Since April 4th I’ve taken a break from editing interviews and have been working on both VeganLinked.com and a new sister site soon to be released. I’m finally feel like VeganLinked.com is getting to a point where I can use it! So, I wanted to test out posting with this update. The pictures are of a house in Asheville, NC I bought in May 2020 for a VeganLinked project doing videos with Dr. Garth Davis. I wanted a studio near him to do videos regularly. I fixed up the house to rent out so I could afford to do this. I put an accessary building on the property and turned it into a studio. Five videos in and Garth had to move back to Houston. So, I switched to doing interviews full throttle around the country. Sometimes I’ll take a break to work on the site hoping one day to have it right enough for the VeganLinked community to use as a tool. The Posts section of the site is starting fresh with just four category options: Recipes, Science, Stories, and Miscellaneous. I think these categories are a great start and I’m excited to see this section grow from there. I also started a forum on the site. I’m not exactly sure if people will want to interact there but if they do it’s ready for topics. I’m not exactly sure which forums and topics would be good there yet. As far as navigating the site, joining, adding listings & events, etc… the user experience should be significantly better than ever now. The login experience should be cleaner and now directs users to their profile. I hope this will encourage them complete their profiles with a picture, banner, and their vegan story. Stories in the blog will be an opportunity to share more elaborately with pictures, video, and offer a way for people to comment on your story. All types of stories may be shared here. And the stories you post here will link back to your profile where all your stories are archived. The same goes for anything you post here such as recipes, science, and misc.. Notifications seem to be working good now. So I’m a little more excited about actually using social features like “friends”, “follow”, and “messages”. I’m still testing thigs though. And I may have found better options for future upgrades if necessary. But this may be just all we need. The default categories show on the listings now in searches, archive pages, and the homepage. I’ve been wanting to display at least that there for many years now. So I’m really happy to have that finally. I think that will draw people into the site more deeper and more intuitively. I also added a much nicer navigation menu on mobile. It’s much more concise only showing a link for “listings & events” instead of having all five of those sections across the navigation. When you select Listings & Events you’re brought to Professionals with a search bar where you can change to the other sections. I had to get the filter to work on the search page before I could do that. So, that’s working good enough now to simplify things. I am trying to get all I can done before I start working on the homepage. I really want to minimalize it strategically. Once I finish wrapping up all the other functionality and flow behind the homepage I’ll start working on the homepage. And maybe then I’ll be done enough to run with it. It’s been over a month straight, long days, every day, working on this site and the new sister site. Soon I’ll have to get back to finishing editing and getting ready for the next tour of interviews for the channel. Manifesting this vision, one year at a time… Read more...

Ingredients

• 6 cups water
• 4 cups chopped green cabbage
• 1 cup sliced celery (about 2 large stalks)
• 1 cup sliced carrot (about 2 medium size)
• 1 cup chopped beet (about 1 medium or 2 smaller)
• 3/4 cup sliced mushrooms
• 1/2 cup diced red onion
• 1 bay leaf
• 1 tsp italian seasoning (or equal parts oregano, basil, marjoram, thyme, rosemary, sage)
• 1/2 tsp paprika
• 1/8 tsp black pepper
• bragg’s liquid aminos to taste (around 3 tbsp as a base)

Directions

wash beets carrots and celery well – cut ends off then chop into pieces around 1/4 inch thick, slice and dice the onion and mushrooms and chop up the cabbage while you bring the water to a boil add all the veggies and the bay leaf and bring to a boil again on high heat – lower heat to medium and cover, simmer for 15 minutes stirring occasionally, add the rest of the herbs and spices and stir again – cook another 10 minutes – add bragg’s, stir, cover, and let sit for 5 to 10 minutes to settle – remove the bay leaf when you find it 🙂 excellent served with millet, buckwheat or any other whole grain, plus lentils beans or tofu and a nice side salad – red cabbage can be used instead of green if you want to go super purple – it just makes a deep burgundy broth instead and has a more powerful flavor – show here served with a millet red lentil mix and a salad (this one features spring mix green pepper cucumber and avocado, topped with lemon juice braggs and dulse flakes) beets are excellent for growing young children too, so the black pepper can be omitted and less seasoning can be used overall for a milder version, mixing the soft sweet veggies with their favorite whole grains – as children in the 70’s and 80’s we were fed a well balanced whole food plant based / primarily macrobiotic diet of whole grains, plenty of raw and cooked vegetables, legumes fruit nuts and seeds, and simple dishes like these were our favorite!

a variation for a one pot meal

with buckwheat and lentils added

Ingredients

• 8 cups water
• 1 cup french green lentils
• 1 cup untoasted buckwheat groats
• 2 cups chopped beet
• 2 cups chopped cabbage
• 1 cup sliced carrot
• 1 cup sliced celery
• 1/2 cup diced onion
• 1/2 cup sliced shiitake mushrooms (or 1 cup white / cremini)
• 1 bay leaf
• 1 tsp italian seasoning
• 1/8 tsp black pepper
• bragg’s liquid aminos to taste

Directions

wash and chop all veggies, wash and rinse lentils and buckwheat and set aside, bring water to a boil, add lentils and buckwheat, then add all the veggies and spices, bring to a boil again on high heat then lower heat to medium, cover and simmer for 25 minutes, stirring occasionally – add a few tbsp of bragg’s at the end and stir well, remove from heat, then let sit covered for 5 minutes before enjoying Read more...

I just wanted to share this simple, base recipe real quick as I’m testing out posting to the site 🙂 I will come back and add the ingredients soon! For now they’re in the video and pictures. I love trying new cookie recipes. So, I’ll be sharing some other goodies I’ve and enjoy. It’s great to archive them here where there is no ads and I can share with others! Read more...

When I went vegan, 5 years ago,  I kept running into the same questions:

• Am I actually getting enough protein?

• What should I add to hit my goal?

I built Plant-Based Protein to make it easy—for me first, and hopefully for you too.

What’s inside:

• Protein Directory: protein content for hundreds of vegan foods (from tofu and tempeh to pea protein and lentils).

• Protein Tracker: a simple way to log your day and see where you’re at.

• Protein Target Calculator: your personalized daily protein goal (no guesswork).

• Protein Comparison: put two foods side-by-side to see which one gets you closer.

I created this because I was tired of overthinking meals and under-hitting my numbers. If you’re trying to eat more plants and still feel strong, I think this will help.

 

I would love you to check it out: www.plant-based-protein.com
Built by a vegan, for vegans. 💚

Read more...

this wholesome carob cake has a brownie-like effect as its a bit more dense and gooey – one piece is satisfying and filling when you need something sweet, and all ingredients are simple healthy and nutritious

dry:

• 1 cup oat flour
• 1/2 cup whole grain flour (i use einkorn but spelt kamut ww etc all work)
• 1/2 cup carob powder
• 3/4 tsp baking soda
• 2 tbsp ground flax meal (golden if available)

wet:

• 1/3 cup unsweetened plain plant yogurt (i use cashew or soy)
• 2 mashed ripe medium size bananas / about 3/4 cup
• 5 pitted medjool dates
• 2 tsp apple cider vinegar
• 3/4 cup hot water

mix dry ingredients in a bowl and combine evenly – add wet ingredients to a blender (you can just add the bananas in chunks to the blender, or mash them first to check the amount but 2 medium size bananas makes about 3/4 cup) and blend well on a higher speed, its nice with some date bits though so you don’t need to blend too long – add the blended wet mix into the dry and with a large spoon mix it all up real good let the mix sit 5 to 10 minutes while oven preheats and you clean up, allowing the flax and oats to absorb some of the moisture – pour mix into a 9×9 baking dish and bake at 350 for 35 minutes – cool completely before cutting – makes 16 squares – refrigerate after the first 24 hours if there is still some left, or refrigerate as soon as they are cool – the consistency is more cakey at room temp and more brownie-like chilled super good if you slice a piece in the middle and stick some nut butter inside like a sandwich – skip the fancy frosting and just go with some almond or peanut butter – makes a healthy treat most any kid would love nutritionally, carob is high in fiber low in fat and has no caffeine – rich in calcium, potassium, iron, magnesium, b-vitamins, riboflavin and niacin, antioxidant polyphenols and prebiotics – its wonderful for digestion it is sweet and nutty on its own with a beautiful smell, and the effect of eating it is very calming – you feel satisfied, nourished, gently comforted and mellow 🙂

background:

growing up we ate whole food plant based and as young children we always had carob instead of cocoa or chocolate, as it had no caffeine and was very nutritious – we didn’t have chocolate until we were older and LOVED everything carob! carob cupcakes, cakes, candies, cookies, breads – it was a favorite for us and all the kids who ate at our house

historic note:

the wild carob / locust bean gum tree grows throughout the mediterranean and has been enjoyed by humans throughout the ages – also widely known as st. john’s bread as st. john the baptist, believed to be a vegetarian (a counter-culture figure of love kindness and compassion for the times) survived on the pods of the locust bean gum tree in the desert – in the early vegetarian communities like seventh day adventists, carob was widely used and enjoyed as a healthier alternative to chocolate (this was before organic cacao was available and cocoa options were pretty limited comparatively), especially for growing children due to its nutrition content – some of the earliest treats available in health food stores were carob since these stores were often run by christian communities who truly did offer a very warm welcoming and abundant array of healthy meat and dairy free alternatives in those early days where there were far less options

environmental note:

carob is far lower impact than cacao – its carbon footprint is significantly smaller, the trees grow well in arid conditions and produce a highly nutritious food, so as far as sustainability carob is a food from the future past that has much to offer to both human and environmental health – grab some and give it a try – if you’ve never had it you’re in for a treat photos show sandwich style, the cake after being refigerated (it gets more gooey), the cake just cooled and unrefrigerated, and the batter for reference – its good every way, even the batter 😉 Read more...

A close reading of the most popular statistic in the pro-meat internet, using the very study it cites.

Part of the same series as Two Acres, The Crop Deaths Argument, Counted Honestly, Rewilding the Plate, and The Pesticide Argument the Carnivore Movement Can’t Answer. Same rule throughout: steelman the other side, then do the arithmetic.

If you spend any time in online debates about meat, you have seen it: a yellow infographic with a donut chart declaring that 86% of global livestock feed is made of materials inedible by humans, with “only 14% edible by humans” stamped in a white circle. It circulates endlessly in regenerative-agriculture and carnivore communities as a one-image rebuttal to any criticism of animal agriculture’s efficiency.

Here is what makes this meme unusually effective: the number is real. It comes from a legitimate, peer-reviewed study led by researchers at the UN Food and Agriculture Organization, Mottet et al. (2017), published in Global Food Security [1]. The graphic even cites it correctly at the bottom.

But the same study that gave the meme its headline number also contains the numbers that quietly undo it: the net loss of human-edible food, the third of the world’s grain, the authors’ own call to reduce grain feeding. The debunk was, in effect, published alongside the claim in 2017. Nearly a decade later, the meme is still shared as though the case were settled in its favor.

A true statistic can still power a false argument, and this one does. Nearly every conclusion the meme invites you to draw is contradicted by the broader evidence, and, remarkably, several are contradicted by the cited study itself. What follows is a point-by-point breakdown, with sources throughout.

Where the 86% number comes from

What study is the 86% livestock feed statistic from?

The study is “Livestock: On our plates or eating at our table? A new analysis of the feed/food debate,” by Anne Mottet, Cees de Haan, Alessandra Falcucci, Giuseppe Tempio, Carolyn Opio, and Pierre Gerber, a team led from the FAO [1].

What is actually inside the 86%?

Livestock consume about 6 billion tonnes of feed (in dry matter) per year, including one third of global cereal production. Of that feed, 86% by dry weight consists of materials “that are currently not eaten by humans”: grass and leaves (46%), crop residues (19%), fodder crops (8%), oilseed cakes (5%), other by-products (5%), and other non-edible materials (3%). Human-edible grains make up 13% and other edibles 1% [1][2]. That is the meme’s donut chart, faithfully reproduced.

The paper was written partly to correct genuinely exaggerated claims, such as the old assertion that every kilogram of meat requires 10 or even 20 kilograms of grain. On that narrow point, it succeeded. The problem is everything the meme built on top of it. There are at least nine distinct issues.

Problem 1: Dry matter is the wrong yardstick

Is livestock feed measured by weight or by nutrition?

The 86% figure is a tally of feed by dry-matter weight. On a scale, a tonne of straw counts exactly the same as a tonne of grain, even though one is fibrous roughage and the other is calorie-dense and protein-dense food. Grass, crop residues, and silage are enormously bulky, so a weight-based accounting is structurally guaranteed to be dominated by them. The 86% is, in a real sense, an artifact of the unit chosen.

What happens when you count feed by calories and protein instead?

Change the denominator and the picture changes dramatically. Cassidy et al. (2013) found that feed crops make up 24% of global crop production by mass, but 36% of global crop calories and a remarkable 53% of global plant protein production, because the crops we feed to animals (maize, soy, oilseed meals) are precisely the densest ones [3]. Measured by the currencies that matter for food security, meaning calories, protein, or economic value, the human-relevant share of livestock feed is far higher than 14%.

Does “inedible to humans” mean it could never have been food?

Note the study’s own careful phrasing: materials “currently not eaten by humans” [1]. Edibility is partly a market classification, not a law of nature. Grain graded and handled as “feed-grade” could in many cases have been grown and handled as food; the category reflects economic choices made because livestock demand exists.

Problem 2: “Only 14%” is a mountain of food

How much human-edible food do livestock actually eat?

The word “only” is doing heroic work in that white circle. Livestock consume roughly 6 billion tonnes of feed dry matter per year [1], so the “mere” 14% that is human-edible works out to roughly 840 million tonnes annually, on the order of the entire global wheat harvest, which runs around 800 million tonnes per year. (The feed figure is dry matter and wheat is reported at harvest moisture, so treat this as an illustrative comparison rather than a precise like-for-like one. The point is the order of magnitude.)

Do livestock really eat a third of the world’s grain?

You do not have to take the arithmetic on faith, because the cited study states the same fact plainly: livestock consume one third of global cereal production [1]. A small percentage of a gigantic number is still a gigantic number.

How many people could be fed with the crops we feed to animals?

To put the opportunity cost in perspective: Cassidy et al. calculated that 36% of the world’s crop calories go to animal feed, and that shifting crops from feed (and biofuel) to direct human consumption could increase available food calories by as much as 70%, enough, in principle, to feed roughly 4 billion additional people [3]. That is a theoretical ceiling, not a policy proposal, but it shows the sheer scale of what “only 14%” represents.

Does this answer the “crops kill animals too” argument?

That acreage multiplier is also why the “crops kill field mice too” gotcha collapses on itself: routing crops through animals commissions more harvested acres per eater, not fewer. The full death-ledger accounting now runs to nine columns, and plant agriculture registers meaningfully in only one of them. The whole ledger is counted out here.

Problem 3: The cited study reports a net loss of human-edible food

How much human-edible feed does 1 kg of meat take?

This is the meme’s most damning problem: its own source contradicts its message. Mottet et al. calculate that producing 1 kg of boneless meat requires, on average, 2.8 kg of human-edible feed in ruminant systems and 3.2 kg in monogastric systems (pigs and poultry) [1]. Read that again: on average, the world’s meat production consumes roughly three times more human-edible food than it produces. Yes, these figures are far lower than the inflated “10 kg or 20 kg of grain” claims of the past, which was the paper’s corrective point, but a 2.8-to-1 or 3.2-to-1 losing ratio is still a losing ratio.

Are grazing cattle a net gain rather than a loss?

The exception the meme universalizes is grazing ruminants: cattle in grazing systems need only about 0.6 kg of human-edible protein to produce 1 kg of protein in milk and meat, making those specific systems net contributors of protein [2]. That is a genuinely interesting finding about one subset of production. It does not describe pigs, poultry, or grain-finished cattle, which is to say it does not describe most of the meat consumed in wealthy countries.

What did the study’s own authors actually conclude?

They call for reducing the use of human-edible grain in feed and improving feed-conversion ratios to prevent further expansion of arable land dedicated to feed production [1][2]. The paper is a case for reforming livestock production, not a vindication of the status quo. Lead author Anne Mottet has been explicit that the work argues for better livestock systems, with animals eating more residues and by-products and less grain, which is precisely the opposite of “nothing to see here.”

Problem 4: “Inedible by humans” is not the same as “costless to humans”

The meme’s implicit logic is that if humans cannot eat it, feeding it to animals is free. Land economics says otherwise.

How much cropland is used to grow animal feed?

Fodder crops, 8% of the donut, are not scavenged waste. They are crops deliberately cultivated on arable land that could grow human food. Overall, the study estimates livestock use about 40% of global arable land [2], and Poore and Nemecek’s global meta-analysis found that 38% of the world’s cropland is used for livestock feed, versus 50% for direct human food [4]. Roughly four of every ten cultivated acres on Earth are feeding animals, not people.

Can grazing land really not grow crops?

By the study’s own accounting, 57% of the land used for feed production is unsuitable for growing food, which means 43% is suitable. More specifically, the paper estimates that of the 2 billion hectares of grassland grazed by livestock, about 700 million hectares could be used as cropland [1][2]. That is an area roughly the size of Australia currently under pasture that could, physically, grow food.

What is the opportunity cost of land that cannot grow crops?

Land that cannot grow wheat can grow forests, store carbon, and host biodiversity. Poore and Nemecek estimate that a global shift away from animal products would reduce agricultural land use by about 75%, freeing an area the size of the US, China, the EU, and Australia combined, while still feeding the world [5][6]. And the carbon that freed land could pull back down is enormous: Hayek et al. (2021) estimate that letting vegetation regrow on land currently used for animal agriculture could sequester around 330 to 550 billion tonnes of CO2, on the order of the entire remaining carbon budget for a 1.5 °C target [18]. What that freed land becomes, and what actually happens to grazed ecosystems when the cattle come off, is the subject of Rewilding the Plate.

How much of the Amazon is cleared for cattle?

“Grass and leaves” in the donut includes pasture that was recently rainforest. The expansion of pasture for beef is the single largest driver of tropical deforestation, responsible for about 41% of it globally, around 2.1 million hectares per year (Pendrill et al. 2019) [7]. And cattle pasture occupies roughly 80% of the deforested land in the Brazilian Amazon [8]. When the meme’s green photograph of dewy grass includes ex-rainforest, “food we can’t eat” starts to look rather different.

Problem 5: The by-product sleight of hand

The donut classifies oilseed cakes (5%) and other by-products (5%) as leftovers that livestock heroically rescue. For some materials, such as brewers’ grains, beet pulp, and citrus pulp, that is fair. For the most important one, it is not.

Is soybean meal really just a by-product of oil?

Soybean meal (cake) is what remains after soybeans are crushed for oil. Calling it a “by-product” implies the oil is the point and the meal is a happy accident. In reality, the meal is most of the bean, about 44 of every 60 pounds crushed, roughly three-quarters of its mass [9]. For most of the crop’s history it was also most of the crush’s value; that value split has only recently shifted toward oil as the US renewable-diesel boom pushed soybean oil’s share past 50% [16]. Either way the meal is no afterthought. And here is the kicker: Mottet et al. themselves flag soybean cake production as a main driver of land use, even while counting it as only 4% of global feed intake [1]. The meme’s own source refuses the “innocent leftover” framing.

How much of the world’s soy goes to animal feed?

Zoom out to the whole soy economy and the picture sharpens: 77% of global soy is used as animal feed, while just 7% goes to direct human foods like tofu, tempeh, and soy milk. Of the total, over a third feeds poultry, about a fifth feeds pigs, and 6% goes to aquaculture [9]. Feed demand does not mop up the soy industry’s waste. Feed demand is the soy industry.

Are crop residues free to feed out?

Even genuine residues are not costless. Crop residues, 19% of the donut, have competing uses, chief among them being returned to the soil to maintain fertility and prevent erosion. “Inedible by humans” quietly conflates “no alternative use” with “no culinary use,” and those are very different claims.

One more cost hides in the feed complex: it is where the food system’s pesticide load concentrates, with federal residue tolerances on feed crops running hundreds to thousands of times those permitted on human produce. That regulatory paper trail is dissected here.

Problem 6: There is a third yardstick, and nobody counts it

Problem 1 showed that switching the denominator from dry weight to calories and protein transforms the picture. But calories and protein are not the only things a crop contains, and the third currency is the one this debate never counts at all.

Is animal feed just nutritionally empty bulk?

Feed crops are not inert filler. Maize bran carries one of the highest concentrations of ferulic acid reported in any plant tissue, and yellow maize is a documented source of lutein and zeaxanthin. Soybeans carry roughly 200 mg of isoflavones per 100 g of dry weight, plus saponins and a range of phenolic acids. Even the grass in the donut chart’s own photograph is real chemistry: fresh pasture runs on the order of 44 to 105 mg/kg of beta-carotene and 21 to 86 mg/kg of alpha-tocopherol on a dry-matter basis [19]. That is precisely the class of compounds, the carotenoids, tocopherols, and polyphenols, that no animal on Earth can synthesize and every animal must obtain by eating plants.

What happens to plant phytochemicals inside a cow?

Most of them do not survive the trip. Rumen microbes degrade a large share of the polyphenol load before it is ever absorbed; soy isoflavones, for instance, are broken down or converted to equol inside the rumen [20]. What does get absorbed is largely conjugated by the liver and excreted, with a substantial fraction of absorbed ferulic acid leaving in the urine [21], and reviews of the field conclude that tissue polyphenol content is not proportional to dietary intake [22].

The fat-soluble survivors are routed preferentially into liver, body fat, and milk fat rather than into skeletal muscle, which is the tissue people actually eat [23]. Cattle are in fact the best carotenoid depositors among common ruminants, which is why beef fat is yellower than the near-white fat of sheep and goats, and why grass-fed butter is yellower than grain-fed [24]. But muscle remains a poor depot. Where the transfer has been measured directly, the carry-over rates are small: studies tracking isoflavones from feed into milk report recoveries in the low fractions of a percent of intake [25].

How much antioxidant capacity actually survives into meat?

The largest survey of its kind assayed more than 3,100 foods. Animal-based foods came in at a median of 0.10 units of antioxidant capacity per 100 g against 0.88 for plant-based foods, a gap of roughly nine to one [26]. The means are far more lopsided still, 0.18 against 11.57, but the authors are candid that the plant mean is inflated by a minority of extreme products among the spices and herbal medicines, so the median is the fairer test. Their own like-for-like comparison is perhaps the cleanest statement of it: measured against the mean for meat and meat products, the fruit, nut, chocolate, and berry categories run from 5 to 33 times higher.

Two honest caveats belong here, because this argument should be held to the same standard as the rest of the post. First, that survey used an in-vitro chemical assay, and its authors say plainly that there is not necessarily a direct relationship between the antioxidant content of a food and the antioxidant activity that follows in a target cell. Antioxidant capacity scores are a measure of chemistry, not a measure of health outcomes. Second, nobody has ever directly assayed ferulic acid or isoflavone residues in beef muscle, so the honest claim is trace amounts by inference rather than a measured zero. The inference is well supported, resting on rumen degradation plus hepatic conjugation plus the measured preference for fat over muscle, but it is an inference, and worth labeling as one.

Does this mean animal foods have no nutritional advantages?

No, and the post should not pretend otherwise. Animal-source foods are genuinely good delivery vehicles for vitamin B12, heme iron, zinc, and preformed retinol, and in parts of the world where diets are narrow, that matters a great deal. Milk really does carry some equol through to the glass. The claim here is narrower and survives all of that: the phytochemical payload of the feed is not concentrated by the animal, it is largely dismantled by it.

