; the rate cannot be adjusted [3]

20 mg

recycled from old red cells daily

against 1 to 2 mg absorbed from food [3]

200-250 mg

iron in one bag of red cells

about 100 to 150 days of normal absorption, at once [9]

The lead story — what happened

An adult carries about three to four grams of iron. Most of it is inside the red blood cells that move oxygen. [5][4]
There is no way to get rid of it on purpose. Humans have no working route for pushing iron out of the body. [2]
About one to two milligrams leak away each day, in sweat, in shed gut and skin cells, and in blood. [3][5]
That leak is fixed. The body cannot turn it up when it holds too much, or turn it down when it is short. [2]
So the only control that can be adjusted is how much gets absorbed. Iron balance is held at the gut wall and nowhere else. [2]
Almost none of your iron comes from food. About 20 mg a day is recycled from worn-out red cells; only 1 to 2 mg is absorbed. [3]
A liver hormone called hepcidin works that control. It sticks to ferroportin, the doorway iron leaves a cell by, and destroys it. [1][3]
Ferroportin is the only known protein that moves iron out of a cell. Shut it and the iron stays where it is. [2][10]
Hepcidin also rises with inflammation. An infection shuts the gut door whether or not you need iron. [3][34]
Ferritin, the storage protein, is a hollow ball that packs iron away. That is containment, not removal. [54]
When storage runs out, iron circulates loose instead of bound to its carrier. Loose iron is the toxic form, and it can enter any organ. [2]
In hereditary haemochromatosis the hepcidin signal is faulty, so the gut keeps absorbing. About 1 in 150 people of northern European ancestry carry the main double gene change. [6][7]
Left unmanaged, that slow loading ends in cirrhosis, or permanent liver scarring, in about 70% of patients. About half of people with the double gene change develop diabetes. [6]
There is no cure. The standard treatment is drawing blood, because most of the body's iron sits inside red cells. [8]
One bag of donated red cells carries 200 to 250 mg of iron, roughly 100 to 150 days of ordinary absorption, delivered at once. [9]
Shortage is the far commoner problem. Iron deficiency is the most common mineral shortage in the world, affecting around two billion people. [25]

What is pushing on this

Shortage worldwide High→

about two billion people short of iron, and the 2025 anaemia targets were missed [25][12]

Loading with no exit Building↑

every transfusion adds 200 to 250 mg the body cannot remove [9]

Tests that argue with the body Building↑

inflammation lifts ferritin, so a normal reading can sit on top of a real shortage [34][3]

Exits we had to invent Steady↑

drawing blood and iron-grabbing drugs are the only ways out, and both are man-made [8][43]

Who is involved

The liver — makes hepcidin, the hormone that decides how much iron gets through [1] The gut lining — the one place where the amount taken in can still be changed [2] Macrophages — recycle about 20 mg of iron a day from worn-out red cells [3] Transfusion medicine — keeps people alive and loads them with iron at the same time [9]

How it unfolded

2.2 billion years ago oxygen fills the air, iron becomes hard to dissolve, and living things build systems to hoard it [4]
The 1970s before iron-removing drugs, children on regular transfusions lived to a median of about 16 [2]
The past decade the discovery of hepcidin as the master regulator of iron transforms how the whole field is understood [10]
2025 iron given by drip cuts heart-failure hospital stays, and a Lancet commission reports the global anaemia targets were missed [29][12]
Next drugs aimed at hepcidin itself are in development, for both loading and shortage [10]

Where this points

Watch the thresholds rather than the treatments. Several 2025 papers argue the ferritin cut-offs for calling someone iron deficient are set too low. Raising one of them doubled the share of young children counted as short. [18][15]

The rest of the day

38 more stories on this beat.

Each with its own sources. None of these is a link to the story above.

02
One in three UK women tested came up short

A retrospective look at 33,029 private health-check samples found anaemia in 6.0% of people overall, 9.9% of women, and outright iron deficiency in 31.6% of women. [14]

Why it matters — These are people well enough, and well off enough, to buy a health check on themselves.
03
Normal blood count, empty stores

In 100 non-pregnant women with a normal level of haemoglobin, the oxygen-carrying protein in blood, 41% were iron deficient on a ferritin test. History of breastfeeding raised the odds nearly sevenfold. [16]

Why it matters — The blood count is the test most people get, and it is the last thing to move.
04
The children's threshold may be set too low

Using a ferritin cut-off of 22 micrograms per litre, iron deficiency in children aged 6 to 59 months reached 34%, against 16% at the WHO figure. [18]

Why it matters — Nothing changed in any child; the definition changed, and the disease doubled.
05
Global anaemia targets were missed

A Lancet Haematology Commission puts anaemia at over 1.9 billion people and calls the 50% reduction target unachievable, proposing 12-22% instead. [12]

