I didn’t like the description of “rusting” so I asked the question of Claude as to what proportion of the body is Iron Oxide or Hydroxide and got an interesting answer:
Claude:
The human body contains very little iron oxide or iron hydroxide in the “rust” sense — but these compounds do appear in specific biological contexts.
Total body iron context
A typical adult has 3–5 g of total iron, distributed as haemoglobin (~65%), myoglobin (~10%), storage iron ferritin/haemosiderin (~20–25%), and enzymes/other (~5%).
Iron oxide (Fe₃O₄ — magnetite)
The most biologically relevant iron oxide is magnetite (Fe₃O₄), not haematite (Fe₂O₃):
- Found in the human brain — estimates range from roughly 5–100 ng per gram of wet tissue in various brain regions, with the cerebellum and brainstem tending to have higher concentrations
- Total brain magnetite content is in the order of micrograms — vanishingly small
- Also reported in the ethmoid sinus region (near the cribriform plate), tentatively linked to magnetoreception research, though this remains controversial
- Found in magnetosomes within certain tissues, possibly of biogenic origin
Pathologically, iron oxide accumulation increases in neurodegenerative diseases (Alzheimer’s, Parkinson’s), where it has been detected in amyloid plaques and is thought to contribute to oxidative stress.
Iron hydroxide / oxyhydroxide (FeOOH)
This is more metabolically central:
- The iron core of ferritin consists largely of ferrihydrite (a hydrated iron oxyhydroxide, approximating Fe₅HO₈·4H₂O), sometimes described loosely as iron(III) hydroxide
- Each ferritin shell can store up to ~4,500 iron atoms as this mineral core
- Given the body’s ~1 g of storage iron in ferritin, this represents a meaningful fraction of body iron in oxyhydroxide form — potentially hundreds of milligrams of FeOOH-equivalent mineral
-
Haemosiderin (degraded ferritin aggregates) also contains amorphous iron oxyhydroxide phases
Summary table
| Compound |
Form |
Estimated quantity |
| Magnetite (Fe₃O₄) |
Brain, ethmoid |
~micrograms total |
| Ferrihydrite/FeOOH |
Ferritin cores |
potentially ~100s mg (iron equivalent) |
| Haematite (Fe₂O₃) |
Not normally present |
negligible |
The ferritin core chemistry is particularly relevant to your work on the SLC25A1→AcCoA→epigenetics axis, since ferritin iron mobilisation is redox-sensitive and intersects with mitochondrial ROS production and the labile iron pool that feeds into Fenton chemistry.
I didn’t want to watch it so I ran the transcript into chatGPT:
Here is a cleaned-up version of the transcript, followed by a summary and critique.
Tidy transcript
Mike:
Recently I had an epiphany that the population is rusting from the inside out. This is leading to increased insulin resistance, cancer, and other diseases. This is affecting about half of the US population, and the Western population in general.
We are talking about an incredibly similar chemical reaction happening inside of us as rust. Whether we are talking about your car or your garden tools, we are not made of metal, but we do of course have some metal in our body, and it is creating a similar result.
We are going to learn about this really important biochemical process, the research on diseases associated with it, populations that have five- to ten-fold greater risk of this happening, and potential solutions.
I just want to mention that I have launched my Greece trip for September 3 to September 10. It is going to be another super delicious adventure. You can click the link below and learn more about that. But let’s get into the main video.
Many of you already know what I am talking about. Others are thoroughly annoyed with me. I am talking about how iron oxidizes and does some oxidative damage. That is the process of rust. We just see red flaky stuff, but on a molecular level, that is what is occurring. In that case it is literally iron and oxygen reacting, pulling an electron out of that iron and creating iron oxide.
Interestingly, ferritin, which is our form of iron storage in the body, is essentially an iron oxide with a bunch of other elements associated with it.
Our own human “rusting” process can happen in a couple of specific ways. One is the buildup from excess iron of a denatured protein called hemosiderin. This is oxidative and is again a complex involving iron oxide. The condition of high iron levels in the body is called hemochromatosis.
We have descriptions in the literature that say the liver is enlarged and has a distinctive chocolate-brown or rust-red color due to massive accumulation of hemosiderin. The pancreas is also intensely pigmented, showing a dark red to brown discoloration, and so on.