Why this reframes the meme’s central move

“Inedible by humans” is offered as though the only question were whether the material could have been a meal. But the feed complex is full of chemistry that is valuable to a human body and is destroyed rather than delivered by routing it through a cow. The loss is not only caloric, and not only nutritional in the protein sense. It is a loss of the phytochemicals and the fiber alike, and almost none of it survives the trip. Whatever else the animal is doing, it is not concentrating this material for us.

Problem 7: A global average is being used to defend systems it does not describe

Do pigs and chickens eat grass?

The 46% “grass and leaves” figure is a global average, pulled upward by vast extensive and pastoralist systems, largely in lower-income countries, where cattle, goats, and sheep graze marginal land. Those systems are the ones the study’s favorable numbers (like the 0.6 protein ratio) describe.

They are not the systems that put meat on plates in the US or Europe. Industrial chicken, pork, and feedlot-finished beef, the overwhelming majority of rich-country meat, run on grain and soy, the exact human-edible and cropland-grown feeds the meme waves away. Recall that Mottet et al.’s conversion figure for monogastrics (pigs and poultry) is 3.2 kg of human-edible feed per kg of meat [1], and pigs and poultry are most of what is actually eaten.

Is grass-fed beef better for the climate?

Globally, grass-fed beef is a rounding error: the Oxford-led Grazed and Confused report found that fully grass-fed beef contributes roughly 1 gram of protein per person per day to the global supply, out of about 32 grams of animal protein consumed per person per day [10]. And a 2025 analysis in PNAS found that even under optimistic rangeland-sequestration assumptions, US grass-fed beef is no less carbon-intensive than industrial beef, and 3 to 40 times as carbon-intensive as a wide range of plant and animal alternatives, roughly 10-fold for common protein-dense foods, while cropland reallocated from beef feed to direct human food delivers 3 to 7 times more protein per unit of land, emissions, and nitrogen [11]. A pastoralist herd in the Sahel cannot vindicate a feedlot burger in Ohio, and it turns out a boutique grass-fed ranch cannot either. For the acre-by-acre arithmetic of what grazing land actually produces versus what cropland can, see Two Acres.

What about regenerative or adaptive multi-paddock grazing?

The strongest reply here is regenerative or adaptive multi-paddock (AMP) grazing, the claim that well-managed cattle build enough soil carbon to cancel their own emissions. It deserves a fair hearing, and the better field studies do find real gains: one southeastern-US comparison measured about 13% more soil carbon on AMP sites than on conventionally grazed ones (Mosier et al. 2021) [17]. But soil-carbon gains are finite, saturate within a few decades, reverse if the management lapses, and vary enormously by site. They are a one-time deposit, not a permanent subsidy. And the objection does not rescue the meme even on its best day, because the PNAS analysis above already builds in optimistic sequestration and grass-fed beef still finishes as the most carbon-intensive protein on the board. When regenerative-grazing advocates pressed exactly this point, the authors replied that such exceptions are real but far too rare to change the result [11].

Problem 8: Feed competition was never the whole efficiency argument

The meme frames the entire debate as a single question, “do livestock eat our food?”, and then declares victory on it. But the efficiency critique of animal agriculture rests on several independent pillars, and the feed/food question is only one.

Why are the conversion numbers this bad?

There is a reason these figures are as poor as they are, and it is not bad farming. The second law of thermodynamics guarantees that no energy transfer is free: some fraction always disperses as unusable heat. Ecologists reduced the consequence to a rule of thumb, that roughly a tenth of the energy at one level of a food chain survives into the next, and while that ten percent is an average rather than a constant (Lindeman’s original measurements ranged from a fraction of a percent up into the thirties), the direction never reverses [27]. A plant banks sunlight as calories. A cow spends most of those calories being a cow. What reaches a plate is the remainder.

This is why the efficiency critique cannot be managed away. Better genetics, adaptive multi-paddock rotation, holistic management: none of them repeal the law. They can argue about the size of the tax. They cannot argue about whether it is levied. Every figure in this section is that tax, scaled to a planet.

Two rebuttals are worth anticipating, because they sound like answers and are not. The first is that most of the loss is respiration, the animal breathing, moving, and staying warm. True, and irrelevant to the accounting: those calories still left the human food supply, and the fact that a cow burned them rather than a person does not put them back. The second is that the feed was inedible anyway, so the loss is free. Also false, as Problem 4 established: growing, harvesting, transporting, and feeding that material consumes land, water, fuel, and fertilizer that could have produced human food directly. Against that expenditure, the return includes not only the meat but the methane, the manure lagoons, and the nutrient runoff.

How much farmland does animal agriculture use?

Livestock use about 83% of the world’s farmland while supplying just 18% of calories and 37% of protein (Poore and Nemecek 2018, the largest meta-analysis of global food systems, covering about 38,700 farms in 119 countries) [4].

How efficient is beef at converting feed into food?

In the US, the average feed-to-food conversion efficiency across livestock categories is 7% to 8% for both calories and protein; for beef it is 3% on both metrics, meaning 97 of every 100 feed calories are lost. Merely reallocating US beef-feed cropland to poultry feed could meet the caloric needs of roughly 120 to 140 million additional people [12].

What share of greenhouse gas emissions comes from livestock?

Livestock account for roughly 12% of anthropogenic greenhouse-gas emissions, about 6.2 Gt CO2-equivalent per year, per FAO’s 2023 assessment, which revised down the older and still widely quoted 14.5% figure from FAO’s 2013 estimate [15]. And the Grazed and Confused team found that even under generous assumptions, soil carbon sequestration from grazing could offset only 20% to 60% of grazing cattle’s own emissions, leading them to conclude that expanding grass-fed ruminant production is “a self-defeating climate strategy” [10][13].

So even if the 86% statistic meant everything the meme wants it to mean, the land, climate, and conversion arguments would stand untouched. Rebutting one claim, and rebutting it badly, settles nothing.

Problem 9: Follow the meme’s logic and you arrive at much less meat

What if livestock only ate food humans cannot eat?

Here is the deepest irony. Suppose we take the meme’s principle completely seriously: livestock should exist to upcycle grass, residues, and true by-products, food we cannot eat, into food we can. Researchers have modeled exactly that world.

Van Zanten et al. (2018), reviewing the literature on livestock raised solely on “leftover streams” and grassland, found such systems could supply about 9 to 23 grams of protein per person per day, a real but modest fraction of the 50 to 60 grams humans need daily [14]. Ruminants fed solely on existing grasslands would supply around 7 grams per person per day [14]. Compare that with the roughly 32 grams of animal protein the average person on Earth already consumes daily [10], and with the far higher intakes typical in Western countries, and the conclusion is unavoidable: “livestock as upcyclers” is an argument for a livestock sector a fraction of the size of today’s, with rich-country meat consumption falling the furthest.

The meme borrows the moral glow of the upcycling model to defend a food system that does not remotely resemble it.

What the graphic gets right, and why it works

Intellectual honesty requires acknowledging what is true here, and it also explains the meme’s persuasive power.

The paper is real, correctly cited, and authored by credible FAO researchers. It genuinely corrected exaggerated feed-conversion claims that circulated for years. Ruminants really can convert cellulose humans cannot digest into nutrient-dense food, and specific grazing systems really are net protein contributors [1][2]. None of that is in dispute.

That is exactly what makes the meme effective: it is true but radically incomplete, the most durable species of misleading content. It answers a question almost nobody was asking (“is livestock feed mostly grain by dry weight?”) and presents the answer as though it resolved the questions people actually ask: whether animal agriculture competes with human food supply (it does, one third of the world’s grain), whether it uses land efficiently (it does not, 83% of farmland for 37% of protein), and whether current consumption levels are defensible on the meme’s own upcycling logic (they are not).

Consider the source

Who made the 86% livestock feed infographic?

The URL on the graphic, sacredcow.info, is not a research institution. It is the promotional website for Sacred Cow: The Case for (Better) Meat, a pro-meat advocacy book and companion film by Diana Rodgers and Robb Wolf. The graphic is, quite literally, marketing material.

Notice also the framing in the meme’s text: the efficiency critique is attributed to “alternative protein companies,” as though concern about feed and land use were a corporate talking point. In reality, the feed/food analysis comes from the FAO itself, the land-use figures from Oxford and Agroscope researchers publishing in Science, the conversion efficiencies from the Weizmann Institute publishing in Environmental Research Letters, and the climate analysis from an Oxford-led international team. The meme cites the academic establishment with one hand and recasts it as industry spin with the other.

The bottom line

The 86% statistic is real. The argument built on it fails on every front. By the very study the meme cites, livestock eat one third of the world’s grain and consume roughly three kilograms of human-edible food for every kilogram of meat they produce, and its authors call for grain feeding to be reduced. The “inedible” feed occupies land, including nearly 40% of the world’s cropland for feed, 700 million hectares of croppable pasture, and millions of hectares of former rainforest, that carries enormous opportunity costs. The “by-products” include a soy industry in which 77% of output feeds animals. And if you actually embrace the meme’s own principle, that livestock should eat only what we cannot, the science says that world contains a small fraction of the meat we currently produce.

Real number. Wrong conclusion. The strongest debunking of this meme was published in 2017, in the journal Global Food Security, by the authors the meme thanks in its own fine print.

Related reading in this series

• Two Acres: What Beef Produces vs. What Plants Can. The acre-by-acre math: roughly 500,000 calories of beef versus 20 to 32 million calories of plants from the same land, and why the corn and soy monoculture exists chiefly to feed animals.
• The Crop Deaths Argument, Counted Honestly. The field-mouse gotcha traced to its academic source, the same-year correction that flipped it five-to-one, and the nine columns of the death ledger the argument never counts: field deaths, slaughter, wild capture, predator control, disease culls, on-farm mortality, pesticide kills, dead zones, and habitat loss.
• Rewilding the Plate. What the freed land becomes: the regenerative-grazing story audited, and the peer-reviewed recovery that follows when the cattle come off.
• The Pesticide Argument the Carnivore Movement Can’t Answer. Feed crops carry the heaviest pesticide tolerances in federal law, and routing calories through animals multiplies the sprayed acreage your diet requires.

References

1. Mottet, A., de Haan, C., Falcucci, A., Tempio, G., Opio, C., and Gerber, P. (2017). Livestock: On our plates or eating at our table? A new analysis of the feed/food debate. Global Food Security, 14, 1–8. sciencedirect.com
2. Mottet, A., et al. (2017). More fuel for the food/feed debate. Feedipedia / FAO Broadening Horizons No. 47. The authors’ own open-access write-up, stating the 0.6 kg protein ratio for grazing cattle, about 40% of arable land used for feed, and about 700 million hectares of grazed grassland that could be cropland. feedipedia.org. Further summarized at whylivestockmatter.org
3. Cassidy, E. S., West, P. C., Gerber, J. S., and Foley, J. A. (2013). Redefining agricultural yields: from tonnes to people nourished per hectare. Environmental Research Letters, 8(3), 034015. iopscience.iop.org
4. Poore, J., and Nemecek, T. (2018). Reducing food’s environmental impacts through producers and consumers. Science, 360(6392), 987–992. (83% of farmland; 18% of calories; 37% of protein; cropland split of 50% food, 38% feed, 12% non-food per Supplementary Table S10.) science.org
5. Ritchie, H., and Roser, M. Half of the world’s habitable land is used for agriculture. Our World in Data. ourworldindata.org
6. Ritchie, H. If the world adopted a plant-based diet, we would reduce global agricultural land use from 4 to 1 billion hectares. Our World in Data. ourworldindata.org
7. Ritchie, H. Drivers of Deforestation. Our World in Data, summarizing Pendrill, F., et al. (2019), Global Environmental Change. (Beef equals 41% of tropical deforestation, about 2.1 million hectares per year.) ourworldindata.org
8. WWF. Unsustainable cattle ranching. (Cattle ranching accounts for about 80% of deforested land in the Brazilian Amazon; Nepstad et al. 2008.) wwf.panda.org
9. Ritchie, H. (2021). Soy. Our World in Data. (77% of global soy to animal feed; 7% to direct human food; poultry about 37%, pigs about 20%, aquaculture 6%. Meal is about 73% of the bean by mass, roughly 44 of every 60 lb crushed.) ourworldindata.org
10. Garnett, T., Godde, C., et al. (2017). Grazed and Confused? Food Climate Research Network, University of Oxford. (Grass-fed beef about 1 g protein per person per day globally; sequestration offsets at most 20% to 60% of grazing cattle emissions; “self-defeating climate strategy.”) oxfordmartin.ox.ac.uk. Coverage: carbonbrief.org
11. Eshel, G., Flamholz, A., Shepon, A., and Milo, R. (2025). US grass-fed beef is as carbon intensive as industrial beef and about 10-fold more intensive than common protein-dense alternatives. PNAS, 122, e2404329122. pnas.org. See also the authors’ reply to a regenerative-grazing critique, PNAS 122, e2506941122: pnas.org
12. Shepon, A., Eshel, G., Noor, E., and Milo, R. (2016). Energy and protein feed-to-food conversion efficiencies in the US and potential food security gains from dietary changes. Environmental Research Letters, 11(10), 105002. iopscience.iop.org
13. Inside Climate News (2017). Eating grass-fed beef isn’t as climate-friendly as you may think. insideclimatenews.org
14. Van Zanten, H. H. E., Herrero, M., Van Hal, O., Röös, E., Muller, A., Garnett, T., Gerber, P. J., Schader, C., and De Boer, I. J. M. (2018). Defining a land boundary for sustainable livestock consumption. Global Change Biology, 24(9), 4185–4194. onlinelibrary.wiley.com
15. FAO (2023). Pathways towards lower emissions: A global assessment of the greenhouse gas emissions and mitigation options from livestock agrifood systems. (Livestock agrifood systems about 6.2 Gt CO2-eq per year, about 12% of anthropogenic GHG emissions, revising down the older 14.5% figure from FAO 2013.) fao.org
16. CME Group (2025), “Soybean Oilshare,” and Janzen, J., and Wang, S. (2025), “The Soybean Industry Response to the Renewable Diesel Boom, Part 3,” farmdoc daily. (A 60-lb bushel yields about 44 lb meal and 11 lb oil; soybean oil’s share of crush value rose from about 25% to 35% up past 50% amid renewable-diesel demand.) cmegroup.com. farmdocdaily.illinois.edu
17. Mosier, S., Apfelbaum, S., Byck, P., Calderon, F., Teague, R., Thompson, R., and Cotrufo, M. F. (2021). Adaptive multi-paddock grazing enhances soil carbon and nitrogen stocks and stabilization through mineral association in southeastern US grazing lands. Journal of Environmental Management, 288, 112409. sciencedirect.com
18. Hayek, M. N., Harwatt, H., Ripple, W. J., and Mueller, N. D. (2021). The carbon opportunity cost of animal-sourced food production on land. Nature Sustainability, 4, 21–24. (Shifting to plant-based production could allow regrowth sequestering about 332 to 547 Gt CO2, roughly 99% to 163% of the CO2 budget for a 66% chance of 1.5 °C.) nature.com
19. Lindqvist, H., et al. (2014). Alpha-tocopherol and beta-carotene contents of forage species. Grass and Forage Science, 69, 356. doi.org
20. Vasta, V., et al. (2019). Invited review: Plant polyphenols and rumen microbiota responsible for fatty acid biohydrogenation, fiber digestion, and methane emission. Journal of Dairy Science, 102(5), 3781–3804. doi.org
21. Adam, A., et al. (2002). The bioavailability of ferulic acid is governed primarily by the food matrix rather than its metabolism in intestine and liver in rats. Journal of Nutrition, 132(7), 1962–1968. doi.org
22. Serra, V., et al. (2021). Dietary polyphenol supplementation in food producing animals: effects on the quality of derived products. Animals, 11(2), 401. doi.org
23. Kalač, P. (2012). Carotenoids, ergosterol and tocopherols in fresh and preserved herbage and their transfer to bovine milk fat and adipose tissues. Journal of Agrobiology, 29, 1. doi.org
24. Yang, A., Larsen, T. W., and Tume, R. K. (1992). Carotenoid and retinol concentrations in serum, adipose tissue and liver and carotenoid transport in bovine, ovine and caprine species. Australian Journal of Agricultural Research, 43, 1809. doi.org
25. Křížová, L., et al. (2021). Production of bovine equol-enriched milk: a review. Animals, 11(3), 735. (Feed-to-milk isoflavone carry-over rates vary by source and intake and are consistently small fractions of intake.) doi.org. See also Mustonen, E. A., et al. (2009), British Journal of Nutrition, 102(11), 1552–1556. doi.org
26. Carlsen, M. H., et al. (2010). The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide. Nutrition Journal, 9, 3. (Table 1: plant-based foods median 0.88 and mean 11.57 mmol/100 g; animal-based foods median 0.10 and mean 0.18; the authors note the plant mean is raised by a minority of very high products among plant medicines, spices, and herbs, and that antioxidant content measured in vitro does not necessarily correspond to activity in a target cell.) doi.org
27. Lindeman, R. L. (1942). The trophic-dynamic aspect of ecology. Ecology, 23(4), 399–417. doi.org

Read more...

The Fire Within: My Battle with Lupus Inflammation

By Cindy Amick

The Fire Within: My Battle with Lupus Inflammation

Six years ago, my life was consumed by an invisible fire. Lupus had taken control of my body, and chronic inflammation was the fuel that kept this destructive blaze burning. I know firsthand what it feels like when your immune system turns against you, creating a cascade of inflammatory responses that affect every aspect of your existence.

Living with lupus inflammation isn’t just about joint pain—though mine felt like molten metal had been poured into my joints. It’s about waking up feeling like you’ve been hit by a truck, even after a full night’s sleep. It’s about watching your face transform with that telltale butterfly rash, Sjogren’s, feeling the burn of mouth ulcers, and struggling through the mental fog that makes simple decisions feel impossible.

What I didn’t understand then was how inflammation was orchestrating this symphony of symptoms. The inflammatory cascade in lupus releases pro-inflammatory cytokines like TNF-α, IL-1, and IL-6 throughout the body. These molecular messengers create immune complexes that deposit in tissues, activating the complement system and causing widespread cellular damage through oxidative stress.

The Conventional Path: Symptom Management Without Answers

Like most lupus patients, I was prescribed the conventional treatment protocol. I didn’t want to see a specialist or take any of the medicines after learning about their side effects and worry about long-term health consequences. I didn’t believe they would make me better. The frustrating reality was that the conventional treatments were only managing symptoms, not addressing the root cause.

I found a disease reversal doctor who showed me a natural way to cure myself of Lupus through hyper nourishing, and that prompted me to take the plant nutrition certificate that my doctor had taken to learn how nutrition healed me. I was convinced by what had happened to me that nutrition could play a more significant role than most people realized in helping the body heal itself.

The Turning Point: Discovering Edestin

My breakthrough came when I stumbled upon research about edestin, a protein found in hemp seeds. Unlike the incomplete proteins we have all been told to consume, Edestin contains all essential amino acids in ratios that perfectly match human tissue requirements. This wasn’t just another plant protein that fit the category of ‘superfood’: it was a level above other superfoods—Edestin was a complete nutritional building block that would support my body’s repair processes at the cellular level.

What fascinated me about Edestin was its unique molecular structure and similarity to human protein that make it easily digestible and highly bioavailable, meaning my body could actually use it efficiently. But more importantly, early research suggested that edestin might have anti-inflammatory properties that could help modulate my overactive immune response.

The Edestin Protocol: A New Approach to Inflammation Management

After successfully completing the disease reversal protocol, I decided to completely change my protein source to edestin and developed what I now call the Edestin Diet Protocol. This wasn’t just about swapping proteins—it was about creating a comprehensive nutritional approach that would support my body’s natural healing mechanisms.

The protocol focuses on:

High-Quality Protein Foundation: Replacing food that cause inflammation with a plant based version featuring edestin as the main protein to ensure optimal amino acid profiles, nutrient dense food for super nourishing, tissue repair and immune function.

Hemp-Based Nutrition: Edestin may be the perfect protein for humans but I’ve also incorporated hemp leaf juice powder into my protocol. This powerful juice contains CBDA which has been shown in preclinical studies to be up to 1,000 times more potent than CBD at activating the 5-HT1A serotonin receptor — a key player in regulating mood, nausea, and stress response:

• Anti-inflammatory Properties –Research shows CBDA may be more effective than CBD at inhibiting COX-2, an enzyme associated with inflammation and pain.
• Nausea & Anxiety Relief – Studies indicate CBDA may be significantly more effective than CBD at activating 5-HT1A serotonin receptors, which help regulate nausea, anxiety, and mood.
• Enhanced Bioavailability – CBDA has shown greater bioavailability than CBD, meaning the body can absorb and utilize it more effectively at lower doses.
• Seizure Prevention – Preliminary research suggests CBDA may have anticonvulsant properties that could help prevent seizures, potentially at lower doses than CBD.
• Cancer Cell Inhibition – Studies have shown CBDA may help inhibit the migration of aggressive breast cancer cells, suggesting potential applications in cancer treatment.
• COVID-19 Prevention – Recent research from Oregon State University found that CBDA may bind to the SARS-CoV-2 spike protein, potentially blocking the virus from entering cells.

The Results: My Personal Transformation

The changes didn’t happen overnight, but they were profound. Within 5 months of starting the super nourishing diet approach that I learned from my doctor, I saw all my symptoms disappear. My morning stiffness decreased, the brain fog lifted, and my energy levels started returning to normal.

Most remarkably, my inflammatory markers began normalizing. The butterfly rash faded, joint pain became manageable and then disappeared, and my energy was high all day long. I was no longer bedridden and worried I would die. I felt happier. For the first time in years, I felt like myself again. It wasn’t just symptom relief—it felt like my body was actually healing.

The results were so encouraging that I knew I had to understand the science behind what was happening. This led me to pursue formal education in plant based nutrition at Cornell University, where I dove deep into nutritional immunology and the mechanisms by which specific nutrients can modulate inflammatory pathways.

The Science Behind the Success

Through my studies, I discovered why the Edestin Protocol was so effective. The key lies in understanding how proper nutrition can support the body’s natural inflammatory resolution processes.

Optimal Amino Acid Balance: Edestin provides all essential amino acids in ratios that support the synthesis of anti-inflammatory compounds and tissue repair proteins. This gives the body the building blocks it needs to heal damaged tissues and regulate immune responses.

Specialized Pro-Resolving Mediators: The GLA in hemp nutrition helps the body produce specialized compounds that actively resolve inflammation rather than just suppressing it. This is fundamentally different from conventional anti-inflammatory drugs.

Antioxidant Support: The chlorophyll and other phytonutrients in hemp leaves provide powerful antioxidant support, helping to neutralize the free radicals that perpetuate inflammatory damage.

Immune System Balance: Rather than suppressing the immune system, the Edestin Protocol appears to help restore proper immune function, allowing the body to distinguish between threats and healthy tissue.

Why Doctors Don’t Know About This Yet

The most frustrating part of my journey has been realizing that most healthcare providers, even the disease reversal doctor who healed me, aren’t aware of these nutritional approaches. Medical education focuses heavily on pharmaceutical interventions, with limited -or no training- in nutritional immunology or the therapeutic potential of specific proteins like edestin. Doctors are not taught about edestin and hemp nutrition in medical school, so they don’t prescribe this kind of diet as a treatment.