Why it matters — The target was set on ambition rather than on what the available tools can do.
06
Two-thirds of anaemia is iron

Dietary iron shortage accounts for roughly 66% of anaemia worldwide: an estimated 444 million cases in men and 825 million in women. [11]

Why it matters — It is a nutrient problem before it is a blood problem.
07
Daily beat alternate-day in one trial

Sixty iron-deficient women were given either 60 mg daily or 120 mg on alternate days for 14 days. Daily doses did better on red cell size and haemoglobin content. [20]

Why it matters — The popular alternate-day advice rests on hepcidin reasoning that this trial did not reproduce.
08
Supplements did not close the gap in pregnancy

In the PANDA trial, 135 non-anaemic pregnant women took iron daily, on alternate days or three times a week. At 26 to 30 weeks, 69% to 72% were still iron deficient. [21]

Why it matters — Pregnancy raises the demand faster than the gut can pass iron through.
09
Vitamin C works by plain chemistry

Ascorbic acid stops iron forming compounds the gut cannot take up, and converts iron to the form the gut transporter accepts. [23][24]

Why it matters — It changes what arrives at the door, not what the door decides.
10
Eggs, calcium and antacids get in the way

Some studies put the drop in iron availability from eggs at about 27% in adults. Calcium supplements, antacids and acid-blocking drugs also cut absorption. [26][22]

Why it matters — Two people eating the same iron can absorb very different amounts of it.
11
Repeat blood donors run low

Among 252 donors at a teaching hospital in Tamale, Ghana, 40.1% had ferritin under 30 ng/mL and 14.7% were iron deficient. Low stores were commoner in repeat donors. [27]

Why it matters — Donation is the body's only real exit, and using it has a price attached.
12
A drip skips the door entirely

Iron given into a vein is taken up completely, bypassing the gut control. A network meta-analysis of 13 trials found it cut heart-failure hospital admissions. [18][30]

Why it matters — The one regulated step in the whole system can simply be gone around.
13
Iron by drip cut heart-failure returns

Across five trials and 6,038 patients, repeat hospital admission or death from heart disease fell by about a quarter in the year after ferric carboxymaltose, one of the iron drips. [29]

Why it matters — Iron deficiency affects between 37% and 75% of heart-failure patients, depending on the definition used. [28]
14
Before surgery, iron cut transfusions

Seven trials covering 975 patients found high-dose iron given by drip reduced the rate of red cell transfusion after cardiac surgery. [31]

Why it matters — A transfusion avoided is 200 to 250 mg of iron the body never has to hold.
15
Iron during infection did not backfire

Across 339,145 records, matched one to one, patients given iron by drip during an infection survived better at 14 and 90 days. The infections were MRSA in the blood, pneumonia, urinary infections and colitis. [32]

Why it matters — Bacteria need iron too, which is why withholding it during infection was the standing worry.
16
The definition changes the disease

In 149 patients with a disease that stiffens the heart with abnormal protein, iron deficiency ranged from 40.9% to 70.5% depending on which of three definitions was used. [33]

Why it matters — Only the two availability-based definitions predicted who died.
17
Inflammation makes the test lie

A review of 47 studies in inflammatory bowel disease found hepcidin rising alongside the two standard blood markers of inflammation, while iron and haemoglobin fell. [34]

Why it matters — The same signal that hides the shortage is the one causing it.
18
Short without being anaemic

Among 112 people with ulcerative colitis and no anaemia, 33.9% were iron deficient. Non-healed gut lining raised the odds eighteenfold. [35]

Why it matters — Anaemia is a late sign, and waiting for it means waiting years.
19
Teen runners fall through the normal range

A Cleveland Clinic clinician notes that everything flags normal between 15 and 200 ng/mL, while many clinicians aim for at least 50 in young athletes. [38]

Why it matters — A high-school cross-country girl runs 30 to 35 miles a week, and a Division I collegiate runner up to 100.
20
Iron fixed the fatigue, not the fitness

In 23 young women with iron-deficiency anaemia, eight weeks of iron improved fatigue scores and muscle endurance. Aerobic fitness and muscle strength did not move. [39]

Why it matters — It is a small uncontrolled trial, and it still separates two claims usually sold together.
21
One iron-removing pill did not move the iron

Across 23 randomised trials and 1,005 patients, deferiprone improved heart pumping. It did not significantly change ferritin, liver iron or the heart scan, and side effects were more common. [43]

Why it matters — The drug exists to install the exit the body lacks, so what it removes is the whole point.
22
Iron-removing drugs helped hormones, cost hearing

Of 60 children on oral iron-removing drugs, 16.6% had some hearing loss. In a Turkish cohort, hormone problems fell from 83% to 25.8% over about six years of treatment. [44][45]