Then we have a second process analogous to human rusting, and that is the Fenton reaction—not to be confused with the fentanyl reaction, which of course we have plenty of in the US. Like how that drug is much more aggressive than many other opioids, this is more aggressive than ordinary rust because it happens faster and creates a byproduct that is even more destructive.
The Fenton reaction occurs when iron and hydrogen peroxide come into contact. It can get out of control with high iron, and hydrogen peroxide is created naturally in the body as a result of metabolism. Things like inflammation and aging can lead to excess levels. When the Fenton process occurs, we get hydroxyl radicals, which are among the most powerful oxidants in the body.
So yes, it is essentially “super-rust,” and that hydroxyl radical can rip through DNA and proteins and lead to lipid peroxidation, which drives atherosclerosis and heart disease, our leading killer.
At this point people might say, “That is hemochromatosis. Those are high levels even for a Western population. I do not have that, so I do not need to worry.” But we need to clarify that a lot of research shows health issues at iron levels that are not considered hemochromatosis, especially in relation to disease risk.
This has to do with the Western diet and other factors. We may have reached a point where our iron levels as a species are higher than they have ever been. Our reference range was essentially created based on what was normal in the population, but is that really the healthiest place to be? There is even a study suggesting that ambient or normal Western levels of iron may be a noxious public health threat affecting large swaths of the population.
In other words, the population is rusting from the inside, like I am talking about here.
One major inspiration for this video was a recent series by Dr. Greger of NutritionFacts.org. He talks about normal ferritin ranges. The ferritin blood test measures iron stores, and a typical normal value is in the 30 to 300 range. Almost universally, vegetarians have lower ferritin levels than non-vegetarians—for example, 39 versus 77. But lower iron stores may actually be better, with a possible target between 15, which is the World Health Organization cutoff for iron deficiency, and 50, because levels above 50 are associated with disease. In that case, vegetarians may be in the sweet spot because iron is a double-edged sword.
I should mention that hemochromatosis is generally over 300 ferritin for men and 200 for women, so much higher.
I had to look into this because 50 seemed pretty low as a risk threshold. Looking at the data on vascular risk, there is a massive increase in risk above 50, but that category can include everyone over 100, 150, and 200. So we do not know exactly where the cutoff for substantially higher risk is until we look at a line graph showing a more linear rise in atherosclerosis risk as ferritin levels increase.
This brings us back to the whole vampire-like history of bloodletting and using leeches to treat disease, especially in overfed royals. Most ironically, this is one of those medieval practices that actually appears to hold up. And I just want to say: leeches—would you consider them vegan? You are just letting them eat you. Comment below.
But yes, studies of phlebotomy or bloodletting show shockingly positive results. For example, we have an older study out of Finland, one of the first to find a strong association. It is the Kuopio Heart Disease Risk study, looking at nearly 3,000 men over nine years.
People who give blood are generally trying to help others, and this is where I think, “Is karma real?” Because after adjustments for other risk factors, blood donors had an 88% reduced risk of heart attacks compared with non-blood donors.
The study is older, and maybe the adjustments were not perfect, but the results are still absolutely insane. The authors cite body iron stores, saying that because high body iron stores have been suggested as a risk factor for acute myocardial infarction, donation of blood could theoretically reduce the risk by lowering body iron stores.
With heart attacks especially, we are talking about arterial lesions where the cap can rupture and clog the artery. Here we have iron oxidation contributing to plaque lesions on the artery wall—sort of like rust spots. I know they are not literally rust, but the parallel is there. It is a bit like rust flakes clogging pipes.
There is also a newer study from the same cohort looking at mortality. Both animal-to-plant protein ratio, as a proxy for animal-to-plant iron intake, and total meat intake were associated with higher mortality. Why? Because heme iron, or animal iron, is a major source of iron in the Western diet. In the same cohort, elevated ferritin levels were associated with higher heart attack risk directly as well.
Another important factor is absorption limits. When we eat plant-based iron, the body knows how to reduce uptake—basically, “that is enough for today.” With heme iron, we do not have the same mechanisms in place. Maybe it is because we also have heme iron in our own body. Regardless, people who already have adequate iron and continue eating heme iron can drive their iron levels higher and higher.