This knowledge gap means that millions of people with lupus and other autoimmune conditions aren’t being offered these potentially transformative nutritional strategies. The research exists, but it’s scattered across various fields and hasn’t yet made its way into mainstream medical practice.

That’s why I’m so passionate about sharing this information. I believe that in the coming years, we’ll see a revolution in how we approach autoimmune diseases—moving from symptom suppression and “managing” to nutritional support for the body’s natural healing mechanisms.

The Edestin Difference: A New Paradigm

What makes the Edestin Protocol different from other nutritional approaches is its focus on providing the body with exactly what it needs to function optimally. Instead of trying to force the immune system into submission, we’re supporting its natural ability to maintain balance.

The results I’ve experienced aren’t unique to me. As more people try this approach, I’m seeing similar improvements in energy, inflammation markers, and overall quality of life. The key is understanding that this isn’t just about taking a supplement—it’s about fundamentally changing how we nourish our bodies.

Looking Forward: Hope for the Future

Six years ago, I couldn’t have imagined writing these words. Lupus had stolen my vitality, my hope, and my sense of control over my own health. Today, I’m not just managing symptoms—I’m thriving.

The Edestin Protocol has given me my life back, and my education in plant nutrition has shown me why it works. This isn’t magic—it’s science. It’s the power of giving our bodies the precise nutrients they need to heal and maintain optimal function.

I know there are thousands of people out there struggling with lupus and other autoimmune conditions, following conventional treatments that only manage symptoms. I want them to know that there may be another way. The research is emerging, the results are encouraging, and the potential for transformation is real.

A Personal Invitation

If you’re living with lupus or chronic inflammation, I encourage you to explore the possibility that your protein choice and overall nutritional approach could be key factors in your healing journey. The Edestin Protocol isn’t just a theory—it’s a lived experience that has transformed my health and my understanding of what’s possible.

While the medical establishment catches up to this research, we don’t have to wait. We can start making informed choices about our nutrition today, supporting our bodies’ natural healing mechanisms and potentially discovering the same transformation I’ve experienced.

The journey from lupus to wellness is possible. It starts with understanding that our bodies want to heal—they just need the right nutritional support to do so.

Read more...

Cholesterol might be the most argued-about molecule on the internet. One camp says it built the modern epidemic of heart disease. Another says it has been framed. Both camps quote studies. So instead of quoting a study, this article is going to walk the entire staircase of evidence, from single molecules at the bottom to global scientific consensus at the top, and let you watch every step point in the same direction.

By the end, you will know what a micelle is, what ApoB means, why your arteries cannot feel themselves being damaged, and why nearly every major scientific body on Earth, across multiple countries and specialties, landed in the same place.

Your Body Already Makes Every Bit of Cholesterol It Needs

What does cholesterol actually do in the body?

Cholesterol itself is not a villain. It is one of the most important molecules you own. It stiffens and stabilizes the membrane of every cell, it is the raw material for testosterone, estrogen, cortisol, and vitamin D, it forms the bile acids that digest your food, and it wraps your nerves in insulation called myelin.

Does your body make its own cholesterol?

It is so important, in fact, that your body refuses to outsource it. Essentially every cell with a nucleus can manufacture its own cholesterol on demand. Your brain is the most dramatic example: it holds roughly a quarter of all the cholesterol in your body, and it makes every bit of it locally, because the blood-brain barrier blocks cholesterol arriving from the bloodstream. Not one molecule of the cholesterol in your brain came from an egg. Your liver, meanwhile, acts as the central bank of the whole system, manufacturing a large share, packaging it for shipment, recalling it from circulation, and converting the excess into bile.

How much cholesterol do you need to eat?

Add it up and your body synthesizes roughly a gram of cholesterol a day, several times more than most people eat. Which is why the National Academies (the Institute of Medicine, in its official Dietary Reference Intakes) concluded there is no dietary requirement for cholesterol at all, and advised keeping intake as low as possible within a nutritionally adequate diet. That single sentence is the anchor for everything that follows: the question was never whether you need dietary cholesterol. You do not. The question is what happens when you eat it anyway.

What is the optimal LDL cholesterol level?

There is a companion fact worth holding onto here. The LDL level that healthy human newborns, contemporary hunter-gatherers, and free-living primates all share, none of whom develop atherosclerosis, sits around 50 to 70 milligrams per deciliter, roughly half the modern Western average. A landmark review in the Journal of the American College of Cardiology made the case directly in its title: optimal LDL is 50 to 70, and that range is not some aggressive medical target but simply the physiologically normal level our species evolved with. Today’s “normal,” in other words, is not the same thing as natural.

What Happens When You Eat It Anyway

How is dietary cholesterol absorbed?

Dietary cholesterol is chewed and swallowed, passing into the small intestine, where bile acids released from the gallbladder emulsify it into tiny water-soluble packages called micelles. Think of a micelle as a soap bubble built to smuggle grease through water. These micelles ferry the cholesterol to the intestinal wall, where a transport protein called NPC1L1 actively pulls it inside the intestinal cells. NPC1L1 is essentially a customs gate for cholesterol, and it is worth remembering that name, because an entire drug class exists just to block it.

Inside the cell, the cholesterol is chemically modified into a storage form called a cholesterol ester and loaded, along with dietary fat, into a massive transport vehicle called a chylomicron. A chylomicron is too large to squeeze into ordinary blood capillaries, so it exits through the porous lymphatic vessels called lacteals, rides the lymphatic system upward, and finally drains into your bloodstream through the thoracic duct near your neck. Within hours of a meal, cargo ships full of dietary fat and cholesterol are sailing through your arteries.

Why do some people absorb more cholesterol than others?

Two honest details make this picture complete. First, the pool of cholesterol sitting in your gut is not only what you ate. Your own liver dumps roughly a gram of cholesterol per day into the intestine through bile, typically more than food contributes. Second, how much of that pooled cholesterol gets absorbed varies wildly between people, from around 20 percent to around 80 percent. Some people are natural hyper-absorbers. Hold that thought, because it explains one of the strangest findings in recent nutrition research.

The Feedback Loop: How Eating Cholesterol Changes Your Blood

How does eating cholesterol raise blood cholesterol?

When dietary cholesterol enters the bloodstream inside chylomicron remnants, those remnants are taken up by the liver, and the liver’s internal cholesterol pool fills up. The liver responds the way any warehouse responds to a full inventory: it stops accepting deliveries. Specifically, it reduces the number of LDL receptors on its surface.

LDL receptors are the vacuum cleaners of the bloodstream. Their discovery won Michael Brown and Joseph Goldstein the 1985 Nobel Prize, and it remains the single most important concept in this entire subject. When the liver displays fewer receptors, LDL particles, the so-called bad cholesterol, are cleared more slowly and circulate longer. The longer a particle drifts, the more time it has to be chemically altered and oxidized into forms your immune system treats as a threat. On top of that, a cholesterol-loaded liver packs its outgoing VLDL particles with extra cholesterol, and those particles degrade in circulation into cholesterol-heavy LDL and remnant lipoproteins, exactly the species best equipped to damage an artery wall.

How much does 100 mg of dietary cholesterol raise LDL?

Now for calibration, because this is where honest science lives. In controlled feeding studies, each additional 100 milligrams of dietary cholesterol per day raises blood cholesterol by only a few points on average. Averages hide the spread: some people barely budge, while the hyper-responders and hyper-absorbers climb sharply. And in the real world, dietary cholesterol almost never travels alone. The foods that carry it, meat, eggs, butter, cheese, are usually the same foods rich in saturated fat, which suppresses LDL receptors even more strongly than cholesterol itself does. The package deal is what shows up in your bloodwork. How the two food kingdoms package their fat so differently, and why the animal-dominant fatty acids are the ones that suppress those receptors, is the whole subject of a companion piece on animal fat, plant fat, and the journey from your plate to your arteries.

Does blocking cholesterol absorption lower LDL on a keto diet?

If you want a hint that the absorption side of this system can dominate, a 2026 case series in Lipids in Health and Disease provides one, and in the interest of full disclosure it comes from the same low-carb research circle behind the retracted KETO-CTA study discussed later in this article. Patients whose LDL had skyrocketed on ketogenic diets were given ezetimibe, the drug that blocks the NPC1L1 customs gate. Their LDL fell by a median of 53 percent, roughly triple the drug’s usual monotherapy response of around 18 percent. Blocking cholesterol absorption at the gut collapsed the problem. The researchers proposed that the diet itself had reprogrammed cholesterol handling. This is a small, selected, uncontrolled group of 14 responders, not a randomized comparison, so it is a hypothesis to test rather than a settled effect. But the direction of the finding is hard to miss: what comes through the intestinal gate can matter enormously.

How Plaque Actually Forms

What is ApoB and why does it matter?

Most explanations of heart disease skip the crime scene. Here is what actually happens, and it is worth reading twice.

Atherosclerosis begins when ApoB-containing lipoproteins, which include LDL, remnants, and their relatives, slip across the endothelium, the one-cell-thick lining of your artery, into the inner arterial wall. ApoB is the large protein wrapped around each of these particles, one copy per particle, like a shipping label. On that protein sit positively charged amino acids, lysine and arginine, and the arterial wall’s structural mesh is built of proteoglycans carrying intensely negatively charged sulfate chains. Positive meets negative, and the two lock together in an electrostatic bond. The particle is trapped.

Why do cholesterol particles get stuck in artery walls?

This is called the response-to-retention model, and it is not a metaphor. In 2002, researchers published one of the most elegant experiments in the history of this field: they genetically engineered mice whose LDL could not bind proteoglycans, changing nothing else about their cholesterol levels. Those mice developed dramatically less atherosclerosis. Same cholesterol, no trapping, little disease. The trap is the disease.

What is a foam cell, and are there really cholesterol crystals in plaque?

Once trapped, the particle oxidizes, and the immune system flags it as toxic debris. Monocytes crawl into the wall and mature into macrophages, garbage-collector cells that gorge on the damaged particles until they become bloated foam cells. Foam cells eventually die and rupture, dumping their fatty cargo into a growing pool of dead material called the necrotic core. Inside that core, cholesterol concentrates until it precipitates into actual sharp-edged crystals, which stab at the surrounding tissue and activate an inflammatory alarm system called the NLRP3 inflammasome. Pathologists looking at plaque under the microscope literally see the needle-shaped voids those crystals leave behind. That is what a heart attack is made of.

A Possible Accelerator: TMAO

Does TMAO from meat and eggs cause heart disease?

There is a second player worth mentioning honestly, because it comes up constantly and deserves neither hype nor dismissal. When gut bacteria digest carnitine, concentrated in red meat, and choline, concentrated in eggs, they produce a compound the liver converts into trimethylamine-N-oxide, or TMAO. In large prospective studies, people with higher blood TMAO have had more cardiovascular events and deaths, and the pooled associations are not trivial. The mechanism proposed from animal work is that TMAO impairs the reverse-cholesterol-transport that clears the wall and nudges the whole trapping process along.

Why does fish raise TMAO more than meat?

The honest framing is that TMAO looks like an accelerator, not the engine. Two facts keep it in its lane. First, when researchers use genetics to test whether TMAO actually causes disease rather than merely marking it, the causal signal largely disappears, which is exactly the pattern of a passenger rather than a driver. Second, and awkwardly for the simple story, fish raises TMAO far more than meat or eggs do, roughly fiftyfold in feeding studies, yet fish is not harmful to the heart. So TMAO belongs in the picture as a plausible contributor and a useful marker, while ApoB remains the thing actually building the plaque. It is a footnote to the mechanism, not a rewrite of it.

Three Ways a Cholesterol-Heavy Diet Loads the Dice

Everything above compresses into three mechanisms.

Does dietary cholesterol raise your ApoB particle count?

More dietary cholesterol and saturated fat means fewer liver LDL receptors, which means more ApoB particles circulating at any moment. Plaque formation is a numbers game played every minute of your life: the more particles drifting past the wall, the more get in, and the more get trapped. This is why cardiologists increasingly measure ApoB directly, because it counts the particles themselves.

Is small dense LDL caused by dietary cholesterol?

Particles that linger get modified and oxidized, becoming stickier to the immune system. One clarification the internet usually gets wrong: the notorious small, dense LDL pattern is driven mostly by high triglycerides and insulin resistance, the territory of refined carbohydrates and excess calories, rather than by dietary cholesterol itself. What a cholesterol-rich diet reliably does is raise the particle number and enrich the remnants, which is damaging enough.

What is remnant cholesterol?

Absorbed dietary cholesterol is packaged directly into chylomicron remnants. Each remnant carries an ApoB label, is small enough to enter the arterial wall, and hauls a far larger cholesterol payload per particle than a standard LDL. When a remnant gets trapped, it delivers a truckload where LDL delivers a suitcase. This is not a footnote: large genetic studies from Copenhagen have shown that the cholesterol carried in these remnants is a causal driver of heart disease in its own right, independent of LDL, which is one reason a fasting lipid panel that ignores them understates risk.

A Word About HDL, Because Someone Will Bring It Up

Is high HDL actually protective?

High-cholesterol, high-fat diets often raise HDL, the so-called good cholesterol, and people wave that number like a shield. The best modern evidence says: put the shield down. Large genetic studies (a method called Mendelian randomization, which we will meet properly in a moment) show that people born with lifelong higher HDL levels do not have less heart disease, and drug after drug designed to raise HDL failed to deliver the expected protection; the one large trial that did show modest benefit produced it in proportion to the drug’s LDL lowering, not its HDL raising. HDL that rises under a saturated-fat-heavy, inflammatory dietary pattern can become protein-depleted, oxidized, and functionally lazy, losing the very anti-inflammatory, cholesterol-clearing behavior that made HDL famous.

ApoB or LDL-C: which number should you trust?

HDL is a system, not a score. ApoB is the score. And when your standard LDL-C number and your ApoB number disagree, which happens most in people with high triglycerides or diabetes, the outcome studies say to trust ApoB, because it counts the actual particles knocking on the artery wall.

One Particle Diet Cannot Touch: Lipoprotein(a)

What is Lp(a)?

Honesty cuts both ways, so here is the cholesterol risk factor that no salad will fix. Lipoprotein(a), written Lp(a) and said “L-P-little-a,” is an LDL-like particle carrying its own ApoB plus an extra tangled protein that makes it especially prone to lodging in artery walls and encouraging clots. It is an independent, causal driver of both heart attacks and calcific narrowing of the aortic valve.

Can diet lower Lp(a)?

Two facts make it important for a piece like this. First, roughly one in five people worldwide carries a high level, so it is common. Second, and unlike almost everything else in this article, your level is set overwhelmingly by the genes you were born with. Diet barely moves it. Exercise barely moves it. Even the statins that lower ordinary LDL do essentially nothing to it, though a new class of drugs built specifically to lower Lp(a) is now deep in testing, with the first cardiovascular-outcome results expected in 2026.

Should you get your Lp(a) tested?

So Lp(a) is the honest asterisk on the whole plant-forward argument. Eating well remains the highest-leverage thing most people can do for the particles diet can move, which is the overwhelming majority of the risk for the overwhelming majority of people. But a person can do everything right and still carry elevated Lp(a), which is exactly why it is worth measuring once in your life, since it almost never needs re-checking, so you know whether you are the exception who needs medical help beyond the plate.

Climbing the Whole Pyramid of Evidence

Science ranks its evidence. Mechanisms and lab work sit at the base, then animal experiments, then human observational studies, then randomized controlled trials, then systematic reviews and consensus statements built from all of it. A claim you can trust is one that survives at every level. Cholesterol’s role in heart disease is arguably the most thoroughly tested claim in all of medicine, so let’s climb.

Level 1: What does the cell biology show?

The base of the pyramid is everything you just read. The LDL receptor system (Brown and Goldstein, Nobel Prize, 1985). The discovery of the NPC1L1 absorption gate in 2004. The oxidation hypothesis, showing how modified LDL becomes immunologically toxic. The response-to-retention model of particles binding proteoglycans. Cholesterol crystals activating the NLRP3 inflammatory alarm. None of this is speculative. It is textbook cell biology, worked out in exhaustive detail over fifty years.

Level 2: What do the animal experiments show?

The proteoglycan-binding mouse experiment described above is the crown jewel: block the trapping, prevent the disease, without changing cholesterol levels at all. The broader animal literature goes back much further. In 1913, Nikolai Anitschkow fed cholesterol to rabbits and produced arterial lesions essentially indistinguishable from human atherosclerosis, and in the century since, the pattern has held with blunt consistency. Raise blood LDL high enough, by diet or by genetics, and virtually every susceptible species tested develops arterial plaque, which regresses when those levels fall. Species that resist atherosclerosis are the species whose LDL stays low.

Level 3: What do genetics and Mendelian randomization show?

Here is where it gets beautiful. You cannot ethically randomize babies to a lifetime of high or low cholesterol. Nature already did.

People born with familial hypercholesterolemia, about 1 in 250, carry broken LDL receptor machinery. Their LDL runs high from birth, and untreated, many suffer heart attacks decades early. Children with two broken copies can have heart attacks before their teens, on any diet.

Now flip the coin. In 2006, researchers studying the gene PCSK9 found people born with mutations that keep LDL modestly lower for life. A 28 percent lower LDL from birth came with an 88 percent reduction in coronary heart disease. Even a 15 percent lifelong reduction cut risk nearly in half. Then in 2014, geneticists found people born with naturally inactivated NPC1L1, the cholesterol absorption gate itself. Their LDL ran only about 12 points lower, yet their heart disease risk was roughly cut in half. These people are, in effect, born on a low-absorption diet, and their arteries show it.

This method, Mendelian randomization, uses randomly inherited genes as a lifelong experiment, immune to the confounding that plagues diet studies. Applied across hundreds of thousands of people, it shows that every unit of lifelong lower LDL delivers about three times the protection of the same reduction started late in life with a pill. The effect is graded, dose-dependent, and shows no threshold below which LDL stops mattering. Exposure over time is the disease. Cardiologists now compress this into a mantra: lower, earlier, longer.

Level 4: What do population studies show?

Zoom out to whole populations and the same picture appears.

Framingham, launched in 1948, put cholesterol on the map as a core risk factor. The MRFIT screening study then followed over 350,000 men and found the relationship between blood cholesterol and fatal heart disease is continuous and graded, with no safe plateau where risk levels off.

The Ni-Hon-San study answered the genetics objection: Japanese men living in Japan had low cholesterol and little heart disease; genetically similar Japanese men in Hawaii had more of both; those in California, eating the most Westernized diet, had the most. Same genes, different food, different arteries.

Does heart disease really start in childhood?

Autopsy studies like Bogalusa delivered the timeline: fatty streaks and early plaques are already present in children and young adults who died of unrelated causes, and the extent of those lesions tracks their LDL levels. This disease starts decades before the first symptom, which is why the American Academy of Pediatrics endorses cholesterol screening for kids at ages 9 to 11. If you are thinking about how this applies to a family, our guide to raising children on a plant-based diet covers the nutrition side in detail.

What do the longest-lived and lowest-heart-disease populations eat?

And then there are the Tsimane of the Bolivian Amazon, described in The Lancet in 2017 as having the lowest levels of coronary artery disease ever recorded in a studied population. Forager-farmers eating a diet around 14 percent fat, built overwhelmingly on plants (plantains, rice, corn, cassava, fruit) plus modest lean wild game and fish, their average LDL sits near 90 and their inflammation from infections is actually high, yet 85 percent of them showed zero coronary artery calcium. The researchers estimated their arteries aged roughly 28 years behind a comparable American population, so a Tsimane in his eighties tended to carry the coronary calcium of an American decades younger. The broader question of what the longest-living people actually eat is worth its own treatment, as is the recurring claim that the Blue Zones research has been debunked, and the specific Saul Newman critique that gets cited for it.

Do plant-based eaters have less heart disease?

Meanwhile the Adventist Health Study-2, following about 96,000 North Americans, found progressively lower cardiovascular and total mortality as diets moved further toward plants. That finding is sometimes waved away on the grounds that Adventists differ from everyone else in more ways than diet, so it is worth knowing it holds outside that community. In the general middle-aged population of the ARIC study, higher adherence to plant-centered eating tracked with less incident cardiovascular disease and lower mortality, and across more than 200,000 people in the Harvard cohorts, a crucial distinction emerged: a plant-based diet built on whole foods predicted less coronary disease, while one built on refined grains, sweets, and sugary drinks predicted more. Plant-based is not a synonym for healthy, and the data say so. A 2023 analysis of roughly 120,000 adults found that every major healthy-eating index, scored independently, predicted lower total and cardiovascular mortality, and a Women’s Health Initiative analysis found the Portfolio pattern, the same cholesterol-lowering food combination tested in the feeding trials above, associated with fewer actual cardiovascular events over years of follow-up.

Observational studies can be confounded, which is exactly why we keep climbing.

Level 5: What do controlled feeding trials show?

Lock people in metabolic wards, control every bite, and blood lipids become predictable enough to write equations, which is literally what Ancel Keys and Mark Hegsted did in the 1960s. The modern summary is the Mensink and Katan meta-analyses of dozens of controlled trials, later commissioned by the World Health Organization: saturated fat raises LDL, polyunsaturated fat lowers it, and dietary cholesterol raises it on average with wide individual variation. This is not epidemiology. This is cause and effect measured in locked kitchens.

How much can food alone lower cholesterol?

Feeding trials also revealed how far food alone can go. The DASH trial dropped blood pressure with fruits, vegetables, and low-fat foods in eight weeks. The Portfolio Diet trial went after LDL directly, stacking four plant tools (viscous fiber from oats, barley and psyllium, soy protein, plant sterols, and almonds) and cut LDL about 29 percent in four weeks, statistically indistinguishable from the 31 percent achieved by a starting dose of lovastatin running head-to-head in the same trial. A meta-analysis of plant sterols alone shows about 2 grams a day lowers LDL 8 to 10 percent, and soluble fiber adds its own reliable reduction. Individual foods carry their own randomized evidence on top of these: oat beta-glucan at 3 grams a day lowers LDL by roughly 5 to 10 points across dozens of trials, which is the basis of the oat health claims approved by the FDA, the European authorities, and Health Canada; soy protein and ground flaxseed each lower it a few points more in their own meta-analyses. Even blood vessels themselves respond: dietary nitrate from beets and greens produced sustained blood pressure drops of roughly 8 points systolic in randomized hypertensive patients.

Does unfiltered coffee raise cholesterol?

A quick aside that surprises people, since it shows the logic cuts both ways: not everything plant-based is automatically cholesterol-neutral. Unfiltered coffee, meaning French press, boiled, Turkish, and espresso, carries oily compounds called diterpenes that raise LDL, enough that heavy unfiltered drinkers can add 10 to 15 points. A paper filter traps them, which is why filtered and instant coffee do not have the effect. The point is not to fear coffee but to notice that the science follows the molecules wherever they lead, not a plant-versus-animal slogan.

Do omega-3 supplements raise LDL cholesterol?

The same honesty applies to a supplement plenty of plant-based readers take. DHA, the long-chain omega-3 sold as algal oil, modestly raises LDL. The effect is specific to DHA rather than EPA, and it runs around 7 to 8 percent in trials using roughly 0.7 to 1.7 grams a day, alongside lower triglycerides and higher HDL. It is partly offset because DHA also shifts LDL toward larger particles, but if your whole cardiovascular strategy is built on keeping LDL and ApoB low, which is the strategy this article has spent ten thousand words defending, it is worth knowing. Two things keep this in proportion. The doses that produce the effect are several times higher than the 250 milligrams a day that plant-based clinicians typically suggest. And the underlying question of whether a healthy vegan needs the supplement at all is genuinely unsettled, with the large trials having failed to move hard outcomes in either direction, which is the subject of a companion piece on whether vegans need DHA and EPA.