Why it matters — Both numbers describe the same trade: damage prevented against damage caused.
23
Brain iron has its own family of diseases

The US neurological institute lists ten disorders in which iron builds up in the brain. One is caused by a faulty version of a protein whose job is clearing extra iron out of cells. [42]

Why it matters — When the local clearing fails, the accumulation is in a place nothing can be drained from.
24
Ferritin doubles as a warning light

Among 200 patients with cirrhosis, 68.5% of those with raised ferritin had a complication within a year, against 27.7% of those with normal levels. [47]

Why it matters — Ferritin is being read less as a fuel gauge and more as a stress signal.
25
Mice were given hepcidin through the gut

Researchers engineered food-grade bacteria to release hepcidin straight into the intestine. It cut the gut's iron transporter and countered overload in mice. [49]

Why it matters — It is an attempt to work the door from the outside rather than replace the hormone in the blood.
26
In India, iron was behind almost all of it

In records of anaemic women aged 15 to 49 across Indian regions, iron deficiency was the leading cause in 95.8% of cases; 9.4% were severe. [51]

Why it matters — Where the food supply is thin, the door is not the limiting step.
27
A quarter of the world is anaemic

The 2021 global burden estimate puts anaemia at 24.8% of the world's population, more than 1.6 billion people, with the heaviest load in under-fives and in women of reproductive age. [13]

Why it matters — The count keeps rising even as the rate per person falls, because there are more people.
28
School-age children were left out of the counting

A review of 55 studies covering 2.1 million children aged 5 to 12 found iron-deficiency anaemia in 9.4%, rising to 21.9% in sub-Saharan Africa and 29.7% in low-income countries. [17]

Why it matters — Programmes are aimed at pre-schoolers and women, so this group is measured last.
29
After menopause the requirement more than halves

The recommended intake for women is 18 mg a day up to age 50 and 8 mg from 51, against 27 mg in pregnancy. Iron-deficiency anaemia still affects up to 17% of Americans over 65. [19]

Why it matters — A gerontologist's line for it: after menopause a woman is not meant to be losing iron, so a falling level needs a reason.
30
Anaemia rides with tuberculosis

Anaemia is reported in 30% to 94% of tuberculosis patients. A study in Port Sudan put iron-deficiency anaemia at about 16% of them. [36]

Why it matters — Long infection keeps hepcidin high, so iron is locked away rather than absent.
31
Telling overload from inflammation in children

Researchers tested 238 children aged 6 to 72 months in southern Peru. They combined the ferritin reading with a second measure, how loaded the blood's iron carrier is, to tell overload apart from inflammation. [37]

Why it matters — Both push ferritin up, and the treatments point in opposite directions.
32
Iron-driven cell death enters the Alzheimer's argument

Ferroptosis, a form of cell death that runs on iron and collapses the cell's own antioxidant defence, is now a live research line in Alzheimer's disease. [41]

Why it matters — Iron accumulates in ageing brains, and nobody has yet shown which way the causation runs.
33
Pregnancy runs on its own numbers

Normal haemoglobin in pregnancy is 11 g/dL or above in the first and third trimesters, and 10.5 in the second. Ferritin under 30 ng/mL is the preferred marker of low stores. [50]

Why it matters — The thresholds move by trimester, which the general iron guidelines do not.
34
A newer red-cell measure catches the mixed cases

Among 672 anaemic patients at one Indian hospital, 38.2% had iron-deficiency anaemia and 41.2% of those also had inflammation running alongside it. [53]

Why it matters — That overlap is exactly where a ferritin number stops being readable.
35
Low iron and ADHD keep turning up together

A review found ADHD symptom severity higher in children with iron deficiency, and an association between the two, though ferritin values overlapped heavily and unmedicated children showed no difference. [40]

Why it matters — It is an association looking for a mechanism, not a finding.
36
Years of transfusion show up in the liver

Among 53 transfusion-dependent beta-thalassemia patients in Sana'a, iron levels were markedly higher in those transfused for more than five years or receiving over 250 mL at a time. [46]

Why it matters — Dose and duration, and nothing the body does, decide the total.
37
Loading happens without transfusions too

Fifty-six Indian children with a form of thalassemia that needs no regular transfusions had liver and heart iron scanned and compared with their ferritin readings. [52]

Why it matters — Ineffective red-cell production suppresses hepcidin, so the gut over-absorbs with no transfusion involved.
38
Stored iron tracks a worse outcome in heart disease

Among 3,424 patients with coronary disease and chronic heart failure, those above a ferritin of about 230 ng/mL did worse than those below it. [48]

Why it matters — High ferritin here is read as inflammation and damage, not as a full tank.
39
Patients rank the symptoms nobody measures

A scoping review of hereditary haemochromatosis found the standard quality-of-life questionnaire covers neither joint pain nor fatigue in detail, the two things patients raise first. [55]

Why it matters — A disease scored on liver enzymes is not scored on how it feels.