Heme iron in particular may overload the transferrin system. Transferrin shields the oxidative effects of iron, but when intake is high, we can get non-transferrin-bound iron, which is primed for the dangerous Fenton reaction, as well as some direct free heme iron in the bloodstream.
There are more studies. One Dr. Greger mentioned that shocked me had to do with insulin resistance. Researchers looking at vegetarians found that they had lower iron levels and much lower insulin resistance. This mirrors the lower type 2 diabetes rates seen in vegans in populations such as EPIC-Oxford or Adventists, where the risk may be 50 to 70 percent lower.
The meat-eaters in the study had higher iron levels, and we are trained to think higher iron is always better. But here researchers asked whether the iron might be connected to insulin resistance. So they took some of the meat-eaters and did phlebotomy until they reached a vegetarian ferritin range. They went from about 70 to the mid-30s, and the result was a 40% drop in insulin resistance.
The iron-lowering trial using blood draws demonstrated that iron is indeed a key factor in the difference in insulin resistance between vegetarians and meat-eaters.
This is also seen in study after study where people with type 2 diabetes have higher serum ferritin. We can talk about causation versus reverse causation—inflammation can also raise ferritin—but that is worth noting too, especially because there are plausible mechanisms.
For example, iron deposits can interfere with both pancreas and liver function in relation to insulin. The pancreas is particularly prone to iron-induced oxidation. It has beta cells that produce insulin, and those can be damaged or killed by iron-induced oxidation, leading to less insulin and poorer blood sugar regulation.
Excess iron in the liver can also short-circuit the off switch for glucose production, meaning more insulin is required because the liver is malfunctioning. So there is a one-two punch affecting internal organs that could worsen insulin resistance.
Then there is another study Dr. Greger discussed. He says: too bad we could not do a blood donation study with cancer—except we can. More than a thousand cancer-free individuals were randomized to give blood donations or not and followed for the next four and a half years. Those who gave blood had a 60% lower risk of dying from cancer.
Wait—they cut the risk of dying from cancer by more than half just by donating blood? That is a massive reduction in cancer death just from donating blood. Again, another shocking result.
Of course, we have to ask whether this is credible and what mechanisms there could be. It appears that iron may feed cancer in a few ways, both directly and indirectly. First, iron-driven oxidation can lead to DNA damage and mutations that initiate cancer. Then there is the fact that cancer cells often express far more iron-uptake receptors on their surface than normal cells, so they love to absorb iron.
Additionally, excess iron can affect immune cells, especially macrophages. It can shift M1 macrophages, which target cancer, toward M2 macrophages, which may suppress immune responses. That is not a good thing.
At this point I want to add a couple more points about bloodletting. Is it all iron? I believe iron plays a huge role, but it is also worth noting that phlebotomy in general can lower blood pressure, blood glucose, and LDL cholesterol. We know LDL is causally involved in atherosclerosis from genetic studies, although LDL and glucose did not fall in all cases, such as one study in people with hemochromatosis.
Still, because we do not have a heme-iron absorption cap and Western populations are getting large amounts of iron from animal foods, I cannot help but think we are eating too many animals—to the point that we benefit from literally draining our own blood of the animal-based iron and fats we have accumulated. That appears to massively lower heart attack risk, insulin resistance, and cancer risk. Clearly we are doing something wrong. We have too much blood in us.
So how many people are actually too high? I would say 75 is a decent risk cutoff. That is where higher insulin resistance shows up. Looking at NHANES US data, the median ferritin for men is 99, and for women it is 40. For postmenopausal women it is around 65.
So it is fair to say that about half of the population is over 75, meaning about half the population is rusting inside to a degree that will increase disease risk.
Finally, there is a population that has it much worse in terms of hemochromatosis, and that is Irish people. Whether because of surviving the potato famine or generally surviving a harsh environment with low iron availability, they may have evolved to use iron more efficiently. This means that when they do get enough iron, they are much more likely to get hemochromatosis.
Certain people in northwestern Ireland are seven times more likely than the general population to get it. This is due to a mutation affecting the brake on our iron system—hepcidin, which I talked about in my “vegans have increased plant iron absorption” video. Hepcidin directly blocks iron absorption, and these individuals do not have enough of it.