Level 6: What do randomized trials with hard endpoints show?

The top interventional tier asks the only question that ultimately matters: fewer heart attacks and deaths, yes or no?

Drug trials proved the pathway end to end. The 4S trial in 1994 cut total mortality 30 percent by lowering LDL. The Cholesterol Treatment Trialists’ meta-analysis of 26 randomized trials and 170,000 people nailed the exchange rate: a little over a fifth fewer major vascular events for every 39-point drop in LDL, and that benefit repeats each year the lowering is maintained. IMPROVE-IT showed that ezetimibe, the drug blocking the same NPC1L1 gate your dietary cholesterol uses, reduced cardiovascular events, demonstrating in a randomized trial that the absorption route itself is causally connected to outcomes. FOURIER drove LDL down to a median of 30 with a PCSK9 blocker and events kept falling, with no lower limit of benefit in sight. And in 2023, CLEAR Outcomes added a fourth independent mechanism: bempedoic acid, which throttles cholesterol synthesis one step upstream of where statins act, lowered LDL and cut cardiovascular events in nearly 14,000 statin-intolerant patients. Four different drug mechanisms now, each poking the system in a different place, one common denominator: fewer ApoB particles, fewer heart attacks.

Do diet trials show fewer heart attacks?

Diet trials with hard endpoints are harder to run, but they exist. The Los Angeles Veterans trial, 846 men fed institutionally for eight years under double-blind conditions, saw fewer atherosclerotic events in the group given vegetable oil in place of animal fat. The Oslo Diet-Heart study and the Finnish Mental Hospital study, both replacing saturated fat with unsaturated fat, reduced coronary events. The Lyon Diet Heart Study put heart attack survivors on a Mediterranean pattern and saw recurrent cardiac events fall by roughly half to 70 percent, a result so strong the trial was stopped early. PREDIMED, the largest modern diet trial, randomized about 7,400 high-risk people and found a Mediterranean pattern with extra olive oil or nuts cut major cardiovascular events by about 30 percent. And when the Cochrane Collaboration, the most methodologically conservative review body in medicine, pooled 15 randomized trials covering 56,675 participants in 2020, including the unfavorable ones, reducing saturated fat cut combined cardiovascular events by about 17 percent, with bigger cholesterol reductions producing bigger event reductions, exactly what the mechanism at Level 1 predicts.

What about the trials that failed?

And the honest tier includes the failures, because failures teach. The Women’s Health Initiative randomized nearly 49,000 women, about 19,500 of them assigned to a generically low-fat pattern, and saw no cardiovascular benefit: cutting fat while eating refined carbohydrates achieves nothing, so quality of replacement is everything. The recovered data from the Sydney and Minnesota experiments of the 1960s and 70s showed that swapping saturated fat for the omega-6 margarines of that era, some laced with trans fats, did not help either. And the great antioxidant-pill trials of the 1990s, ATBC and CARET, found that isolated beta-carotene capsules not only failed but increased lung cancer in smokers. The lesson repeats three times: the operative unit of nutrition is the whole food pattern, not an extracted nutrient in a capsule or an industrial substitution. The fuller version of that argument, including why the “seed oils are poison” claim fails at every rung of this same ladder, is laid out in our post on saturated fat, seed oils, and heart disease.

Level 7: Can arteries actually heal?

Dean Ornish’s Lifestyle Heart Trial randomized patients with established coronary disease to an intensive program built on a very-low-fat plant-based diet plus exercise and stress management. At one year, arteries in the lifestyle group had measurably widened while controls narrowed, and over five years the gap grew, with more cardiac events piling up in the control group. Caldwell Esselstyn’s plant-based case series points the same direction, and it is only fair to label it honestly: case series sit low on the pyramid, but they are consistent with everything above them. Meanwhile intravascular ultrasound trials of intensive drug therapy, ASTEROID and GLAGOV, showed that once LDL is pushed into the 60s, plaque volume physically shrinks. Multiple roads, one destination: get ApoB low enough, for long enough, and arteries stop getting worse and start getting better.

Level 8: What did the expert consensus panels conclude?

In 2017, the European Atherosclerosis Society convened its consensus panel to answer the question once and for all. They integrated over 200 studies covering more than 2 million participants and over 20 million person-years of follow-up, spanning genetics, epidemiology, and randomized trials. Their title was not hedged: “Low-density lipoproteins cause atherosclerotic cardiovascular disease.” Not associated with. Cause. The 2020 update deepened the mechanism and said it again. When separate methods with completely different weaknesses all land on the same answer, scientists call it consilience, and this is the cleanest example modern medicine has.

What the World’s Major Scientific Bodies Concluded

Walk the directory of major health and science organizations and notice how rare the disagreement is.

The National Academies (Institute of Medicine) set no requirement for dietary cholesterol and advised intake as low as possible within a nutritionally adequate diet, precisely because your body makes its own and the relationship between intake and heart disease risk is positive with no identified safe threshold.

Why did the 300 mg cholesterol limit disappear in 2015?

The Dietary Guidelines for Americans dropped its old 300-milligram daily cap in the 2015 cycle, a change endlessly misquoted as an exoneration. It helps to know that two different documents are involved. The advisory committee, a panel of scientists who spend years assembling the evidence, writes a scientific report. The government then writes the actual policy document. They are not the same thing, and the difference matters here.

The 2015 committee said the cap should not be carried forward, on the grounds that available evidence showed no appreciable relationship between dietary cholesterol and blood cholesterol, and that cholesterol was not a nutrient of concern for overconsumption. Two pieces of context rarely survive the screenshot. That phrasing overstated its own cited source, the 2013 review by the American Heart Association and the American College of Cardiology, which had actually concluded there was insufficient evidence, a weaker and different claim. And average American intake was already below 300 milligrams, so the number had largely stopped doing work.

Did the Dietary Guidelines say dietary cholesterol doesn’t matter?

The final Guidelines did not repeat the no-relationship line. They said explicitly that dropping the cap does not suggest dietary cholesterol is no longer important, and kept the National Academies’ advice to eat as little dietary cholesterol as possible while eating a healthy pattern, noting that the foods highest in it are generally the same foods highest in saturated fat. The 2020 to 2025 edition kept that language. And in 2019 the American Heart Association effectively corrected the committee’s wording, concluding that dietary cholesterol does raise LDL in a dose-dependent way, on the order of a couple of points per hundred milligrams under a linear model and more under nonlinear ones.

What did the 2025 scientific advisory committee actually conclude?

The 2025 Dietary Guidelines Advisory Committee, the scientific panel again, reaffirmed the saturated fat limit based on what it called a large body of consistent, high-quality evidence, and noted that more than 80 percent of Americans aged one and older already exceed it. Two of its underlying systematic reviews are worth knowing about, because they are exactly the sort of thing that never makes it into a screenshot. The committee was the first to formally evaluate saturated fat at the level of foods rather than nutrients, and it graded as strong the finding that replacing butter with plant oils and spreads lowers LDL. It also found, at moderate strength, that swapping red and processed meat for plant proteins like beans, lentils, nuts, and soy is associated with less cardiovascular disease. Its dietary patterns review, graded strong, concluded that patterns higher in vegetables, fruits, legumes, nuts, whole grains, and unsaturated fats, and lower in red and processed meat, refined grains, and sugary foods, are associated with lower cardiovascular risk, consistently across racial and socioeconomic groups. One point of provenance worth stating plainly: the headline conclusion that swapping saturated fat for polyunsaturated fat reduces coronary events was graded strong by the 2020 committee, and the 2025 committee carried it forward rather than rerunning it.

What changed in the 2025 to 2030 Dietary Guidelines?

Then came the part worth watching closely. The 2025 to 2030 Guidelines released in January 2026 dropped the dietary cholesterol language entirely, moved eggs to the front of the protein list, and floated butter and beef tallow as fat options, none of which came from the committee’s scientific report. What the political stage could not remove was the number that covers those same foods: saturated fat stays limited to under 10 percent of calories. The American College of Cardiology said the obvious thing out loud, that if you actually eat the newly recommended servings of animal protein and the proposed healthy fats, you will blow past that 10 percent limit, and researchers including Harvard’s Frank Hu called the document internally inconsistent. Christopher Gardner, a Stanford scientist who sat on the advisory committee itself, said of the accompanying food pyramid featuring red meat and cheese at the top that it goes against decades of evidence and research.

That is the real lesson about the guidelines, and it cuts against the way they are usually attacked. The weak link was never the scientists. It is what happens to their report afterward, in the room where industry has a seat and the panel does not. When a document written under pressure to rehabilitate animal fat still cannot delete the saturated fat cap, that tells you where the evidence floor is.

What do the AHA, WHO, and other health bodies say?

The American Heart Association’s 2017 Presidential Advisory concluded that replacing saturated fat with unsaturated fat lowers cardiovascular events on the order of what statins achieve, and its 2019 science advisory on dietary cholesterol recommended keeping intake low while acknowledging that healthy people can fit up to about an egg a day within an otherwise healthy pattern, with more caution for those with high LDL or diabetes.

The American College of Cardiology and AHA cholesterol guidelines, and the European Society of Cardiology and European Atherosclerosis Society guidelines, treat LDL and ApoB as causal targets, with the explicit principle that lower, earlier, and longer is better.

The World Health Organization (2023) recommends holding saturated fat under 10 percent of energy and replacing it with polyunsaturated fats and plant foods, alongside its long-running push to eliminate trans fats.

The International Agency for Research on Cancer classified processed meat as a Group 1 carcinogen, the same certainty category as tobacco, and red meat as probably carcinogenic, and the World Cancer Research Fund advises eating little, if any, processed meat. The American Cancer Society’s 2020 guideline centers vegetables, fruits, whole grains, and legumes while limiting red and processed meat. That classification gets challenged regularly, and we took apart one of the most-shared versions of that challenge in our look at why a single subgroup number did not overturn the meat and bowel cancer link.

The American Diabetes Association’s standards of care list Mediterranean and plant-based patterns among recommended eating approaches. The American Academy of Pediatrics endorses universal childhood lipid screening because the disease begins in childhood.

The Academy of Nutrition and Dietetics, the largest organization of food and nutrition professionals in the United States, approved a fresh position paper in January 2025, in effect through 2032: appropriately planned vegetarian and vegan dietary patterns are nutritionally adequate and can offer long-term health benefits, particularly for cardiometabolic disease.

The American College of Lifestyle Medicine goes furthest, recommending an eating plan based predominantly on a variety of minimally processed vegetables, fruits, whole grains, legumes, nuts, and seeds, as first-line treatment.

Health Canada’s 2019 food guide moved plant protein to the front of the plate. And the EAT-Lancet Commission, first in 2019 and again in its 2025 update built by experts from more than 35 countries, concluded that a plant-rich planetary health diet could prevent roughly 15 million premature deaths per year worldwide. The Global Burden of Disease project puts dietary risks at around 11 million deaths annually, driven mostly by too few whole plant foods and too much sodium and processed food.

That is not a conspiracy of agreement. That is what it looks like when the pyramid holds from bottom to top. If you want the broader picture of how a plant-predominant diet performs across every nutrient and outcome, not just lipids, start with our overview of plant-based nutrition.

Why “But I Feel Great” Cannot Settle This

Can you feel plaque building in your arteries?

Here is the cruelest design flaw in human anatomy: the inside of your arteries cannot feel anything.

The endothelium has no pain receptors. You cannot feel an ApoB particle slipping through the lining. You cannot feel the electrostatic snap of a positively charged protein locking onto a negatively charged proteoglycan. You cannot feel a macrophage bloating into a foam cell. A plaque can grow silently for thirty or forty years without narrowing the artery enough to cause chest pain, and remember, the Bogalusa autopsies found the process already underway in teenagers. For a shocking number of people, the very first symptom of heart disease is the heart attack.

Is erectile dysfunction an early warning sign of heart disease?

There is one early warning worth knowing about, though, at least for men: because the arteries feeding the penis are narrower than the coronary arteries, the same systemic plaque process tends to show up there first, and erectile dysfunction precedes a cardiac diagnosis in a large share of cases, often by several years. It is less a separate problem than the same disease knocking on an earlier door.

Why does losing weight make a bad diet look good?

This is why the “I feel amazing” testimonial, the backbone of every diet movement, is scientifically worthless as a safety claim. And there is a specific illusion worth naming. When someone drops weight on a meat-heavy, low-carb diet, shrinking fat tissue genuinely lowers systemic inflammatory signals like IL-6 and CRP. Blood sugar steadies. Energy climbs. Every dashboard light the person can feel turns green. Meanwhile, if that diet is loaded with saturated fat and cholesterol, LDL receptors stay suppressed, ApoB particles stay elevated, and the silent trapping inside the arterial wall continues every minute, felt by no one. Systemic inflammation and local arterial inflammation are two different fires, and losing weight only douses one of them. Tools like the Dietary Inflammatory Index consistently score whole plant foods, with their fiber and polyphenols, as anti-inflammatory, while patterns heavy in processed and high-heat-cooked animal products, with their advanced glycation end-products, heme iron, and saturated fat, lean pro-inflammatory. This particular trap, weight loss flattering every marker you can see while the ones you cannot see quietly worsen, is worked through in detail in a companion piece on why a whole-food plant-based diet is the easiest way to reach and keep a healthy weight, which also makes the case that the same mechanism taking the weight off is the one lowering your ApoB, rather than the two working against each other.

What did the KETO-CTA study actually find?

The low-carb world is currently running this exact experiment on itself. Lean, athletic keto and carnivore followers, so-called lean mass hyper-responders, routinely post LDL levels of 300 to 500 alongside glowing energy and perfect glucose numbers, and argue the LDL must be harmless in their context. Their own community crowdfunded a CT imaging study of 100 such people, the KETO-CTA trial, to make the case. It did not cooperate. Over a single year, coronary plaque advanced in this group, with the amount of soft plaque already present at the start being the strongest predictor of how much more accumulated. The trial’s first published paper, titled to argue that plaque predicts plaque while ApoB does not, was then retracted in 2026, and the retraction is more revealing than the paper: the authors themselves requested it after concluding that the outside vendor’s automated plaque measurements were unreliable. When the pre-registered results were re-reported using different plaque-measurement software, the progression was still there: a median noncalcified plaque increase of 5.6 cubic millimeters, a 37 percent relative rise over one year. About 15 percent of participants did show some regression, and that re-report remains a non-peer-reviewed preprint. The much-quoted claim that ApoB failed to predict who progressed is close to meaningless in a group where everyone’s ApoB was sky-high already, which is like finding that height fails to predict basketball skill in a room full of people who are all seven feet tall. And in 2026, researchers from that same circle published the ezetimibe paper described earlier, actively studying how to bring the number down. Feeling great is a mood. ApoB is a measurement. Only one of them predicts your arteries.

The Nitrate Paradox: Same Atoms, Opposite Fates

Are the nitrates in vegetables bad for you?

One oxygen atom separates nitrate from nitrite, yet in food they behave like different substances, because chemistry never acts alone. It acts inside a food matrix.

Dietary nitrate from leafy greens and beets is concentrated by your salivary glands, converted by oral bacteria into nitrite, and then transformed in the body into nitric oxide, the master signaling molecule that relaxes blood vessels, lowers blood pressure, and helps maintain the endothelial lining that keeps lipoproteins moving instead of sticking. Randomized trials of dietary nitrate show real, sustained blood pressure reductions. Crucially, plants deliver their nitrate bundled with vitamin C and polyphenols, which block those nitrogen compounds from taking a darker chemical path.

Why are the nitrites in processed meat different?

Meat is the mirror image. It contains little beneficial nitrate but is often cured with added nitrite, which meets the abundant amines of animal protein, frequently under high cooking heat and catalyzed by heme iron, and forms nitrosamines, among the more potent DNA-damaging, inflammation-driving compounds in the food supply. This chemistry sits behind the IARC’s decision to classify processed meat as a Group 1 carcinogen. Same nitrogen, opposite outcomes, and the difference is the plant matrix wrapped around it. If the mirror-image worry is on your mind, that plant foods carry their own dangerous compounds, we handled that argument separately in our post on plant toxins and antinutrients.

What Plant Foods Actually Do, Mechanism by Mechanism

Everything in this article so far describes a trap: particles in the blood, charges on a protein, a mesh in the wall. Plant-predominant eating disarms that trap at nearly every stage, and the mechanisms are worth spelling out.

1. Do plants contain cholesterol, and what are phytosterols?

Plants contain essentially zero cholesterol. Every gram of plant food displacing animal food is dietary cholesterol that never enters the customs gate. Better still, plants carry phytosterols, cholesterol look-alikes that compete for space in the micelles and at the NPC1L1 gate, so cholesterol gets crowded out of absorption. Concentrated to about 2 grams a day, plant sterols alone cut LDL 8 to 10 percent in randomized trials.

2. How does fiber lower cholesterol?

Remember that your own bile pours roughly a gram of cholesterol and a load of bile acids into the gut daily, and most of it normally gets reabsorbed and recycled. Soluble fiber from oats, barley, beans, and psyllium binds those bile acids and escorts them out of the body. The liver must then pull cholesterol out of your bloodstream to manufacture replacement bile. This is not a folk theory: it is the exact working mechanism of an entire prescription drug class, the bile acid sequestrants. Oatmeal and beans are, quite literally, edible versions of that drug, and this is the mechanism that reaches even the cholesterol your body made itself.

3. Does swapping saturated fat for unsaturated fat lower LDL?

Replacing saturated fat with the unsaturated fats of nuts, seeds, olives, and avocados causes the liver to display more LDL receptors, the vacuum cleaners come back online, and LDL and ApoB fall. This is the most consistent finding in the entire controlled-feeding literature. Polyunsaturated fats also signal through liver receptors like PPAR-alpha to reshape the particles the liver ships out, and larger, more buoyant particles appear somewhat less prone to retention in the wall, though the headline effect is simpler: fewer ApoB particles in circulation, period. Two of the loudest objections to this swap, that seed oils are inflammatory and that coconut oil is a health food, each get a full evidence-pyramid treatment of their own: saturated fat, seed oils, and heart disease and coconut oil versus animal fat.

Do you need to eat fat to absorb antioxidants?

Before we get to what these compounds do inside a lipoprotein, there is a step that has to happen first, and it explains why a plate of undressed vegetables is not the same as a plate of vegetables eaten with a little fat. Carotenoids are locked inside plant cells and are stubbornly insoluble in water, so they reach your bloodstream by exactly the route this article already walked you through for cholesterol. Fat in the meal triggers bile release, the bile and fat assemble into mixed micelles, the micelles ferry the freed carotenoids to the intestinal wall, and the cells pull them in using the very same transporters that handle cholesterol, NPC1L1 among them. From there they are packed into chylomicrons, ride the lymphatic system up through the thoracic duct, and only then get handed off to LDL and HDL in circulation. No fat, no micelles, no delivery.

This is not a subtle effect. In one controlled crossover study, salads eaten with fat-free dressing produced almost no measurable carotenoid absorption at all, essentially a flat line, while the same salads with full-fat dressing produced a large rise that kept climbing as the fat went up. In another, adding avocado to a salad multiplied beta-carotene absorption more than fifteenfold, roughly quintupled lutein, and, importantly for the whole-food argument, the avocado fruit worked just as well as extracted avocado oil. Cooking helps too, by breaking down the cell walls that trap these compounds: lycopene absorption from tomato paste ran several times higher than from the same amount of fresh tomato. Heat and fat together, as in a simmered sauce with a little olive oil, are better than either alone.

The practical amount is small. Roughly three to five grams of fat in a meal, about what is in a spoonful of tahini, a quarter of an avocado, or a small handful of walnuts, is enough to unlock most of it, though a large pile of vegetables keeps absorbing more with more fat. Two honest limits on this advice. It applies to the fat-soluble compounds, so the carotenoids, vitamin E, vitamin K from greens, and provitamin A, and not to water-soluble ones like vitamin C and the anthocyanins, which do not need fat at all. And responses vary a lot between people, with some individuals absorbing very little regardless. It is also the one real caveat on very-low-fat plant-based eating, and it is easily solved without reaching for refined oil or animal products, since whole-food fats do the job.

4. How do antioxidants protect LDL particles?

So that is how the cargo gets loaded. Fat-soluble antioxidants from plants, carotenoids like lycopene and beta-carotene, ride inside the lipid core of LDL particles, alongside the body’s own coenzyme Q10, while water-soluble compounds like vitamin C and anthocyanins patrol the surrounding plasma. They sacrifice themselves to neutralize free radicals before those radicals can oxidize the particle’s fats or warp its ApoB protein.

Here is why that matters for the trap specifically. The initial ionic handshake between ApoB and a proteoglycan is reversible: a native, healthy particle that gets snagged can let go and drift back into circulation unharmed. What makes retention permanent is what happens next. Stuck in the subendothelial space, cut off from the plasma’s antioxidant supply, a snagged particle begins to oxidize, and oxidized particles aggregate, fuse, bind to the matrix through additional mechanisms, and get devoured by macrophages, none of which can be undone. Worse, the oxidation products inflame the wall and push it to manufacture even more proteoglycan mesh, so every particle that turns toxic strengthens the trap waiting for the next one. A particle that arrives loaded with carotenoids and vitamin E is carrying its own fire extinguisher into the one compartment where the bloodstream cannot help it: it resists the conversion, stays detachable, and starves that feedback loop. One precision note for the careful reader: heavy oxidation does not tighten the original ionic bond itself (it actually degrades the positively charged lysines), so the honest statement is not that oxidation strengthens the handshake, but that oxidation converts a temporary snag into permanent capture through aggregation, immune uptake, and wall remodeling.

Why do antioxidants need to be resupplied?

There is a detail in that last paragraph worth slowing down for, because it explains why the arterial wall is such a dangerous neighborhood. When we say an antioxidant sacrifices itself, that is not a metaphor. It is an electron transfer. Vitamin E quenches a lipid radical by handing over an electron, and in doing so becomes a mild radical itself. It does not have to stay spent, because the body runs a regeneration relay: vitamin C restores the vitamin E, glutathione helps restore the vitamin C, and glutathione itself is restored by NADPH, whose electrons come from glucose metabolism. Antioxidant protection is not a fixed tank that drains. It is a resupply line, and the line is powered by the same fuel oxidation that produced the radicals in the first place. One honest note for the careful reader: the vitamin C to vitamin E step is well demonstrated in the test tube, and researchers still debate how much of it operates in living tissue. The glutathione and NADPH steps are settled biochemistry.

Now recall where a retained particle sits. Wedged in the subendothelial space, it is separated from that circulating relay. Its onboard vitamin E and carotenoids can absorb the first hits, but there is no ready resupply to regenerate them. The extinguishers discharge and nobody refills them, which fits the response-to-retention model exactly. Retention is not only a location problem. It is a supply-line problem.

This is also why coenzyme Q10 keeps turning up on both sides of the story. CoQ10 is a working component of the mitochondrial electron transport chain, ferrying electrons as part of energy production, and it is simultaneously a lipid antioxidant carried inside LDL particles, where it shields the particle’s fats from peroxidation. In supplementation studies, LDL lipid oxidation stays slow until the ubiquinol aboard has been nearly all consumed. The same molecule is both a piece of the engine and a piece of the fire suppression.