02 Lesson why it matters

The only choice you get is the first one

When a thing cannot be taken back out, all the care has to happen before it goes in, and everything after that is holding, not fixing.

The twist

When nothing can be taken back out, all the care has to happen before it goes in, and that judgement is made on what is already stored.

How it works

Iron carries oxygen, and loose iron burns cells
The body has no way to push it back out
Losses are fixed at one to two milligrams a day
So all the control moves to the moment of absorption
After that, all the body can do is wrap iron up
And wrapping has a limit that refusing does not

Where you've seen this

Nuclear waste

nothing destroys it, so every plan is about where to keep it and for how long

Lead paint in old houses

it is sealed over rather than taken out, and the seal is the part with a lifespan

Anything put online

a page comes down, but the decision that mattered was made before it went up

The catch

The pattern holds only while the exit is really shut. Medicine's answer to iron overload is to install the exit the body never had: a needle, or a molecule that carries iron away.

And the whole of it

Almost everything your body settled about iron today was settled before you tasted anything, by a hormone reading a total you have never seen. Most of us carry totals like that, in a body, in a house, in a stretch of ground. They were set at a door we were not standing at.

03 Truth what's really going on

What is really going on

Iron has no exit, so everything that looks like managing it is really one decision taken at the gut wall, over and over, for a whole lifetime.

Why it works on us — Iron is sold as something you top up, because a bottle of pills is easy to picture and a running lifetime total is not.

Who gains

Makers of iron given by drip — the 2023 European cardiology guideline recommends ferric carboxymaltose for symptomatic iron-deficient heart failure, which turns a drug into routine care. [28]
Supplement sellers — tiredness is the hallmark symptom and needs no test to self-diagnose, and a Harvard piece describes women over 60 taking iron just in case. [19]
Whoever sets the ferritin threshold — moving the cut-off to 22 micrograms per litre took childhood iron deficiency from 16% to 34% without one child changing. [18]
People with the haemochromatosis gene change in countries that screen — drawing blood before organs load prevents the cirrhosis that follows about 70% of unmanaged cases. [6][8]

Who pays

Women of reproductive age — losses are greater in menstruating women because of the blood, and 18.9% of 2,709 Swiss women were iron deficient. [5][15]
Repeat blood donors — donation is the body's only real exit, and 40.1% of 252 donors at one Ghanaian centre had ferritin under 30 ng/mL. [27]
Children on lifelong transfusions — each bag delivers 200 to 250 mg with no way out, and the drugs that remove it cost some of them hearing and hormone function. [9][44][45]
Anyone whose iron is tested during an infection — inflammation raises ferritin and hepcidin at once, so the number reads fine while the supply is short. [34][53]
School-age children in low-income countries — programmes are aimed at under-fives and at women, and 29.7% of 5-to-12-year-olds in low-income countries are anaemic from iron. [17]

What nobody knows yet

Open questions from across today’s stories — ours included.

01

Where the line for iron deficient actually sits.

The WHO ferritin cut-off gives 16% of children aged 6 to 59 months as deficient; a 2025 analysis using 22 micrograms per litre gives 34%. In Swiss women, haemoglobin only started falling below 28.5. [18][15]
02

Whether daily or alternate-day iron works better.

The hepcidin argument says alternate days should absorb more. A 60-woman trial found daily doses did better on red cell measures, and the pregnancy trial found no schedule prevented deficiency. [20][21]
03

How many people carrying the haemochromatosis gene change ever get ill.

About 1 in 150 people of northern European ancestry carry the main double change, and many never show it. Nobody can say in advance which ones will. [7][6]
04

Whether one of the iron-removing pills actually removes iron.

Across 23 randomised trials and 1,005 patients, deferiprone improved heart pumping but did not significantly change ferritin, liver iron or cardiac MRI. [43]
05

Whether iron drives brain disease or just collects in it.

Iron builds up in Alzheimer's and Parkinson's brains and iron-driven cell death is a live research line, but the direction of cause has not been shown. [41][42]
06

How much of the tiredness blamed on low iron is low iron.

In 23 young women, eight weeks of iron improved fatigue scores and muscle endurance but not aerobic fitness or strength. The trial had no control group. [39]
07

Whether giving iron during an infection helps or feeds it.

A study of 339,145 records found better survival with iron given by drip during infection. It was retrospective and matched, not a trial, and bacteria do need iron. [32]
08

How to read a ferritin number in someone who is inflamed.

Inflammation lifts ferritin and hepcidin together. A Peruvian study of 238 children needed a combined measure to separate overload from inflammation at all. [37][34]

04 Hope carry this

Your body never grew a way to let iron go. We built two in a single lifetime, and the children who once lived to about sixteen now grow up.