White Europeans are highest on average. Indigenous North Americans are also quite high, and then it declines from there.
A small side tangent: the thiamine-IBS video I made is also related. People of Irish descent are higher in that area as well, again probably because of historically low nutrient availability in their environment. If you want to learn about thiamine, digestion, and dysautonomia, check out my thiamine video.
In the end, it is obviously a smarter choice to rely on plant iron in a world where many diseases are driven or worsened by excess iron. Again, we cannot control heme-iron absorption in the same way. We may have around 150 million people with ferritin above 75, which lines up fairly well with insulin resistance numbers.
So whether it is iron in animal products alone, or the animal fat as well, bloodletting appears to be literally saving lives—creating an 88% lower heart attack risk or a 60% lower cancer mortality risk. Those numbers are mind-boggling. I think we should study this more and really examine that cutoff.
Are there maybe negatives to being between 15 and 30 ferritin for things like hair or other issues? I want to see all the research.
Let me know what you think in the comments. Have you had your iron checked? Do you have iron issues? Do you think my assessment that the population is rusting is accurate? Feel free to like, subscribe, and all that good stuff. Check out my Greece trip linked below if you are interested. Thanks for watching. I will see you in the next one.
Summary
The speaker argues that excess iron in modern Western populations acts like a kind of internal “rusting,” driving oxidative damage, insulin resistance, cardiovascular disease, and cancer. He uses the analogy of rust to describe iron oxidation and highlights two related processes: iron accumulation in tissues, particularly as hemosiderin in overt iron overload states such as hemochromatosis, and the Fenton reaction, where iron and hydrogen peroxide generate highly reactive hydroxyl radicals.
He then extends the argument beyond overt hemochromatosis. His main claim is that disease risk may rise well below classical iron-overload thresholds, especially at ferritin levels that are still considered normal in standard laboratory ranges. He suggests that modern reference ranges reflect what is common, not necessarily what is optimal.
The video links lower ferritin, especially in vegetarians and vegans, with better metabolic health. The speaker highlights studies suggesting that blood donation or phlebotomy lowers heart attack risk, insulin resistance, and possibly cancer mortality. He interprets this as support for the idea that excess iron is a widespread and underappreciated driver of chronic disease.
A major dietary theme is that heme iron from animal foods is more dangerous than plant iron because it is more readily absorbed and less tightly regulated. He argues that Western diets push ferritin into a harmful range and that plant-based eating may protect partly by lowering body iron stores.
He also discusses high-risk populations, especially people of Irish ancestry, because of hereditary hemochromatosis-related variants affecting iron regulation. He closes by suggesting that a substantial share of the population may have ferritin levels high enough to increase disease risk and that this possibility deserves much more study.
What is novel in the video
The core biology is not novel. The links between iron, oxidative stress, hemochromatosis, the Fenton reaction, and tissue damage are well known. The relationship between iron overload and risk in certain diseases is also established.
What is more distinctive is the framing and synthesis:
-
“Rusting from the inside” as a unifying metaphor: This is rhetorically strong and helps connect iron chemistry, oxidative stress, atherosclerosis, diabetes, and cancer into a single narrative.
-
The claim that “normal” Western ferritin may often be too high: That is the main provocative thesis.
-
The emphasis on blood donation/phlebotomy as a possible broad public-health intervention rather than just a treatment for hemochromatosis: That is presented much more strongly than is typical.
-
The interpretation of vegetarian/vegan metabolic advantages partly through iron burden rather than only through fat, weight, fiber, or calorie density: This is an interesting angle, though not demonstrated as the dominant explanation.
So the novelty is mainly interpretive, not mechanistic.
Critique
Strengths
The talk does a good job explaining why iron is biologically double-edged. Iron is essential, but its redox activity can become harmful when poorly controlled. The speaker is right that ferritin is not simply “the higher the better,” and he is also right that overt iron overload can damage the liver, pancreas, and heart.
The video is also strongest when it points out that “normal range” does not necessarily mean “optimal range.” That is an important general principle in medicine.
Another strength is that the speaker does not rely only on abstract chemistry; he tries to connect mechanisms to epidemiology and intervention studies, especially phlebotomy trials and observational work on ferritin and disease risk.