Which reframes the dietary question. Reactive oxygen species are not a malfunction. They arise from ordinary electron leak at the transport chain, the unavoidable byproduct of running an oxygen-powered metabolism. You cannot stop generating them. What you decide, three times a day, is whether the food arriving alongside that metabolism brings replacement crew. Whole plant foods deliver carotenoids, tocopherols, vitamin C, and polyphenols packed in with their calories. Animal foods, as the companion piece on dietary fat documents, arrive with very little.

Do antioxidant supplements work as well as antioxidant foods?

Diet also rewrites the particle’s flammability from the inside. LDL whose core is built from oleic acid, the monounsaturated fat of olives, avocados, and most nuts, is measurably harder to oxidize in laboratory assays than LDL enriched in more chemically fragile fats. And notice what nature did with its most oxidizable fats: it packed the polyunsaturated oils of nuts and seeds inside the food supply’s richest sources of vitamin E, shipping the fuel and the fire suppression in the same shell. The matrix, again. Early laboratory work even suggests some polyphenols may interfere with the ApoB-proteoglycan binding step itself, though that line of research is young and should be labeled as such. And one caution the pyramid taught us: this protection travels in food, not capsules. Isolated antioxidant pills failed their trials. The matrix is the medicine.

5. Can diet stabilize existing plaque?

Polyphenols stimulate the endothelium to produce more nitric oxide, measurably improving arterial function in human trials, tightening the leaky lining that let particles through in the first place. Plant-heavy patterns lower inflammatory signals like CRP, slowing the recruitment of new macrophages to existing plaques. No diet dissolves a calcified core, and no honest writer will tell you otherwise. But shrinking the soft lipid core and thickening the fibrous cap converts a rupture-prone plaque into a stable one, and rupture, not narrowing, is what kills. The regression trials, from Ornish’s lifestyle program to the intensive LDL-lowering imaging studies, show arteries genuinely improving once particle counts fall far enough for long enough.

Put plainly: plant-forward eating lowers how many particles are circulating, changes what those particles carry, protects them from turning toxic, and repairs the wall they threatened. It is the only dietary pattern that addresses every stage of the trap at once, which is presumably why it keeps winning at every level of the evidence pyramid.

Where Honest Scientists Still Argue

No credible overview pretends the field has zero live debates, so here are the real ones, stated fairly.

Are eggs actually bad for you?

Dietary cholesterol by itself is the most modest lever in the chain. Its average effect on blood levels is smaller than saturated fat’s, individual responses vary enormously, and the egg literature in particular is genuinely mixed. The split is real and worth seeing: one large analysis of US cohorts found that more dietary cholesterol and more eggs tracked with modestly higher cardiovascular risk, while a set of three big prospective cohorts with a meta-analysis found no association between up to one egg a day and heart disease in Western populations. That disagreement is exactly why the AHA allows an egg a day for healthy people and why guidelines pivoted to food patterns instead of milligram caps. The strongest, least disputed claim was never “an egg will kill you.” It is that ApoB-containing particles cause atherosclerosis, and that the overall package of cholesterol-and-saturated-fat-rich animal foods raises them, while plant-predominant patterns lower them.

What about the PURE study and full-fat dairy?

Some observational findings genuinely complicate the simple story. The large PURE cohort found weaker links between saturated fat and mortality across 18 countries, and full-fat fermented dairy keeps turning up more neutral than its saturated fat content predicts, which is a real finding about food matrices rather than a rehabilitation of saturated fat, and notably the same butter that sits inside those neutral total-dairy averages shows clear harm when it is studied on its own. The best reading of the total evidence is that the replacement matters: swapping saturated fat for whole-food unsaturated fats and intact carbohydrates helps, swapping it for refined starch and sugar does not, which is exactly what the Women’s Health Initiative failure demonstrated.

Does keto work?

Low-carb diets produce real short-term wins. Weight loss, improved glycemic control, higher energy. Those results are genuine and worth acknowledging. The unresolved cost is long-term ApoB exposure in the people whose LDL climbs, and the early imaging data from the low-carb community’s own trial is, so far, not reassuring.

What could the evidence reviewers not conclude?

Honesty here means reporting what the same committees could not conclude. When the federal reviewers assessed dietary cholesterol and cardiovascular disease directly, they judged the evidence insufficient to assign a grade at all, precisely because dietary cholesterol travels with saturated fat in real food and the two are hard to separate. They also could not grade higher-fat versus lower-fat dairy, could not draw a conclusion on ultra-processed foods and cardiovascular disease, and when they attempted the food-level saturated fat question as a whole, they found the overall body of evidence too thin for a verdict, even while individual comparisons within it graded strong. Those gaps are real. What they do not do is undercut the causal core, which rests on genetics, mechanism, and drug trials rather than on food-frequency questionnaires.

What is not seriously disputed?

The core: LDL and ApoB cause atherosclerotic disease, established by consilience across genetics, mechanism, epidemiology, and randomized trials; lowering them, by any of several mechanisms including diet, lowers events; and no major scientific body anywhere on Earth recommends eating more cholesterol or more saturated fat.

The Bottom Line

• Your body manufactures every milligram of cholesterol it needs. The dietary requirement is zero, per the National Academies.
• Heart disease is a particle-trapping disease. ApoB particles enter the artery wall, bind electrostatically to proteoglycans, oxidize, and build plaque. Block the trapping and disease largely fails to develop.
• Cholesterol-and-saturated-fat-rich eating raises the number of trapped particles by suppressing LDL receptors and enriching remnants. The process is silent for decades because arteries cannot feel.
• Every tier of evidence, from Nobel-winning cell biology to genetic natural experiments to metabolic ward feeding studies to randomized trials to the consensus of essentially every major scientific body across multiple countries, points the same direction.
• Plant-predominant eating attacks the mechanism at all five stages: zero cholesterol in, bile drained out, receptors restored, particles shielded from oxidation, endothelium healed. Its results show up in everything from four-week LDL trials to multi-decade population outcomes.
• Feeling great is not a cardiovascular test. ApoB is. If you take one action from this article, ask for that number, and ask once for your lipoprotein(a), the one cholesterol particle diet cannot fix.

Keep Reading

This post is the molecular core of a larger series that walks the same evidence pyramid through the dietary fat debate. If cholesterol was your way in, these take the neighboring threads further.

• Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries follows a fatty meal from bile to chylomicron to liver, contrasts how the animal and plant kingdoms package their fat, and dismantles the “I feel great on keto” illusion that this post’s silent-damage section only introduces.
• Saturated Fat, Seed Oils, and Heart Disease climbs every rung of the evidence hierarchy to answer whether replacing saturated fat with plant oils helps, and takes the “seed oils are poison” claim apart level by level.
• Coconut Oil vs. Animal Fat settles the saturated fat that keeps getting a health halo, and shows why “but it raises HDL” cannot rescue it, an argument this post’s HDL section sets up.
• Why a Whole-Food Plant-Based Diet Is the Easiest Way to Lose Weight takes on the mirror image of this post’s warning: almost any diet that removes weight will flatter the markers you can see, which is exactly why the silent ones matter.
• Plant-Based Nutrition is the wider overview, covering every nutrient and outcome rather than lipids alone, and the vitamin B12 question covers the one supplement this way of eating genuinely requires.
• The Adventist Health Studies and what the longest-living people eat go deeper on the population evidence this post summarizes in a paragraph.

For readers who want to explore the primary research further, NutritionFacts.org maintains a free, ad-free library of evidence reviews.

This article is for education, not personal medical advice. If you take lipid-lowering medication or have existing heart disease, work with your clinician before making major dietary changes.

References

1. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science. 1986;232:34-47. doi:10.1126/science.3513311
2. Dietschy JM, Turley SD. Cholesterol metabolism in the central nervous system during early development and in the mature animal. J Lipid Res. 2004;45:1375-1397. doi:10.1194/jlr.R400004-JLR200
3. Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. National Academies Press; 2005.
4. Altmann SW, et al. Niemann-Pick C1 Like 1 protein is critical for intestinal cholesterol absorption. Science. 2004;303:1201-1204. doi:10.1126/science.1093131
5. Williams KJ, Tabas I. The response-to-retention hypothesis of early atherogenesis. Arterioscler Thromb Vasc Biol. 1995;15:551-561. doi:10.1161/01.atv.15.5.551
6. Skalen K, et al. Subendothelial retention of atherogenic lipoproteins in early atherosclerosis. Nature. 2002;417:750-754. doi:10.1038/nature00804
7. Steinberg D, et al. Beyond cholesterol: modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 1989;320:915-924. doi:10.1056/NEJM198904063201407
8. Duewell P, et al. NLRP3 inflammasomes are required for atherogenesis and activated by cholesterol crystals. Nature. 2010;464:1357-1361. doi:10.1038/nature08938
9. Cohen JC, Boerwinkle E, Mosley TH Jr, Hobbs HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. N Engl J Med. 2006;354:1264-1272. doi:10.1056/NEJMoa054013
10. Myocardial Infarction Genetics Consortium Investigators. Inactivating mutations in NPC1L1 and protection from coronary heart disease. N Engl J Med. 2014;371:2072-2082. doi:10.1056/NEJMoa1405386
11. Ference BA, et al. Effect of long-term exposure to lower low-density lipoprotein cholesterol beginning early in life on the risk of coronary heart disease: a Mendelian randomization analysis. J Am Coll Cardiol. 2012;60:2631-2639. doi:10.1016/j.jacc.2012.09.017
12. Ference BA, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2017;38:2459-2472. doi:10.1093/eurheartj/ehx144
13. Boren J, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease: pathophysiological, genetic, and therapeutic insights. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2020;41:2313-2330. doi:10.1093/eurheartj/ehz962
14. Kannel WB, Dawber TR, Kagan A, Revotskie N, Stokes J. Factors of risk in the development of coronary heart disease: the Framingham Study. Ann Intern Med. 1961;55:33-50. doi:10.7326/0003-4819-55-1-33
15. Stamler J, Wentworth D, Neaton JD. Is the relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 1986;256:2823-2828. doi:10.1001/jama.1986.03380200061022
16. Kagan A, et al. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California. J Chronic Dis. 1974;27:345-364. doi:10.1016/0021-9681(74)90014-9
17. Kaplan H, et al. Coronary atherosclerosis in indigenous South American Tsimane: a cross-sectional cohort study. Lancet. 2017;389:1730-1739. doi:10.1016/S0140-6736(17)30752-3
18. Orlich MJ, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Intern Med. 2013;173:1230-1238. doi:10.1001/jamainternmed.2013.6473
19. Berenson GS, et al. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med. 1998;338:1650-1656. doi:10.1056/NEJM199806043382302
20. Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr. 2003;77:1146-1155. doi:10.1093/ajcn/77.5.1146
21. Mensink RP; World Health Organization. Effects of Saturated Fatty Acids on Serum Lipids and Lipoproteins: A Systematic Review and Regression Analysis. WHO; 2016.
22. Hegsted DM, McGandy RB, Myers ML, Stare FJ. Quantitative effects of dietary fat on serum cholesterol in man. Am J Clin Nutr. 1965;17:281-295. doi:10.1093/ajcn/17.5.281
23. Appel LJ, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med. 1997;336:1117-1124. doi:10.1056/NEJM199704173361601
24. Jenkins DJ, et al. Effects of a dietary portfolio of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive protein. JAMA. 2003;290:502-510. doi:10.1001/jama.290.4.502
25. Ras RT, Geleijnse JM, Trautwein EA. LDL-cholesterol-lowering effect of plant sterols and stanols across different dose ranges: a meta-analysis of randomised controlled studies. Br J Nutr. 2014;112:214-219. doi:10.1017/S0007114514000750
26. Brown L, Rosner B, Willett WW, Sacks FM. Cholesterol-lowering effects of dietary fiber: a meta-analysis. Am J Clin Nutr. 1999;69:30-42. doi:10.1093/ajcn/69.1.30
27. Kapil V, Khambata RS, Robertson A, Caulfield MJ, Ahluwalia A. Dietary nitrate provides sustained blood pressure lowering in hypertensive patients: a randomized, phase 2, double-blind, placebo-controlled study. Hypertension. 2015;65:320-327. doi:10.1161/HYPERTENSIONAHA.114.04675
28. Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383-1389. doi:10.1016/S0140-6736(94)90566-5
29. Cholesterol Treatment Trialists (CTT) Collaboration. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet. 2010;376:1670-1681. doi:10.1016/S0140-6736(10)61350-5
30. Cannon CP, et al. Ezetimibe added to statin therapy after acute coronary syndromes (IMPROVE-IT). N Engl J Med. 2015;372:2387-2397. doi:10.1056/NEJMoa1410489
31. Sabatine MS, et al. Evolocumab and clinical outcomes in patients with cardiovascular disease (FOURIER). N Engl J Med. 2017;376:1713-1722. doi:10.1056/NEJMoa1615664
32. Leren P. The Oslo Diet-Heart Study: eleven-year report. Circulation. 1970;42:935-942. doi:10.1161/01.cir.42.5.935
33. Turpeinen O, et al. Dietary prevention of coronary heart disease: the Finnish Mental Hospital Study. Int J Epidemiol. 1979;8:99-118. doi:10.1093/ije/8.2.99
34. de Lorgeril M, et al. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation. 1999;99:779-785. doi:10.1161/01.cir.99.6.779
35. Estruch R, et al. Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. N Engl J Med. 2018;378:e34. doi:10.1056/NEJMoa1800389
36. Howard BV, et al. Low-fat dietary pattern and risk of cardiovascular disease: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial. JAMA. 2006;295:655-666. doi:10.1001/jama.295.6.655
37. Ramsden CE, et al. Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis. BMJ. 2013;346:e8707. doi:10.1136/bmj.e8707
38. Ramsden CE, et al. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968-73). BMJ. 2016;353:i1246. doi:10.1136/bmj.i1246
39. Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med. 1994;330:1029-1035. doi:10.1056/NEJM199404143301501
40. Omenn GS, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease (CARET). N Engl J Med. 1996;334:1150-1155. doi:10.1056/NEJM199605023341802
41. Ornish D, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet. 1990;336:129-133. doi:10.1016/0140-6736(90)91656-U
42. Ornish D, et al. Intensive lifestyle changes for reversal of coronary heart disease. JAMA. 1998;280:2001-2007. doi:10.1001/jama.280.23.2001
43. Esselstyn CB Jr, Gendy G, Doyle J, Golubic M, Roizen MF. A way to reverse CAD? J Fam Pract. 2014;63:356-364b. PubMed 25198208
44. Nissen SE, et al. Effect of very high-intensity statin therapy on regression of coronary atherosclerosis: the ASTEROID trial. JAMA. 2006;295:1556-1565. doi:10.1001/jama.295.13.jpc60002
45. Nicholls SJ, et al. Effect of evolocumab on progression of coronary disease in statin-treated patients: the GLAGOV randomized clinical trial. JAMA. 2016;316:2373-2384. doi:10.1001/jama.2016.16951
46. Voight BF, et al. Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study. Lancet. 2012;380:572-580. doi:10.1016/S0140-6736(12)60312-2
47. Sacks FM, et al. Dietary fats and cardiovascular disease: a presidential advisory from the American Heart Association. Circulation. 2017;136:e1-e23. doi:10.1161/CIR.0000000000000510
48. Carson JAS, et al. Dietary cholesterol and cardiovascular risk: a science advisory from the American Heart Association. Circulation. 2020;141:e39-e53. doi:10.1161/CIR.0000000000000743
49. Grundy SM, et al. 2018 AHA/ACC multisociety guideline on the management of blood cholesterol. Circulation. 2019;139:e1082-e1143. doi:10.1161/CIR.0000000000000625
50. Mach F, et al. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J. 2020;41:111-188. doi:10.1093/eurheartj/ehz455
51. World Health Organization. Saturated Fatty Acid and Trans-Fatty Acid Intake for Adults and Children: WHO Guideline. 2023.
52. Bouvard V, et al; International Agency for Research on Cancer Monograph Working Group. Carcinogenicity of consumption of red and processed meat. Lancet Oncol. 2015;16:1599-1600. doi:10.1016/S1470-2045(15)00444-1
53. Rock CL, et al. American Cancer Society guideline for diet and physical activity for cancer prevention. CA Cancer J Clin. 2020;70:245-271. doi:10.3322/caac.21591
54. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian dietary patterns for adults: a position paper of the Academy of Nutrition and Dietetics. J Acad Nutr Diet. 2025;125(6):831-846. doi:10.1016/j.jand.2025.02.002
55. American College of Lifestyle Medicine. Dietary Lifestyle Position Statement for Treatment and Potential Reversal of Disease.
56. US Department of Agriculture, US Department of Health and Human Services. Dietary Guidelines for Americans, 2025-2030. Released January 2026. dietaryguidelines.gov
57. Dietary Guidelines Advisory Committee. Scientific Report of the 2025 Dietary Guidelines Advisory Committee. 2024. doi:10.52570/DGAC2025
58. Willett W, et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet. 2019;393:447-492. doi:10.1016/S0140-6736(18)31788-4
59. Rockstrom J, Thilsted SH, Willett WC, et al. The EAT-Lancet Commission on healthy, sustainable, and just food systems. Lancet. 2025;406:1625-1700.
60. GBD 2017 Diet Collaborators (Afshin A, et al). Health effects of dietary risks in 195 countries, 1990-2017: a systematic analysis for the Global Burden of Disease Study. Lancet. 2019;393:1958-1972. doi:10.1016/S0140-6736(19)30041-8
61. Shivappa N, Steck SE, Hurley TG, Hussey JR, Hebert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr. 2014;17:1689-1696. doi:10.1017/S1368980013002115
62. Hord NG, Tang Y, Bryan NS. Food sources of nitrates and nitrites: the physiologic context for potential health benefits. Am J Clin Nutr. 2009;90:1-10. doi:10.3945/ajcn.2008.27131
63. Gonzalez J, Wageman J, Buchanan L, Kalayjian T, Calkins M, Saner E, Soto-Mota A. Increased LDL-C reduction with ezetimibe in ketogenic diet-induced hypercholesterolemia. Lipids Health Dis. 2026. doi:10.1186/s12944-026-03034-w
64. Dehghan M, et al. Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): a prospective cohort study. Lancet. 2017;390:2050-2062. doi:10.1016/S0140-6736(17)32252-3
65. Health Canada. Canada’s Food Guide. 2019. food-guide.canada.ca
66. Kronenberg F, et al. Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement. Eur Heart J. 2022;43:3925-3946. doi:10.1093/eurheartj/ehac361
67. Reyes-Soffer G, et al. Lipoprotein(a): a genetically determined, causal, and prevalent risk factor for atherosclerotic cardiovascular disease: a scientific statement from the American Heart Association. Arterioscler Thromb Vasc Biol. 2022;42:e48-e60. doi:10.1161/ATV.0000000000000147
68. Nissen SE, et al. Bempedoic acid and cardiovascular outcomes in statin-intolerant patients (CLEAR Outcomes). N Engl J Med. 2023;388:1353-1364. doi:10.1056/NEJMoa2215024
69. Zhong VW, et al. Associations of dietary cholesterol or egg consumption with incident cardiovascular disease and mortality. JAMA. 2019;321:1081-1095. doi:10.1001/jama.2019.1572
70. Drouin-Chartier JP, et al. Egg consumption and risk of cardiovascular disease: three large prospective US cohort studies, systematic review, and updated meta-analysis. BMJ. 2020;368:m513. doi:10.1136/bmj.m513
71. Varbo A, et al. Remnant cholesterol as a causal risk factor for ischemic heart disease. J Am Coll Cardiol. 2013;61:427-436. doi:10.1016/j.jacc.2012.08.1026
72. Nordestgaard BG. Triglyceride-rich lipoproteins and atherosclerotic cardiovascular disease: new insights from epidemiology, genetics, and biology. Circ Res. 2016;118:547-563. doi:10.1161/CIRCRESAHA.115.306249
73. Sniderman AD, et al. Apolipoprotein B particles and cardiovascular disease: a narrative review. JAMA Cardiol. 2019;4:1287-1295. doi:10.1001/jamacardio.2019.3780
74. Soto-Mota A, Norwitz NG, Manubolu VS, et al. RETRACTED: Longitudinal data from the KETO-CTA study: plaque predicts plaque, ApoB does not. JACC Adv. 2025;4:101686. (Retracted 2026 at the request of the authors and the editors.) doi:10.1016/j.jacadv.2025.101686
75. Budoff MJ, Kinninger A, Manubolu VS, Norwitz NG, Feldman D, Soto-Mota A. The impact of sustained LDL-C elevation on plaque changes: primary coronary plaque progression results from the Keto CTA study. medRxiv. Posted January 16, 2026. doi:10.64898/2026.01.15.26343955 (Preprint; not peer-reviewed.)
76. Zhang Y, Chadaideh KS, Li Y, et al. Butter and plant-based oils intake and mortality. JAMA Intern Med. 2025;185(5):549-560. doi:10.1001/jamainternmed.2025.0205
77. Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2020;8:CD011737. doi:10.1002/14651858.CD011737.pub3
78. Anitschkow N, Chalatow S. On experimental cholesterin steatosis and its significance in the origin of some pathological processes (1913). English reprint: Arteriosclerosis. 1983;3:178-182.
79. Dayton S, Pearce ML, Hashimoto S, Dixon WJ, Tomiyasu U. A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation. 1969;40(suppl II):II-1-II-63.
80. Astrup A, Magkos F, Bier DM, et al. Saturated fats and health: a reassessment and proposal for food-based recommendations. J Am Coll Cardiol. 2020;76:844-857. doi:10.1016/j.jacc.2020.05.077
81. O’Keefe JH Jr, Cordain L, Harris WH, Moe RM, Vogel R. Optimal low-density lipoprotein is 50 to 70 mg/dl: lower is better and physiologically normal. J Am Coll Cardiol. 2004;43(11):2142-2146. doi:10.1016/j.jacc.2004.03.046
82. Wang Z, Klipfell E, Bennett BJ, et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature. 2011;472(7341):57-63. doi:10.1038/nature09922
83. Koeth RA, Wang Z, Levison BS, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med. 2013;19(5):576-585. doi:10.1038/nm.3145
84. Heianza Y, Ma W, Manson JE, Rexrode KM, Qi L. Gut microbiota metabolites and risk of major adverse cardiovascular disease events and death: a systematic review and meta-analysis of prospective studies. J Am Heart Assoc. 2017;6(7):e004947. doi:10.1161/JAHA.116.004947
85. Schiattarella GG, Sannino A, Toscano E, et al. Gut microbe-generated metabolite trimethylamine-N-oxide as cardiovascular risk biomarker: a systematic review and dose-response meta-analysis. Eur Heart J. 2017;38(39):2948-2956. doi:10.1093/eurheartj/ehx342
86. Cho CE, Taesuwan S, Malysheva OV, et al. Trimethylamine-N-oxide (TMAO) response to animal source foods varies among healthy young men and is influenced by their gut microbiota composition: a randomized controlled trial. Mol Nutr Food Res. 2017;61(1):1600324. doi:10.1002/mnfr.201600324
87. Whitehead A, Beck EJ, Tosh S, Wolever TMS. Cholesterol-lowering effects of oat beta-glucan: a meta-analysis of randomized controlled trials. Am J Clin Nutr. 2014;100(6):1413-1421. doi:10.3945/ajcn.114.086108
88. Blanco Mejia S, Messina M, Li SS, et al. A meta-analysis of 46 studies identified by the FDA demonstrates that soy protein decreases circulating LDL and total cholesterol concentrations in adults. J Nutr. 2019;149(6):968-981. doi:10.1093/jn/nxz020
89. Hadi A, Askarpour M, Ziaei R, et al. Effect of flaxseed supplementation on lipid profile: an updated systematic review and dose-response meta-analysis of sixty-two randomized controlled trials. Pharmacol Res. 2020;152:104622. doi:10.1016/j.phrs.2019.104622
90. Schoeneck M, Iggman D. The effects of foods on LDL cholesterol levels: a systematic review of the accumulated evidence from systematic reviews and meta-analyses of randomized controlled trials. Nutr Metab Cardiovasc Dis. 2021;31(5):1325-1338. doi:10.1016/j.numecd.2020.12.032
91. Montorsi P, Ravagnani PM, Galli S, et al. The artery size hypothesis: a macrovascular link between erectile dysfunction and coronary artery disease. Am J Cardiol. 2005;96(12B):19M-23M. doi:10.1016/j.amjcard.2005.07.006
92. Dietary Guidelines Advisory Committee. Scientific Report of the 2015 Dietary Guidelines Advisory Committee. USDA and HHS; February 2015. Part D, Chapter 1. dietaryguidelines.gov
93. US Department of Health and Human Services, US Department of Agriculture. 2015-2020 Dietary Guidelines for Americans. 8th ed. Released January 2016. Chapter 1. dietaryguidelines.gov
94. US Department of Agriculture, US Department of Health and Human Services. Dietary Guidelines for Americans, 2020-2025. 9th ed. December 2020. dietaryguidelines.gov
95. Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk. Circulation. 2014;129(25 Suppl 2):S76-S99. doi:10.1161/01.cir.0000437740.48606.d1
96. Vincent MJ, Allen B, Palacios OM, Haber LT, Maki KC. Meta-regression analysis of the effects of dietary cholesterol intake on LDL and HDL cholesterol. Am J Clin Nutr. 2019;109(1):7-16. doi:10.1093/ajcn/nqy273
97. American College of Cardiology. How do the 2025-2030 Dietary Guidelines for Americans measure up for cardiovascular health? Expert Analysis. January 27, 2026.
98. McNamara DJ. The fifty year rehabilitation of the egg. Nutrients. 2015;7(10):8716-8722. doi:10.3390/nu7105429
99. Dietary Guidelines Advisory Committee, Nutrition Evidence Systematic Review. Food Sources of Saturated Fat and Risk of Cardiovascular Disease: A Systematic Review. USDA; 2024. doi:10.52570/NESR.DGAC2025.SR24
100. Dietary Guidelines Advisory Committee, Nutrition Evidence Systematic Review. Dietary Patterns and Risk of Cardiovascular Disease: A Systematic Review. USDA; 2024. doi:10.52570/NESR.DGAC2025.SR13
101. Dietary Guidelines Advisory Committee, Nutrition Evidence Systematic Review. Types of Dietary Fat and Cardiovascular Disease: A Systematic Review. USDA; 2020. doi:10.52570/NESR.DGAC2020.SR0501
102. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and unhealthful plant-based diets and the risk of coronary heart disease in U.S. adults. J Am Coll Cardiol. 2017;70(4):411-422. doi:10.1016/j.jacc.2017.05.047
103. Kim H, Caulfield LE, Garcia-Larsen V, Steffen LM, Coresh J, Rebholz CM. Plant-based diets are associated with a lower risk of incident cardiovascular disease, cardiovascular disease mortality, and all-cause mortality in a general population of middle-aged adults. J Am Heart Assoc. 2019;8(16):e012865. doi:10.1161/JAHA.119.012865
104. Shan Z, Wang F, Li Y, et al. Healthy eating patterns and risk of total and cause-specific mortality. JAMA Intern Med. 2023;183(2):142-153. doi:10.1001/jamainternmed.2022.6117
105. Glenn AJ, Lo K, Jenkins DJA, et al. Relationship between a plant-based dietary portfolio and risk of cardiovascular disease: findings from the Women’s Health Initiative prospective cohort study. J Am Heart Assoc. 2021;10(16):e021515. doi:10.1161/JAHA.120.021515
106. Brown MJ, Ferruzzi MG, Nguyen ML, Cooper DA, Eldridge AL, Schwartz SJ, White WS. Carotenoid bioavailability is higher from salads ingested with full-fat than with fat-reduced salad dressings as measured with electrochemical detection. Am J Clin Nutr. 2004;80(2):396-403. doi:10.1093/ajcn/80.2.396
107. Unlu NZ, Bohn T, Clinton SK, Schwartz SJ. Carotenoid absorption from salad and salsa by humans is enhanced by the addition of avocado or avocado oil. J Nutr. 2005;135(3):431-436. doi:10.1093/jn/135.3.431
108. White WS, Zhou Y, Crane A, Dixon P, Quadt F, Flendrig LM. Modeling the dose effects of soybean oil in salad dressing on carotenoid and fat-soluble vitamin bioavailability in salad vegetables. Am J Clin Nutr. 2017;106(4):1041-1051. doi:10.3945/ajcn.117.153635
109. Kopec RE, Cooperstone JL, Schweiggert RM, et al. Avocado consumption enhances human postprandial provitamin A absorption and conversion from a novel high-beta-carotene tomato sauce and from carrots. J Nutr. 2014;144(8):1158-1166. doi:10.3945/jn.113.187674
110. Gartner C, Stahl W, Sies H. Lycopene is more bioavailable from tomato paste than from fresh tomatoes. Am J Clin Nutr. 1997;66(1):116-122. doi:10.1093/ajcn/66.1.116
111. Goltz SR, Campbell WW, Chitchumroonchokchai C, Failla ML, Ferruzzi MG. Meal triacylglycerol profile modulates postprandial absorption of carotenoids in humans. Mol Nutr Food Res. 2012;56(6):866-877. doi:10.1002/mnfr.201100687
112. Reboul E. Mechanisms of carotenoid intestinal absorption: where do we stand? Nutrients. 2019;11(4):838. doi:10.3390/nu11040838
113. Goulinet S, Chapman MJ. Plasma LDL and HDL subspecies are heterogeneous in particle content of tocopherols and oxygenated and hydrocarbon carotenoids. Arterioscler Thromb Vasc Biol. 1997;17(4):786-796. doi:10.1161/01.ATV.17.4.786
114. Upritchard JE, Sutherland WH, Mann JI. Effect of supplementation with tomato juice, vitamin E, and vitamin C on LDL oxidation and products of inflammatory activity in type 2 diabetes. Diabetes Care. 2000;23(6):733-738. doi:10.2337/diacare.23.6.733
115. Theobald HE, Chowienczyk PJ, Whittall R, Humphries SE, Sanders TAB. LDL cholesterol-raising effect of low-dose docosahexaenoic acid in middle-aged men and women. Am J Clin Nutr. 2004;79(4):558-563. doi:10.1093/ajcn/79.4.558
116. Bernstein AM, Ding EL, Willett WC, Rimm EB. A meta-analysis shows that docosahexaenoic acid from algal oil reduces serum triglycerides and increases HDL-cholesterol and LDL-cholesterol in persons without coronary heart disease. J Nutr. 2012;142(1):99-104. doi:10.3945/jn.111.148973
117. Buettner GR. The pecking order of free radicals and antioxidants: lipid peroxidation, alpha-tocopherol, and ascorbate. Arch Biochem Biophys. 1993;300(2):535-543. doi:10.1006/abbi.1993.1074
118. Mohr D, Bowry VW, Stocker R. Dietary supplementation with coenzyme Q10 results in increased levels of ubiquinol-10 within circulating lipoproteins and increased resistance of human low-density lipoprotein to the initiation of lipid peroxidation. Biochim Biophys Acta. 1992;1126(3):247-254. PubMed 1637852
119. Brand MD. Mitochondrial generation of superoxide and hydrogen peroxide as the source of mitochondrial redox signaling. Free Radic Biol Med. 2016;100:14-31. PubMed 27085844
120. Tabas I, Williams KJ, Boren J. Subendothelial lipoprotein retention as the initiating process in atherosclerosis: update and therapeutic implications. Circulation. 2007;116(16):1832-1844. doi:10.1161/CIRCULATIONAHA.106.676890
121. Hevonoja T, Pentikainen MO, Hyvonen MT, Kovanen PT, Ala-Korpela M. Structure of low density lipoprotein (LDL) particles: basis for understanding molecular changes in modified LDL. Biochim Biophys Acta. 2000;1488(3):189-210.
122. Linus Pauling Institute, Oregon State University. Vitamin E. Micronutrient Information Center. lpi.oregonstate.edu doi:10.1161/JAHA.120.021515

Read more...

The easiest way to cook lentils is as follows: 1. Rinse the Lentils: Place the desired amount of lentils in a sieve and rinse them under cold water to remove any debris or dust. 2. Boil Water: In a pot, use about 3 cups of water for every 1 cup of lentils. Bring the water to a boil. 3. Add Lentils: Once the water is boiling, add the rinsed lentils to the pot. 4. Simmer: Reduce the heat to a simmer. Cook uncovered, stirring occasionally. The cooking time will depend on the type of lentils:

• Green or Brown Lentils: 20-30 minutes
• Red or Yellow Lentils: 15-20 minutes
• Black Beluga Lentils: 25-30 minutes
• French Green Lentils (Lentilles du Puy): 25-30 minutes

5. Check for Doneness: Lentils are done when they are tender but not mushy. You can taste a few to check. 6. Drain and Season: If there is any excess water, drain the lentils. Season with salt and any other desired spices or herbs. That’s it! Cooked lentils can be used in a variety of dishes, including soups, salads, stews, and as a protein-rich side dish.

Fun Facts

Lentils are a highly nutritious and healthy food, making them an excellent choice for a balanced diet. Here are some key points about their nutritional value and health benefits:

Nutritional Profile

1. Protein: Lentils are rich in protein, providing about 18 grams per cooked cup (198 grams). They are an excellent plant-based protein source, making them a staple in vegetarian and vegan diets. 2. Fiber: Lentils are high in dietary fiber, offering about 15.6 grams per cooked cup. This fiber includes both soluble and insoluble types, which are beneficial for digestive health and maintaining regular bowel movements. 3. Vitamins and Minerals: Folate (Vitamin B9): Lentils are particularly high in folate, with one cup providing about 90% of the daily recommended intake. Folate is crucial for DNA synthesis and repair, and it is especially important during pregnancy for fetal development. Iron: They provide a good amount of iron, which is essential for the formation of hemoglobin and preventing anemia. One cooked cup contains about 6.6 mg of iron. Magnesium: Lentils contain magnesium, which is important for muscle and nerve function, blood glucose control, and bone health. Potassium: They offer a substantial amount of potassium, which helps regulate fluid balance, muscle contractions, and nerve signals. Zinc: Lentils provide zinc, important for immune function, protein synthesis, and DNA synthesis. 4. Low in Fat: Lentils are low in fat, making them a heart-healthy food choice. They contain virtually no saturated fat.

Health Benefits

1. Heart Health: The high fiber, folate, and potassium content in lentils contribute to heart health. Fiber helps reduce blood cholesterol levels, which can lower the risk of heart disease. Potassium helps manage blood pressure, and folate helps prevent homocysteine buildup, reducing cardiovascular risk. 2. Digestive Health: The high fiber content promotes healthy digestion and prevents constipation. Fiber also acts as a prebiotic, feeding the beneficial bacteria in the gut. 3. Blood Sugar Control: Lentils have a low glycemic index, meaning they release energy slowly and help stabilize blood sugar levels. This makes them a good choice for people with diabetes or those managing their blood sugar. 4. Weight Management: High in protein and fiber, lentils can help promote a feeling of fullness and reduce overall calorie intake, which can be beneficial for weight management. 5. Nutrient Density: Lentils provide a wealth of nutrients without being calorie-dense, making them a nutritious option for overall health and wellness. 6. Antioxidant Properties: Lentils contain various bioactive compounds, including polyphenols, which have antioxidant properties. These compounds help combat oxidative stress and inflammation in the body. Incorporating lentils into your diet can offer numerous health benefits due to their rich nutritional profile. They are versatile and can be used in soups, stews, salads, and many other dishes, making it easy to enjoy their health benefits. Read more...

If you have ever heard someone claim that “studies show vegans live longer,” or, from the other camp, that “studies show pescetarians outlive vegans,” there is a good chance both claims trace back to the same source: the Adventist Health Studies out of Loma Linda University. These are some of the most cited datasets in all of nutrition science, and for good reason. They are also some of the most misquoted.

Loma Linda is the one Blue Zone built on a formal prospective cohort rather than on age-record demography, which is why it keeps surfacing whenever the Blue Zones are attacked or defended. If that is the fight that brought you here, two companion pieces handle it directly: Saul Newman vs. the Blue Zones takes the demographic critique apart claim by claim, and Blue Zones Debunked or Not answers the ten pillars of the viral anti-Blue-Zones narrative, including the Loma Linda Fallacy specifically. This post does the narrower job neither of those stops for: what the Adventist mortality data itself can and cannot establish.

Why Adventists Get Studied So Much

Seventh-day Adventists are a researcher’s dream population. Their church discourages smoking and alcohol and encourages a plant-forward diet, but individual members vary widely. Some are lifelong vegetarians, some are vegan, some eat fish, and roughly half eat meat. That variation inside a community sharing similar values, social structures, and health behaviors makes it possible to compare diets while holding many confounders more constant than in the general population.

How many Adventist Health Studies are there?

Three major waves. The Adventist Mortality Study (1960 to 1965) followed about 23,000 California Adventists and first showed they dramatically outlived the general population. Adventist Health Study 1 (1974 to 1988) followed about 34,000 California Adventists and linked vegetarian diets, nut consumption, and other habits to lower heart disease risk. Adventist Health Study 2 (AHS-2) enrolled roughly 96,000 people across the United States and Canada starting in 2002 and sorted them into five groups: vegan, lacto-ovo vegetarian, pesco-vegetarian, semi-vegetarian, and non-vegetarian. AHS-2 is where nearly all the vegan-specific findings come from.

Do Adventists really live 10 years longer?

Fraser and Shavlik’s 2001 analysis of the AHS-1 cohort found California Adventists had higher life expectancy at age 30 than other white Californians by 7.28 years in men (95% CI 6.59 to 7.97) and 4.42 years in women (95% CI 3.96 to 4.88). Adventist vegetarian men lived about 9.5 years longer than California men generally, and vegetarian women about 6.1 years longer.

Read that 9.5-year figure carefully, because it gets abused in both directions. It reflects the combined vegetarian-Adventist lifestyle package, meaning diet plus non-smoking, normal body weight, exercise, and nut consumption, measured against the general public. It is not the effect of a vegetarian diet in isolation, which within the cohort is considerably smaller. Still, the dose-response signature is what matters: same religion, same town, more of the behavior, more of the benefit. That is precisely why Loma Linda holds up as the cleanest of the Blue Zones even when the others are contested. AHS-1, however, had too few vegans to analyze them separately.

How Vegans Fared on Disease Outcomes

Here is the part that gets lost in the lifespan debate. Across most disease endpoints in AHS-2, vegans did remarkably well, frequently better than every other dietary group.

Do vegans weigh less?

There was a clean stepwise gradient in body weight across diet groups. In Tonstad’s 2009 analysis, average BMI ran 23.6 in vegans, 25.7 in lacto-ovo vegetarians, 26.3 in pesco-vegetarians, and 28.8 in non-vegetarians. Vegans sat in normal-weight territory while non-vegetarian Adventists averaged in the overweight range. No other dietary pattern in the study matched the vegans on this. (If body weight is your main interest, the mechanism gets its own treatment in the healthiest way to lose weight.)

Do vegans get less diabetes?

Vegans had both the lowest prevalence and the lowest incidence of diabetes of any group. Prevalence ran 2.9 percent in vegans against 7.6 percent in non-vegetarians. In the incidence analysis, the significant vegetarian groups showed roughly 38 to 62 percent lower odds of developing diabetes than non-vegetarians: vegans lowest at an odds ratio of 0.38, then semi-vegetarians at 0.49 and lacto-ovo vegetarians at 0.62. The pesco estimate (0.79) was not statistically significant. Those models already adjusted for BMI, so the reduction held up even after accounting for body weight, which suggests the protection was not purely about being thinner.

Do vegans have lower blood pressure?

Yes, and the lowest prevalence of hypertension. In white participants, vegans had about 0.37 times the odds of hypertension versus non-vegetarians, attenuating toward 0.53 after adjusting for BMI, so part of the effect ran through lower body weight. In black participants, vegans and lacto-ovo vegetarians combined had about 0.56 times the odds. Vegetarian patterns overall were associated with roughly half the odds of metabolic syndrome (0.44). Vegans and other vegetarians also had lower total cholesterol and lower LDL; for why that pathway matters so much, see cholesterol from the ground up.

Do vegans get less cancer?

In the 2013 cancer incidence analysis, vegetarians as a whole had modestly lower overall cancer risk than non-vegetarians, and vegans showed the largest reduction of any group (hazard ratio about 0.84), including a notable reduction in female-specific cancers (about 0.66). The authors speculated the latter might relate to higher soy intake. A 2025 follow-up on site-specific cancers reported lower risk across vegan, lacto-ovo, and pesco-vegetarian groups, with notable protection for breast and prostate cancer in younger vegans.

Colorectal cancer is the interesting exception. When a 2015 AHS-2 paper looked specifically at colorectal cancers, the pesco-vegetarians posted the strongest and only individually significant reduction (0.57). Vegans came in at 0.84, not statistically significant, with a wide interval overlapping the pesco estimate, so the two were not statistically distinguishable. (For the related and much-abused claim that a recent pooled analysis showed vegans with higher colorectal risk, see why a 40 percent number in one subgroup did not overturn the meat and bowel cancer link.)

What about dairy specifically?

Worth noting because it is the same cohort answering a question the diet-group comparison cannot. Fraser’s 2020 analysis of dairy, soy, and breast cancer in AHS-2 found dairy milk associated with roughly 50 percent higher breast cancer risk comparing high to low intake (hazard ratio 1.50, 95% CI 1.22 to 1.84), while soy milk showed no such association. That is a within-Adventist comparison, which sidesteps a great deal of the confounding that plagues comparisons against the general public.

So on the disease scoreboard, covering weight, diabetes, blood pressure, metabolic syndrome, overall cancer incidence, and cholesterol, vegans in AHS-2 were consistently at or near the top. On the handful of endpoints where another group edged ahead, the differences between the plant-forward groups sat within the statistical noise. Which makes the mortality finding all the more interesting.

The Lifespan Finding, and Why It Is Shakier Than You Have Heard

The 2013 mortality paper followed 73,308 AHS-2 participants for a mean of 5.79 years and reported adjusted all-cause mortality hazard ratios versus non-vegetarians: 0.85 for vegans (95% CI 0.73 to 1.01), 0.91 for lacto-ovo vegetarians, 0.81 for pesco-vegetarians (95% CI 0.69 to 0.94), and 0.92 for semi-vegetarians.

Do pescatarians live longer than vegans?

This is the origin of that claim, and here is everything wrong with reading it that way.

The difference is not statistically significant. The vegan and pesco-vegetarian confidence intervals (0.73 to 1.01 and 0.69 to 0.94) overlap almost entirely. Each group’s point estimate sits comfortably inside the other group’s interval. The study was never designed to compare vegans against pescetarians head to head; both were compared against non-vegetarians. The pesco estimate cleared the significance threshold and the vegan estimate narrowly missed it, which is a statement about statistical precision in a smaller group, not evidence that fish eaters outlive vegans. Treating a four-percentage-point gap between point estimates with overlapping intervals as a ranking is exactly the overinterpretation confidence intervals exist to prevent.

Did the study separate whole-food vegans from junk-food vegans?

No. Diet was assessed once, at baseline, by food frequency questionnaire, and classification rested purely on how often people ate animal products. A vegan subsisting on refined carbohydrates and processed meat analogues was categorized identically to a vegan eating legumes, vegetables, and whole grains. This is not hypothetical in this particular population: the Adventist community essentially invented the American meat-analogue industry, and Loma Linda brand vegan hot dogs and canned mock meats were staples for generations of Adventists.

Why does that matter as a confounder? Because “vegan” spans everything from a highly processed pattern built on refined flour, mock meats, oils, and sugar to a whole-food pattern built on legumes, leafy greens, whole grains, fruits, vegetables, mushrooms, nuts, seeds, seaweeds, herbs, and spices. Those two diets differ radically in fiber, potassium, antioxidants, phytonutrients, sodium, and refined carbohydrate load, which are the very things thought to drive the benefits of plant-based eating. Later research using plant-based diet indices confirmed the intuition: healthful plant-based patterns are associated with lower coronary heart disease risk while unhealthful plant-based patterns are associated with higher risk. If even a modest fraction of AHS-2’s vegans were mock-meat-and-white-bread vegans, their outcomes would drag down the group average and mask the results of the whole-food vegans, and the study had no way to separate the two.

Did the study measure B12?

Not in the mortality models. Those controlled for age, race, sex, smoking, exercise, education, personal income, marital status, alcohol, geographic region, menopause, and hormone therapy. No serum B12, no supplement-use adjustment.

This matters because B12 is the one nutrient a vegan diet cannot reliably supply. It is made by bacteria, not by plants or by animals themselves, and it concentrates in animal tissue. An unsupplemented vegan is therefore on a slow countdown as liver stores run down, which the Merck Manual puts at roughly three to five years. What goes wrong when they do? Quite a lot. B12 deficiency causes megaloblastic anemia. It damages the myelin sheathing of nerves, producing peripheral neuropathy, balance problems, and in advanced cases irreversible spinal cord degeneration. It is linked to cognitive decline and depression. And it drives up homocysteine, an amino acid associated with increased risk of stroke, cardiovascular disease, and possibly dementia. Elevated homocysteine is a particularly cruel confounder here, because it could quietly offset some of the cardiovascular benefit vegans would otherwise gain from low cholesterol and blood pressure.

To be fair, AHS-2 was not entirely blind on this. Smaller calibration substudies did measure B12 intake and blood biomarkers, and the picture they found was bimodal. Median B12 intake was actually highest in vegans, around 9.4 micrograms per day, because those who supplement lean heavily on fortified foods and pills. Yet about 15 percent of vegans still fell below the estimated requirement, against roughly 11 percent of lacto-ovo vegetarians and 6 percent of non-vegetarians. So the concern is not that all vegans were deficient. It is that a meaningful tail was, and the main mortality analysis could not adjust for where any individual fell.

Where does B12 in meat actually come from?

Worth pausing on, because the supply chain is not what most people assume. Ruminants get cobalt supplements so their gut bacteria can synthesize B12, and monogastric livestock get B12 directly in feed premixes. According to Ullmann’s Encyclopedia of Industrial Chemistry, the feed sector accounts for roughly 55 percent of global B12 sales, with food and pharmaceutical uses making up the other 45 percent. The supplement is in the supply chain either way. Vegans just need to take it directly. That full argument, including why the popular “90 percent” version of this statistic is wrong, lives in the great B12 gamble.

How long had they been vegan?