Weaknesses and limitations
The biggest problem is overreach. The video moves too quickly from:
- iron can promote oxidative damage,
- high iron overload states are harmful,
- some studies associate higher ferritin with worse outcomes,
to the much broader conclusion that roughly half the population is materially “rusting” in a clinically important way because ferritin is above about 75.
That conclusion is much stronger than the evidence presented in the transcript.
Ferritin is not a pure iron-overload marker
A major issue is that ferritin is also an acute-phase reactant. It rises with inflammation, infection, liver disease, metabolic syndrome, alcohol use, and other conditions. So when ferritin correlates with diabetes, cardiovascular risk, or cancer outcomes, ferritin may partly be:
- a marker of iron stores,
- a marker of inflammation,
- a marker of liver injury,
- or all three at once.
The speaker acknowledges reverse causation briefly, but he does not really grapple with how much of the association may be due to confounding.
Intervention evidence is intriguing but not definitive
The strongest part of the case would be randomized trials of iron reduction with hard clinical endpoints. The transcript mentions striking effects of blood donation/phlebotomy, but those effect sizes are so large that they invite caution. Very large benefits in observational studies can reflect healthy-donor bias, selection effects, or residual confounding. Even randomized evidence needs careful scrutiny around study design, endpoint definition, sample size, and replication.
So while the intervention signal is interesting, the video presents it with more confidence than seems warranted.
Mechanistic plausibility does not prove population-level causality
The mechanism is plausible: iron can catalyze oxidative chemistry, damage beta cells, worsen liver insulin handling, and potentially support tumour biology. But mechanistic plausibility is not enough to show that iron is a major driver of common chronic disease in the general population at ordinary ferritin ranges.
The jump from “possible mechanism” to “dominant public-health explanation” is not justified in the transcript.
Overemphasis on heme iron as the central explanation
The video strongly favors the view that animal-derived heme iron is a main driver of excess body iron and disease. That may be part of the story, but the transcript underplays other major determinants:
- total energy excess,
- adiposity,
- inflammation,
- genetics,
- alcohol,
- liver status,
- age,
- sex and menopausal status,
- insulin resistance itself.
In other words, iron may be one contributor without being the master variable.
The “half the population” claim is too assertive
The argument that half the population is at increased risk because half may be above ferritin 75 is not adequately supported here. Even if ferritin above 75 correlates with some outcomes in some cohorts, that does not establish a universal pathological threshold across sexes, ages, ethnic groups, and clinical contexts.
That is probably the weakest inferential step in the talk.
Some statements are oversimplified or slightly muddled
A few examples:
- Ferritin is described in a somewhat loose way as essentially an iron oxide storage form. That is directionally understandable for a lay audience, but biochemically it is more specific and regulated than the metaphor suggests.
- The hepcidin discussion is somewhat simplified and may confuse regulatory pathways with particular population histories.
- The evolutionary claims about Irish ancestry and famine adaptation are presented too confidently relative to the evidence shown in the transcript.
- “We have too much blood in us” is rhetorically memorable but biologically imprecise.
Rhetoric sometimes outpaces nuance
The talk is engaging, but phrases like:
- “the population is rusting,”
- “super-rust,”
- “we are eating too many animals to the point that draining our blood saves lives,”
make the argument vivid at the cost of precision. This can be useful for public communication, but it risks pushing a contested hypothesis as though it were near-settled.
Bottom line
This is a thought-provoking and mechanistically plausible argument that excess iron, even below classic hemochromatosis thresholds, may contribute to cardiometabolic disease and cancer risk more than is usually appreciated.
Its main value is as a hypothesis-generating synthesis:
- iron is biologically dangerous when poorly controlled,
- ferritin may deserve more nuanced interpretation,
- heme iron and iron reduction strategies may be worth more study.
But the video overstates the certainty and scale of the evidence. The transcript does not justify the strongest claims, especially that around half the population is pathologically “rusting” or that ferritin above 75 is a broadly applicable disease threshold.
A fair verdict would be:
Interesting, biologically plausible, and worth investigating further — but presented with substantially more confidence than the evidence in the transcript can support.
If you want, I can also turn this into a claim-by-claim table with columns for claim, evidence presented, strengths, weaknesses, and verdict.