Nobody controlled for it, and this is the underappreciated problem. It is tempting to assume Adventist vegans were lifelong adherents. The data say otherwise. A 2017 AHS-2 analysis of lifetime dietary patterns found only 29 percent of elderly participants kept the same dietary pattern across their lifetime, and among those stable participants, lifetime vegans were just 1.1 percent against 31.2 percent for lifetime lacto-ovo vegetarians. The dominant pattern was conversion: about 60 percent of diet-switchers moved toward more plant-based eating as they aged. Lifelong lacto-ovo vegetarianism is genuinely common among Adventists. Lifelong veganism is rare.

That matters twice over. A recent convert spent most of their life eating differently, so a six-year mortality follow-up largely measures the residue of a previous diet: decades of one eating pattern, credited to a few years of another. And the direction of conversion creates a reverse-causation problem, a kind of unhealthy-convert bias analogous to the sick-quitter effect in smoking research, where recent quitters can look worse than continuing smokers because people often quit precisely when they get a frightening diagnosis. Some older adults go vegan because something is already wrong: a heart diagnosis, prediabetes, a cancer scare, a doctor’s ultimatum. Their subsequent illness gets tallied in the vegan column even though the disease predates the diet.

In fairness, this cuts less cleanly than it might. The study excluded people with prior cardiovascular disease and cancer at baseline, and the same 2017 analysis found that switchers moving away from plant-based eating tended to be the less healthy ones. The honest position is that single-baseline diet measurement leaves the direction genuinely uncertain, not that reverse causation is proven.

Was the vegan category reliably measured?

It was the noisiest of the lot. AHS-2’s own methodological work on dietary recall found classification was more reliable for lacto-ovo vegetarian and non-vegetarian patterns than for vegan, semi-, and pesco-vegetarian ones. Small group (7.6 percent of the cohort), unstable membership, fuzzy edges. Worth understanding which direction that pushes: non-differential measurement error blurs groups into each other, which makes real differences look smaller than they are. Poor classification means the observed estimate probably sits closer to 1.0 than the truth does. It widens the error bars rather than shrinking the effect.

Finally, the follow-up was short. A mean of 5.79 years is a brief window for diet-driven differences in chronic disease mortality to express themselves, especially in a category dominated by recent converts.

Does confounding only run one way?

No, and honesty requires saying so. Everything above would tend to make vegans look worse than they are. At least one significant bias runs the other way. Even within a cohort sharing the Adventist baseline of no smoking, no alcohol, and strong community, people who adopt the most restrictive diet plausibly self-select on conscientiousness, health interest, and a range of behaviors the study never measured: sleep, stress, healthcare use, supplement habits, general diligence. That is a healthy-user effect operating inside the cohort, and it would inflate the vegan results.

So the confounders do not stack neatly in one direction. Some push the vegan estimate down, at least one pushes it up, and nobody can say which dominates. Which is precisely why the honest verdict is that these groups cannot be ranked against each other. Not that vegans would win a fair fight, but that the fight was not measured well enough to call.

The Category Problem: “Vegan” Is Not a Diet

There is a deeper issue running underneath every limitation above, and it is conceptual rather than statistical.

Vegetarianism, as measured in these studies, is a dietary classification. Veganism is not. It is an ethical position, a commitment to avoid exploiting animals as far as is practicable, and the eating pattern is a downstream consequence rather than the thing itself. That distinction has real analytical teeth, because a category defined by what people refuse tells you almost nothing about what they actually eat.

What does “vegan” actually tell you about someone’s diet?

Very little on its own. Compare the two people this label covers. One eats lentils, leafy greens, whole grains, mushrooms, nuts, seeds, seaweeds, herbs, and spices, tracks B12, and has built the diet deliberately. The other eats refined bread, fries, sugar, and mock meats, and has never thought about a nutrient. Both are equally vegan by the ethical definition, because both avoid animal products. Nutritionally they have almost nothing in common.

Meanwhile a pescatarian gets marine omega-3s built into the pattern by default, and any diet including eggs, dairy, or fish supplies more B12 than a vegan diet does without anyone having to think about it. Even lacto-ovo vegetarians are not fully covered, running an intermediate rate of inadequate B12 intake between vegans and omnivores. The vegan category is simply the extreme case: the one where nutrient adequacy depends most heavily on individual knowledge and effort rather than being supplied by the food itself. (What “done properly” looks like in practice is the subject of the nutritional case for plants over animals.)

Are vegans healthier just because they are health-obsessed?

This is the standard objection, and the motivation research suggests the sorting may run opposite to the assumption. Rosenfeld’s work describes a consistent gradient: the less restrictive a form of animal-product avoidance, the more likely it is driven by health rather than ethics. Pescatarians are predicted to be largely health-motivated, consistent with dietary guidelines recommending regular fish intake. Vegans, by contrast, report stronger ethical motivations concerning animals and the environment than health motives.

Follow that through and it inverts the objection. If health-motivated people preferentially land at pescatarianism while ethically motivated people land at veganism, the health-optimizer bias favors the pesco group. Someone who restricts less is more often doing it for their health. Someone who went vegan over factory farming may never have considered their B12 status at all, and has no dietary source of it.

Two caveats keep this honest. The motivation research comes from general populations, not from Adventists, whose members sit inside a church promoting plant-based eating on both religious and health grounds, so the sorting is plausible in AHS-2 but not demonstrated there. And the heterogeneity argument cuts both ways: if the vegan category is too internally varied to trust when results look unfavorable, it is equally varied when results look good.

What the 2024 Update Added

Most discussion of AHS-2 mortality still cites the 2013 paper, but the group published a substantially longer follow-up in 2024, extending mortality tracking through 2015 across roughly 88,400 participants and 12,515 deaths. It is the more current answer, and it complicates both sides of the argument.

The headline result got stronger. Vegetarians overall showed a hazard ratio of 0.89 (95% CI 0.83 to 0.95) for all-cause mortality at age 65, a tighter and more clearly significant estimate than 2013 produced. Cause-specific reductions were substantial: renal failure 0.52, infectious disease 0.57, diabetes 0.51, ischemic heart disease 0.73, and select cardiac causes 0.75. Notice where those land. They are precisely the endpoints the disease markers predict, which is the healthspan-to-lifespan link showing up in the data rather than being assumed.

Do vegans specifically live longer?

Here the honest answer is messier, and it belongs in the post rather than tucked away. In the minimally adjusted model, vegans died at about 0.88 times the rate of their meat-eating co-religionists. Once the model layered in BMI, exercise, and prevalent disease, the all-vegan advantage shrank toward statistical noise, with a hazard ratio drifting up around 0.97 to 0.99. The clearest surviving signal was in men: vegan men at 65 carried a hazard ratio near 0.72, but by 85 that edge had vanished, the point estimate sitting just above 1.0.

One methodological note worth flagging: adjusting for BMI is arguably over-adjustment, because lower body weight is one of the mechanisms by which a plant-based diet extends life. Controlling it away quietly controls away part of the diet’s own benefit.

Do vegetarians have higher stroke and dementia risk?

In the oldest participants, several neurological causes went the other way: stroke at 1.17 (1.02 to 1.33), dementia at 1.13 (1.00 to 1.27), and Parkinson’s disease at 1.37 (0.98 to 1.91), all estimated at age 85. Two of those three intervals touch or cross 1.0, so this is a signal worth watching rather than a settled finding. It is a real one, though, and it belongs in any honest accounting.

One important wrinkle: in this analysis “vegetarian” pools all the meatless groups including pesco-vegetarians, so these numbers are not vegan-specific. That actually complicates the tidy nutrient story, because pescatarians eat fish and carry omega-3 levels similar to omnivores, yet the elevated signal still appears. Whatever drives it is not purely an absence of marine omega-3s.

Even so, the explanations the authors themselves reached for were nutrient gaps, not plant foods. They wrote that the higher stroke, dementia, and Parkinson’s risk might relate to vitamin B12 deficiency and to low intake and blood levels of EPA and DHA. The omega-3 half fits awkwardly with the pescatarian wrinkle just noted, which is precisely why they flagged these deaths as meriting further study rather than declaring the mechanism solved. But the shape of the argument holds: if elevated neurological mortality in older plant-based eaters traces to nutrient shortfalls, B12 most cleanly, that points toward supplementation and monitoring, not away from plants.

What Happens Outside the Adventist Cohort

A fair critique deserves a fair hearing, and this is the strongest one available: if the plant-based advantage is real, it should show up somewhere other than Loma Linda. The picture outside is genuinely mixed, and pretending otherwise would repeat the selective reading this post criticizes.

Does the vegan advantage disappear outside Loma Linda?

Partly. EPIC-Oxford, the UK cohort of about 48,000 meat eaters, fish eaters, and vegetarians, found no significant all-cause mortality difference between vegetarians and comparable non-vegetarians. That is a real result and it should temper any confident claim that vegetarians simply outlive everyone.

The same cohort, however, reported 22 percent lower ischemic heart disease in vegetarians and vegans over 18 years of follow-up, alongside a higher rate of hemorrhagic stroke. Researchers suspect very low B12 status played a role, in a cohort that largely predates routine supplementation. It is a split result, and citing only one half of it would be dishonest in either direction.

Do vegans break bones more easily?

This is the finding least often mentioned by plant-based advocates, so it goes here rather than in a footnote. EPIC-Oxford found vegans had 2.31 times the risk of hip fracture compared with meat eaters (95% CI 1.66 to 3.22), equivalent to about 15 additional hip fractures per 1,000 people over ten years. That figure was already adjusted for BMI, and it attenuated only slightly after further adjustment for dietary calcium and protein.

It is not a B12 story. The mechanism points at low body weight combined with lower calcium, protein, and vitamin D intake. Which makes it the same category of finding as the neurological signal: a nutrient-adequacy problem with an identifiable remedy, not evidence that plants are harmful. Note the pattern across cohorts. The vegan-adverse findings cluster around nutrients that a plant-exclusive diet supplies poorly unless someone plans for them, rather than around anything plants do to you.

The Honest Summary

The Adventist Health Studies are strong evidence that plant-predominant diets are associated with better health, and within them, vegans posted the best results on most disease outcomes: lowest weight, lowest diabetes risk, lowest blood pressure, least metabolic syndrome, lowest cholesterol, and the largest reduction in overall cancer incidence. On the few endpoints where they did not come out on top, the gap between vegans and the leading group was statistically a tie. The 2024 follow-up strengthened the overall vegetarian mortality result while adding genuine complications in the very old, which the study’s own authors attributed to nutrient gaps rather than to plant foods.

What the studies do not show is that pescetarians outlive vegans. That claim rests on a statistically non-significant difference between two overlapping confidence intervals, in an analysis riddled with unmeasured variables: B12 status, diet quality in a population where mock meats were a cultural institution, and how recently anyone had converted in a group where lifetime adherents were vanishingly rare. Add a follow-up under six years and a single baseline diet snapshot in the category the study’s own methods papers flagged as least reliable, and the ranking simply is not there.

This is an argument about uncertainty, not a claim that vegans would have won a cleaner study. Some unmeasured variables would drag the vegan estimate down; healthy-user self-selection would push it up. The point is that the instrument was not precise enough to separate these groups, which is exactly why a four-percentage-point gap between point estimates should not be treated as a finding by anyone, in either direction.

The right conclusion is not “veganism failed” or “fish is the secret.” Within this cohort, all the plant-forward patterns clustered together at lower mortality than meat eating, and the data are too coarse to rank them against each other. The defensible claim is narrower and more useful than a ranking: a well-constructed whole-food plant-based diet, with deliberate attention to B12, calcium, iodine, and vitamin D, produced the best outcomes measured here, and probably better than measured, since the measured group included plenty of people not eating that way at all. What the evidence does not support, and what no serious advocate should claim, is that avoiding animal products by itself is sufficient. The ethics set the constraint. Nutrition is a separate discipline, and it still has to be done.

Related Reading

• What the Longest-Living People Actually Eat places this cohort inside the full hierarchy of diet-and-longevity evidence: the gradient, the randomized trials, the identical-twin study, and the genetics.
• Saul Newman vs. the Blue Zones takes the demographic critique apart claim by claim.
• Blue Zones Debunked or Not answers the ten pillars of the anti-Blue-Zones narrative, including the Loma Linda Fallacy.
• The Great B12 Gamble covers B12 biochemistry, deficiency markers, and where supplemental B12 actually goes.
• Cholesterol, From the Ground Up walks the evidence staircase on LDL and ApoB causality.
• No, a 40 Percent Number in One Subgroup Did Not Overturn the Meat and Bowel Cancer Link handles the recent colorectal cancer claim.

References

1. Orlich MJ, Singh PN, Sabaté J, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Internal Medicine. 2013;173(13):1230-1238. PubMed
2. Abris GP, Shavlik DJ, Mathew RO, et al. Cause-specific and all-cause mortalities in vegetarian compared with those in nonvegetarian participants from the Adventist Health Study-2 cohort. American Journal of Clinical Nutrition. 2024;120(4):907-917. PubMed
3. Fraser GE, Shavlik DJ. Ten years of life: Is it a matter of choice? Archives of Internal Medicine. 2001;161(13):1645-1652. PubMed
4. Tonstad S, Butler T, Yan R, Fraser GE. Type of vegetarian diet, body weight, and prevalence of type 2 diabetes. Diabetes Care. 2009;32(5):791-796. PubMed
5. Tonstad S, Stewart K, Oda K, et al. Vegetarian diets and incidence of diabetes in the Adventist Health Study-2. Nutrition, Metabolism and Cardiovascular Diseases. 2013;23(4):292-299. PubMed
6. Pettersen BJ, Anousheh R, Fan J, et al. Vegetarian diets and blood pressure among white subjects: results from the Adventist Health Study-2. Public Health Nutrition. 2012;15(10):1909-1916. PubMed
7. Fraser GE, Katuli S, Anousheh R, et al. Vegetarian diets and cardiovascular risk factors in black members of the Adventist Health Study-2. Public Health Nutrition. 2015;18(3):537-545. PubMed
8. Rizzo NS, Sabaté J, Jaceldo-Siegl K, Fraser GE. Vegetarian dietary patterns are associated with a lower risk of metabolic syndrome. Diabetes Care. 2011;34(5):1225-1227. PubMed
9. Tantamango-Bartley Y, Jaceldo-Siegl K, Fan J, Fraser G. Vegetarian diets and the incidence of cancer in a low-risk population. Cancer Epidemiology, Biomarkers and Prevention. 2013;22(2):286-294. PubMed
10. Orlich MJ, Singh PN, Sabaté J, et al. Vegetarian dietary patterns and the risk of colorectal cancers. JAMA Internal Medicine. 2015;175(5):767-776. PubMed
11. Fraser GE, Butler FM, Shavlik DJ, et al. Longitudinal associations between vegetarian dietary habits and site-specific cancers in the Adventist Health Study-2 North American cohort. American Journal of Clinical Nutrition. 2025;122(2):535-543. PubMed
12. Fraser GE, Jaceldo-Siegl K, Orlich M, et al. Dairy, soy, and risk of breast cancer: those confounded milks. International Journal of Epidemiology. 2020;49(5):1526-1537. PubMed
13. Martins MCT, Jaceldo-Siegl K, Orlich M, et al. A new approach to assess lifetime dietary patterns finds lower consumption of animal foods with aging in a longitudinal analysis of a health-oriented Adventist population. Nutrients. 2017;9(10):1118. Full text
14. Martins MCT, Jaceldo-Siegl K, Fan J, et al. Short- and long-term reliability of adult recall of vegetarian dietary patterns in the Adventist Health Study-2. Journal of Nutritional Science. 2015;4:e11. PubMed
15. Damayanti D, Jaceldo-Siegl K, Beeson WL, et al. Foods and supplements associated with vitamin B12 biomarkers among vegetarian and non-vegetarian participants of the Adventist Health Study-2 calibration study. Nutrients. 2018;10(6):722. PubMed
16. Haddad EH, Jaceldo-Siegl K, Oda K, Fraser GE. Associations of circulating methylmalonic acid and vitamin B-12 biomarkers are modified by vegan dietary pattern in adult and elderly participants of the Adventist Health Study 2 calibration study. Current Developments in Nutrition. 2020;4(2):nzaa008. PubMed
17. Tong TYN, Appleby PN, Bradbury KE, et al. Risks of ischaemic heart disease and stroke in meat eaters, fish eaters, and vegetarians over 18 years of follow-up: results from the prospective EPIC-Oxford study. BMJ. 2019;366:l4897. PubMed
18. Tong TYN, Appleby PN, Armstrong MEG, et al. Vegetarian and vegan diets and risks of total and site-specific fractures: results from the prospective EPIC-Oxford study. BMC Medicine. 2020;18:353. Full text
19. Appleby PN, Crowe FL, Bradbury KE, et al. Mortality in vegetarians and comparable nonvegetarians in the United Kingdom. American Journal of Clinical Nutrition. 2016;103(1):218-230. PubMed
20. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and unhealthful plant-based diets and the risk of coronary heart disease in U.S. adults. Journal of the American College of Cardiology. 2017;70(4):411-422. PubMed
21. Rosenfeld DL. The psychology of vegetarianism: recent advances and future directions. Appetite. 2018;131:125-138. PubMed
22. Rosenfeld DL, Tomiyama AJ. How proximal are pescatarians to vegetarians? An investigation of dietary identity, motivation, and attitudes toward animals. Journal of Health Psychology. 2021;26(5):713-727. PubMed
23. Hopwood CJ, Bleidorn W, Schwaba T, Chen S. Health, environmental, and animal rights motives for vegetarian eating. PLOS ONE. 2020;15(4):e0230609. Full text
24. Merck Manual Professional Edition. Vitamin B12 deficiency. Reference
25. Ullmann’s Encyclopedia of Industrial Chemistry. Vitamins, 6. B Vitamins. Wiley-VCH. Reference

Read more...

Short version: every essential nutrient a human needs is available from plants, plus a reliable vitamin B12 source and, for most people, a little sunshine or a vitamin D supplement. That much is settled by the world’s major dietetic bodies. The more interesting claim is the one this post defends in full: across the entire hierarchy of evidence, from mechanism through randomized trials to decades-long cohorts, whole-food plant-based patterns do not merely match animal-based ones. They beat them on cholesterol, on diabetes, on mortality, and on the antioxidant and fiber payload that animal foods cannot supply at all. Once you accept that animal products are not nutritionally necessary, everything they cost beyond your own plate becomes a cost we are choosing to impose.

This post walks the whole case. It covers macronutrients and micronutrients in enough detail to show that the bases are genuinely covered, then climbs the evidence hierarchy tier by tier, answers the twenty objections you will actually meet online, sets out what the major health organizations conclude, and ends with a practical plan and an honest accounting of where the evidence is softer than the headline.

The nine findings this post defends

1. Plant proteins are complete and sufficient. All nine essential amino acids appear in ordinary plant foods, and the protein-combining rule was retracted by the author who popularized it.
2. Carbohydrate quality matters, quantity does not. Fiber, absent from every animal food, is the marker that separates the healthy carbohydrates from the harmful ones.
3. Saturated fat raises LDL, and LDL causes atherosclerosis. This is confirmed at every evidence tier, by feeding trials, by genetics, and by three unrelated drug classes.
4. The micronutrient gaps are few and manageable. B12 must be supplemented or fortified, exactly as it is for the animals people eat. Vitamin D and iodine warrant a little attention. Omega-3 is a lower priority than people assume, and the rest follows from eating whole plants.
5. The evidence hierarchy points one way. Randomized trials, giant cohorts, and mechanism all converge, and consilience across independent methods is what makes the case.
6. Plants carry what animal foods lack. Plant foods average roughly 64 times more antioxidants than animal foods, the anti-inflammatory payload tied to lower cardiovascular, brain, immune, and cancer risk.
7. Beyond the eater, it is not close. Even the lowest-impact animal products typically exceed the highest-impact plant proteins on climate, land, and water.
8. The institutional verdict is lopsided. Pediatric, cardiology, oncology, diabetes, and dietetic bodies across four continents affirm well-planned plant-based eating. No animal-centered pattern carries anything close to that support.
9. Every stock objection has a short answer. Twenty of them are answered below, from “where do you get your protein” to “correlation isn’t causation.”

This is a companion piece to earlier deep dives on vitamin B12, saturated fat and seed oils, cholesterol, dietary fat, and plant “toxins” and anti-nutrients. It uses the same rule throughout: climb the whole hierarchy of evidence, steelman the other side, then follow the data.

The shortcut that makes the rest of this optional

Do you have to micromanage nutrients on a plant-based diet?

No, and this is worth saying before the detail begins. Eat a variety across these categories, most days: legumes, greens, grains, fruits, vegetables, mushrooms, nuts, seeds, seaweed, and herbs and spices. That is the entire strategy. Rotate through those and the nutrients largely take care of themselves; variety is the trick that lets you stop counting. It feels like a lot to keep track of for about a week, and then it becomes second nature. Everything below is the biochemistry of why that one habit works. Read it if you are curious, skim it if you already trust the pattern.

Protein, carbohydrate, and fat on a plant-based diet

Can you get enough protein without meat?

The oldest objection to plant eating is also the weakest. Every essential amino acid is synthesized by plants, and all nine appear in ordinary plant foods. What is true is a matter of dose and distribution, not possibility: some plant proteins run lower in specific amino acids, grains in lysine and legumes in methionine, and plant proteins carry slightly lower digestibility scores than animal proteins. The fix is almost insultingly easy: eat legumes and grains across the day and hit adequate total protein. The National Academies set the protein RDA at 0.8 g/kg body weight and an Acceptable Macronutrient Distribution Range of 10–35 percent of energy, which is a wide and easily met window.

Do you need to combine plant proteins at every meal?

No. The “incomplete protein, must combine at every meal” idea traces to a 1971 popularization that its own author, Frances Moore Lappé, retracted in the 1981 revision of the same book. Your liver keeps a standing pool of amino acids that makes meal-by-meal combining unnecessary.

Which plant proteins are highest quality?

Soy is a standout: soy protein isolate lands in the high-quality band on DIAAS, with studies reporting values from roughly 90 to 98 depending on the reference pattern, and it delivers ample leucine. Pea protein sits just below it. Seitan is dense in protein, and legumes and whole grains round it out. Protein quality matters most at the per-meal level for athletes and older adults, who should simply eat somewhat larger, varied protein portions, roughly 0.4–0.5 g/kg per meal for plant-focused lifters versus 0.3–0.4 for animal-focused. Modeling studies of completely plant-based diets scaled to the calorie needs of male bodybuilders and of professional American footballers found they supply enough protein and leucine to maximize hypertrophy and strength.

Is plant protein better than animal protein for longevity?

Real-world outcomes favor plants. In the Harvard Nurses’ Health Study and Health Professionals Follow-up Study, Song and colleagues followed 131,342 people and found substituting plant protein for animal protein associated with lower all-cause and cardiovascular mortality. The NIH-AARP analysis of more than 400,000 participants found replacing 3 percent of energy from animal protein with plant protein associated with roughly 10 percent lower all-cause mortality.

Do carbohydrates cause disease, or does carbohydrate quality?

The carbohydrate wars conflate two different things: quantity and quality. The evidence overwhelmingly indicts refined flour and sugar while exonerating, indeed celebrating, whole intact carbohydrates: beans, oats, sweet potatoes, whole fruit. The defining nutrient here is fiber, which is entirely absent from all animal foods. The National Academies set the carbohydrate RDA at 130 g/day, based explicitly on the average minimum amount of glucose used by the brain, an Acceptable Macronutrient Distribution Range of 45–65 percent of energy for carbohydrate against 10–35 percent for protein and 20–35 percent for fat, and total-fiber Adequate Intakes of 38 g/day for men and 25 g/day for women. The same framework set no Tolerable Upper Intake Level for saturated fat, trans fat, or cholesterol, because incremental intake above zero raises coronary risk. Each should be kept as low as possible while eating a nutritionally adequate diet.

How much fiber do you need, and does it actually extend life?

Fiber’s payoff is large. Reynolds and colleagues’ 2019 Lancet series, pooling 185 prospective studies across roughly 135 million person-years plus 58 clinical trials, found the highest fiber consumers had 15–30 percent lower all-cause and cardiovascular mortality, with risk reduction greatest at 25–29 g/day and dose-response curves hinting at still greater benefit above that. Notably, that same series found low glycemic-index diets offered only limited, inconsistent protection. It is fiber and whole grains, not glycemic index, that carry the signal. Aune’s 2016 BMJ dose-response meta-analysis of 45 studies found whole-grain intake associated with substantial reductions in coronary disease, cancer, and total mortality, with benefit accruing up to two or three servings a day.

Do high-carbohydrate diets cause diabetes?

The populations that ate the most carbohydrate historically had the least of the disease. Okinawans, the Tarahumara, and rural China all had very low rates of heart disease and diabetes on diets built overwhelmingly on starch. And whole fruit associates with lower type 2 diabetes risk even as fruit juice raises it. Same sugar, different matrix.

Does saturated fat really raise cholesterol?

Yes, and the causal chain is airtight at every tier, as covered in depth in the seed oils piece. Mensink and Katan’s controlled feeding trials show saturated fat raises LDL and polyunsaturated fat lowers it, with the WHO-commissioned review of 84 trials grading the evidence high-certainty. The 2020 Cochrane review of 15 randomized trials and 56,675 participants found reducing saturated fat cut cardiovascular events by 17 percent. The 2017 American Heart Association Presidential Advisory led by Frank Sacks concluded that replacing saturated fat with polyunsaturated vegetable oil cuts cardiovascular disease by roughly 30 percent, comparable to statin treatment. And the 2017 European Atherosclerosis Society consensus, drawing on more than 200 studies, over 2 million participants, and more than 20 million person-years, concluded that LDL meets every criterion for causality. Mendelian randomization and three unrelated drug classes seal it: it is the LDL.

Are seed oils bad for you?

The linoleic acid evidence points the opposite way. Marklund’s 2019 Circulation pooled analysis of 30 cohorts across 13 countries found higher linoleic acid biomarkers associated with about 22 percent lower cardiovascular-disease mortality, alongside modestly lower total cardiovascular disease and ischemic stroke. Crucially, arachidonic acid was not associated with higher risk, which undercuts the inflammation storyline entirely. Wu’s 2017 pooled analysis in Lancet Diabetes and Endocrinology, covering 20 cohorts, found higher linoleic acid biomarkers associated with substantially lower type 2 diabetes risk. Mozaffarian’s 2010 trial meta-analysis found replacing saturated with polyunsaturated fat cut coronary events by about 19 percent overall, roughly 10 percent per 5 percent of energy swapped. Cochrane’s dedicated omega-6 review found no evidence of harm.

Why is grass-fed beef still a poor source of omega-3?

This is the wrinkle specific to animal fat that the “grass-fed is different” defense never survives. Fat from cattle, sheep, and goats does not reach your plate as it left the pasture. It passes first through the rumen, a fermentation chamber whose microbes hydrogenate unsaturated fats, and that has three consequences worth knowing. The process is incomplete, so beef and dairy fat carry a few percent naturally occurring trans fats as trans-18:1 and CLA intermediates. Beef fat’s most abundant saturate is palmitic acid, the classic LDL-raiser, much of it synthesized by the animal itself regardless of what it ate. And those same microbes saturate omega-3s just as readily as omega-6s, which is precisely why even grass-fed beef remains a mediocre omega-3 source. If you want the test case, it is butter: microbially pre-processed ruminant fat that still reliably raises LDL against nearly any plant oil it is compared with. The chemistry is taken apart in the dietary fat deep dive and in coconut oil versus animal fat.

Where do vegans get omega-3s without fish?

The plant world supplies ALA abundantly through flax, chia, walnuts, and hemp, and a tablespoon of ground flax or chia daily covers the baseline recommendation. Grind the flax, because whole seeds pass through largely intact. Your body converts ALA onward to EPA and DHA, and while the conversion percentage is often quoted as tiny, that framing is misleading: a tracer percentage measures what fraction of one labeled dose ends up as DHA, not whether total synthesis meets what your tissues actually need. And here is the point almost everyone misses about fish: the EPA and DHA in fish originate in algae. Fish merely accumulate it up the food chain, so anyone who does want preformed EPA and DHA can go straight to the source and skip the mercury.

Do vegans need to take a DHA or EPA supplement?

Probably not, if you are a healthy adult eating a well-planned whole-food vegan diet. This deserves a straight answer because the topic attracts loud certainty in both directions, and neither end is supported. There is no good evidence that a healthy vegan needs to supplement DHA and EPA, and no good evidence that a small dose causes harm. The large long-chain omega-3 supplement trials failed to improve hard cardiovascular outcomes, and they failed for cognition too, including a 2026 trial that proved the DHA actually reached the brain and still found no benefit. Meanwhile plant omega-3 has its own favorable outcome data. Taking roughly 250 mg a day of algae-derived EPA and DHA as cheap insurance is a defensible choice with a very small downside; skipping it is also defensible. What matters far more, and should be handled first, is B12, vitamin D, iodine, sleep, exercise, blood pressure, and overall diet quality. If you are spending worry-energy on DHA while your B12 status is unverified, the priorities are backwards. The groups with the strongest case for supplementing anyway are pregnant and breastfeeding women, infants and young children, APOE4 carriers, people with a confirmed very low omega-3 index, those with FADS variants that impair conversion, and older adults with cognitive concerns. The full evidence review, including the trials, the risks, the measurement problem, and where the experts genuinely disagree, is in Do vegans need DHA and EPA?

One practical note that follows from the biochemistry: ALA and linoleic acid compete for the same enzymes, so while there is no reason to fear omega-6 as a class and the ratio panic is overblown, cooking everything in high-linoleic oils such as sunflower, corn, or safflower works against your conversion. Olive, avocado, or no added oil are better defaults. There is also no need to megadose ALA; moderate and consistent beats heroic.

Covering every micronutrient without animal products

A well-planned entirely plant-based diet is endorsed as healthful and nutritionally adequate by the world’s major dietetic bodies, and the full institutional roll call is below. Here is the honest, precise rundown of the nutrients that deserve attention. Remember the shortcut: eat the ten categories with variety and almost all of this happens on autopilot. This section is the why, not a to-do list.

Do vegans have to take a B12 supplement?

Yes, and it is the one non-negotiable. As argued at length in The Great B12 Gamble, B12 is not an animal nutrient; it is a microbial one, made by bacteria and archaea, not by plants or animals. Every animal on Earth gets it from microbes. Modern farmed animals get it from feed supplements, since fermentation-produced cyanocobalamin is a regulator-approved feed additive for all animal species and animal feed accounts for a large share of global B12 production, or, for ruminants, from supplemented cobalt so their rumen microbes can build it. Vegans simply take the direct route and skip the middleman. Practical dosing: 50 mcg daily or 2,000 mcg weekly of cyanocobalamin, chewed, or fortified foods spread across at least three servings a day. There is no reliable plant source. Spirulina and most algae contain inactive pseudo-B12. Deficiency is not a vegan disease either: per NHANES data, roughly 3.6 percent of all US adults are outright deficient and about 12.5 percent insufficient, and US policy already tells everyone over 50, omnivores included, to get B12 from fortified foods or supplements.

How do vegans get vitamin D and iodine?

Vitamin D comes from sunlight, fortified foods, and supplements. D2 is plant and fungal in origin; D3 raises status more effectively and is now available in vegan lichen-derived form. For iodine, iodized salt is the simplest source. Seaweed works but requires moderation. Kelp in particular can massively overshoot the requirement, so treat it as a garnish rather than a staple.

How do vegans get enough zinc and selenium?

Selenium is covered by a couple of Brazil nuts a day, though do not overdo them; they are extremely concentrated. Zinc comes from legumes, seeds, and whole grains. Phytate reduces absorption somewhat, and soaking, sprouting, fermenting, and leavening as in sourdough all mitigate it, with fermentation studies showing phytate reductions up to roughly 86 percent (Nsabimana and colleagues, Frontiers in Nutrition, 2024). The fuller treatment, including why phytate is arguably working for you as an iron-chelating antioxidant, is in Plant “Toxins” and “Anti-Nutrients”.

Is plant iron as good as heme iron from meat?

Plant iron’s regulated absorption is a feature, not a bug: your body takes up more when stores are low and less when they are full. Pair it with vitamin C to boost uptake, which raises non-heme iron absorption several-fold, and keep tea and coffee about an hour away from iron-rich meals. By contrast, heme iron’s unregulated absorption is exactly why it is associated with higher disease risk. Meta-analyses find higher heme-iron intake associated with roughly 16 percent greater type 2 diabetes risk per 1 mg/day and with increased colorectal cancer risk, while non-heme and total iron show no such consistent harm. Ferritin is worth a nuance: very high ferritin often reflects inflammation and iron overload, not health.

Can you get enough calcium without dairy?

Yes. Fortified plant milks, calcium-set tofu, and low-oxalate greens such as kale, bok choy, and collards are well absorbed; bioavailability from low-oxalate greens actually exceeds that from cow’s milk. Spinach is high-oxalate, so its calcium is poorly available; choose the low-oxalate greens for calcium. Oxalate is a real concern only for stone-formers and extreme mono-eaters.

What about vitamin A, vitamin K2, choline, and creatine?

Vitamin A comes from carotenoids in orange and dark-green vegetables, converted to retinol as needed. Choline comes from soy, quinoa, broccoli, legumes, and cruciferous vegetables; scare pieces overstate the concern for anyone eating varied whole plants. Creatine, carnitine, and taurine are non-essential; the body synthesizes them endogenously. Athletes may optionally supplement creatine, and interestingly vegans tend to show a larger performance response because they start with lower muscle stores.

Do vegans need to worry about vitamin K2?

No, and this one deserves a plain statement: there is no evidence that vegans have any difficulty with vitamin K status, and no vitamin-K2 deficiency syndrome has ever been documented in vegans. The requirement authorities set is for vitamin K itself, and they set it on K1 alone. The European Food Safety Authority, reviewing both forms in 2017, concluded the evidence on menaquinones was insufficient to set any dietary reference value and set an Adequate Intake for phylloquinone only, at 70 micrograms a day for adults. K1 is abundant in leafy greens, and the body handles the rest by converting K1 to the tissue form MK-4, with gut bacteria making additional menaquinones. Natto exists for anyone who wants a direct, concentrated K2 food. Treat K2 anxiety as what it mostly is: supplement marketing in search of a problem.

Do plants have more antioxidants than animal foods?

Dramatically more, and this is the category that gets almost no attention in the “can you get enough on plants” conversation, because the answer runs the other way. It is animal foods that come up short. When researchers assayed the total antioxidant content of more than 3,100 foods, beverages, and spices for a landmark 2010 analysis (Carlsen and colleagues, Nutrition Journal), plant foods averaged on the order of 64 times more antioxidants than animal foods. That headline figure is a mean, lifted by antioxidant-dense plants like herbs, spices, and berries, but the gap holds even for ordinary foods, and most animal foods sit near zero. This is not a trivia point. The same antioxidant and anti-inflammatory compounds packed into plants are the ones tied to lower rates of exactly the diseases that dominate modern mortality, with a whole-food plant-rich pattern associated with better cardiovascular health, healthier brain aging and lower dementia risk, better-regulated immune function, and lower rates of several cancers, while patterns heavy in processed and high-heat-cooked animal products score as pro-inflammatory. One practical consequence follows directly from the chemistry: many of these compounds are fat-soluble, so they need a little dietary fat in the same meal to be absorbed at all, riding out of the intestine packaged in chylomicrons before being handed off to LDL and HDL for delivery around the body. The full journey, how these compounds travel from your plate through bile and the lymphatic system into the very lipoprotein particles that build plaque, and the electrostatic reason those particles get trapped in artery walls, is traced in the next section.

There is a reason the gap runs this way, and it is not arbitrary. Plants synthesize these compounds as their own survival chemistry. A leaf running photosynthesis in full sun generates reactive oxygen continuously and cannot walk away from UV or drought, so it banks the results in its leaves, skins, and seeds. Animals synthesize essentially none of them. No animal builds a carotenoid from scratch, which is why the pink of salmon and the yellow of a yolk are borrowed pigments that vanish when the feed changes. What an animal does eat is largely degraded in the gut, conjugated and excreted, or deposited into liver, fat, and milk rather than into the muscle on your plate. Routing a calorie through an animal costs you most of the energy, as the land arithmetic shows, and most of the protective chemistry that came with it. Why that matters inside a blood vessel is covered in cholesterol and dietary fat.

Why can you eat until full on a plant-based diet and still lose weight?

Whole plant foods are diluted by fiber and water, so they carry fewer calories per gram. On a metabolic ward where every calorie was measured, people eating a minimally processed plant-based diet spontaneously ate roughly 689 fewer calories a day than the same people on an animal-based ketogenic diet, with no difference in reported hunger or meal enjoyment. You can eat until comfortably full and still run a deficit, which is the whole argument of the best diet to lose weight.

What does a day of plant-based eating actually look like?

Everything above collapses into one habit: eat a variety across legumes, greens, grains, fruits, vegetables, mushrooms, nuts, seeds, seaweed, and herbs and spices, then add a B12 supplement, get some sun or a vitamin D pill, and use iodized salt or a little seaweed. If you would like that turned into a concrete daily checklist, Dr. Greger’s Daily Dozen is an excellent evidence-based template that maps almost exactly onto these categories, and the Go Vegan page collects starter resources, meal ideas, and further reading. A useful concrete target, and the one landed on in Plant “Toxins” and “Anti-Nutrients”, is roughly 30 different plants a week. Variety is simultaneously your nutritional safety margin and your biggest lever on microbiome health.

The journey: how plant antioxidants reach your arteries

This is the part of the story that usually gets skipped, and it is the most interesting thing in nutrition science that almost nobody explains. We have said that plants carry vastly more antioxidants and that saturated fat raises LDL. But how does a carotenoid in a raw carrot end up embedded inside a lipoprotein particle in your bloodstream, and why does it matter once it gets there? The answer runs through bile, the lymphatic system, and a piece of basic electrostatics happening in your artery wall right now.

This section follows the plant compounds. The same journey told from the cholesterol side, with the particle counts, the drug trials, the genetics, and the full plaque pathology, is in Cholesterol, from the ground up, and the two are worth reading together.

How do plant antioxidants actually get into your bloodstream?

Start in the mouth. Carotenoids, tocopherols, and vitamin K sit locked inside plant cell walls, bound up in the food matrix, and chewing and cooking are what liberate them. Once released in the small intestine they hit a problem: they are fat-soluble molecules in a watery environment, and they cannot simply diffuse across the gut wall on their own.

This is where bile arrives. Bile salts, made in the liver from cholesterol and released from the gallbladder when fat shows up, act as biological detergents. They emulsify dietary fat into progressively smaller droplets and then assemble the products into mixed micelles, tiny spherical carriers with their water-hating tails pointed inward and their water-loving surfaces pointed out. Fat-soluble plant compounds hitch a ride in the oily core. This is the concrete reason a fat-free salad wastes much of its own nutrition: no fat means little bile response, few micelles, and the carotenoids pass straight through you. It is also why the three to five grams of whole-food fat mentioned earlier is not a nutritional nicety but a transport requirement. The effect is not subtle: in controlled crossover studies, salads eaten with fat-free dressing produced almost no measurable carotenoid absorption at all, while adding avocado multiplied beta-carotene absorption more than fifteenfold, with the whole fruit working as well as extracted oil. Cooking helps too, by breaking the cell walls that trap these compounds.

Micelles ferry their cargo to the surface of the enterocytes lining the intestine, which absorb the contents. Inside the enterocyte, the fats are reassembled and packaged with cholesterol, phospholipids, fat-soluble vitamins, carotenoids, and a structural protein called apolipoprotein B-48 into a chylomicron.

Why do fats travel through the lymphatic system instead of the blood?

Here is an elegant detail. Chylomicrons are enormous as biological particles go, far too large to squeeze into the capillaries that drain the gut. So they take a different road. They enter the lacteals, specialized lymphatic vessels sitting inside each intestinal villus, travel up the lymphatic system, and empty into the bloodstream at the thoracic duct near the left collarbone.

The consequence is significant: dietary fat and everything riding with it bypasses the liver’s first pass and enters the general circulation directly. Your bloodstream receives these plant compounds before your liver ever sees them. After a fatty meal, the volume of chylomicrons is enough to make blood plasma visibly cloudy.

How do antioxidants end up inside LDL particles?

Once in circulation, lipoprotein lipase strips triglycerides out of the chylomicrons to feed muscle and fat tissue, leaving shrunken chylomicron remnants that the liver clears. The liver then exports its own lipids as VLDL, built around a bigger protein called apolipoprotein B-100. As VLDL sheds triglyceride it becomes denser, passing through an intermediate form and ending as LDL, the particle everyone argues about.

Throughout this remodeling, the fat-soluble plant compounds move between particles by exchange and transfer, so they end up distributed across LDL and HDL. This matters more than it sounds. Vitamin E and carotenoids are not floating loose in your blood; they are physically embedded in the lipid core and surface of the very particles that drive atherosclerosis. Roughly three-quarters of circulating carotenoids ride inside LDL. Vitamin C, which is water-soluble and travels separately in plasma, regenerates vitamin E after it has done its job, so the water-soluble and fat-soluble halves of a plant-rich diet work as a relay team.

Why does LDL get stuck in artery walls?

Now the electrostatics, and this is the piece worth understanding, because it explains why LDL is dangerous in a way that “cholesterol is bad” never does.

The inner lining of your arteries, the intima, contains a mesh of proteoglycans: core proteins carrying long sugar chains called glycosaminoglycans, including chondroitin sulfate, dermatan sulfate, and heparan sulfate. Those chains bristle with sulfate and carboxyl groups, which makes them strongly and persistently negatively charged.

Apolipoprotein B-100, the single large protein wrapped around every LDL particle, carries regions dense in lysine and arginine, two amino acids whose side chains are positively charged at body pH.

Opposite charges attract. When an LDL particle drifts into the artery wall, its positively charged apoB region binds directly to the negatively charged proteoglycan chains, and the particle is snagged. This is the response-to-retention model of atherosclerosis, and it reframes the whole disease: the problem is not cholesterol circulating in your blood, it is cholesterol-carrying particles being electrostatically trapped in your arterial wall. Retention is the initiating event, and everything else follows from it. The most elegant demonstration came in 2002, when researchers engineered mice whose LDL could not bind proteoglycans, changing nothing else about their cholesterol levels. Those mice developed dramatically less atherosclerosis. Same cholesterol, no trapping, little disease.

One detail matters enormously and is usually left out: that initial handshake is reversible. A healthy, unoxidized particle that gets snagged can let go again and drift back into circulation unharmed. Retention only becomes permanent because of what happens next, which is where the antioxidants earn their place in this story.

Two things make retention worse. The more apoB particles in circulation, the more binding attempts per unit time, which is why particle number predicts risk better than the cholesterol those particles happen to carry. And small dense LDL particles both penetrate the intima more easily and bind proteoglycans more avidly, because the conformation of apoB on a smaller particle exposes the positively charged binding site more readily.

How does trapped LDL turn into plaque?

A snagged particle is a sitting target, and it is now cut off from the antioxidant supply circulating in your plasma. Held in place, its polyunsaturated fatty acids are attacked by reactive oxygen species and by enzymes including myeloperoxidase and lipoxygenases, and the particle oxidizes. That is the one-way door. Oxidized particles aggregate, fuse, bind the surrounding matrix through additional mechanisms, and get devoured by macrophages, none of which can be undone. Worse, the oxidation products inflame the wall and prompt it to manufacture still more proteoglycan mesh, so every particle that turns toxic strengthens the trap waiting for the next one.

This is exactly where a particle arriving loaded with carotenoids and vitamin E has an advantage. It is carrying its own fire extinguisher into the one compartment where the bloodstream can no longer help it, so it resists the conversion, stays detachable, and starves that feedback loop. A precision note for the careful reader, since this is easy to state sloppily: oxidation does not tighten the original ionic bond itself, as it actually degrades the positively charged lysines. The accurate claim is that oxidation converts a temporary snag into permanent capture through aggregation, immune uptake, and wall remodeling.

Oxidation changes its identity. The normal LDL receptor no longer recognizes it, but macrophage scavenger receptors do, and, critically, those receptors are not downregulated by cholesterol loading the way the normal receptor is. The macrophage keeps eating. It engorges with oxidized lipid, becomes a foam cell, and dies there, leaving its lipid contents behind. Repeat this across years and you get a fatty streak, then a lipid core, then a fibrous cap, then a plaque. If that cap ruptures, you have a heart attack.

Where do plant foods interrupt this process?

At four separate points, which is why the effect compounds:

• Fewer particles to trap. Replacing saturated fat with plant fats lowers LDL and apoB particle number, which directly reduces the number of retention events. This is the dominant lever by a wide margin.
• Antioxidants stationed inside the particle. The vitamin E and carotenoids that traveled in from your food are physically located in the particle that is at risk of being oxidized, which is the most favorable position an antioxidant could occupy.
• Less oxidative and inflammatory burden overall. Fiber, polyphenols, and nitrate-rich vegetables support endothelial function and nitric oxide availability, keeping the lining less permeable and less inflamed in the first place.
• Less cholesterol absorbed to begin with. Fiber and plant sterols reduce cholesterol uptake in the gut, and animal foods are the only dietary source of cholesterol at all.

If the antioxidant mechanism is real, why did antioxidant supplements fail?

This question deserves a direct answer, because the mechanism above is genuinely elegant and elegance is exactly how people get misled.

Isolated high-dose antioxidant supplements have repeatedly failed in randomized trials, and beta-carotene supplements actually increased lung cancer in smokers. That result is real and it should keep everyone honest. The likely explanations are that a single antioxidant in pharmacological isolation behaves differently from the matrixed, low-dose, mutually regenerating mixture in whole food, and that overwhelming the system can blunt useful adaptive signaling rather than help it.

So the calibrated conclusion is this. Particle number is the lever with the strongest evidence, and it is moved by replacing animal fat with plant foods. The antioxidant story is a well-supported mechanism and a good reason to prefer whole foods over pills, but it is not a licence to believe a supplement will protect your arteries. The trials say it will not. What the trials do support is eating the food, which delivers the antioxidants, the fiber, the sterols, and the lower saturated fat load in one package, at the same time. The dietary fat side of this journey, how the two food kingdoms package their fat so differently, is in the deep dive on dietary fat. The full evidence pyramid for LDL causality, from Nobel-winning cell biology through Mendelian randomization to the drug trials, is in Cholesterol, from the ground up.

What the evidence shows, tier by tier

The single most important epistemic point in this whole post: no one tier settles a question this big. Consilience across all of them does. Here is what each level shows, with its limitations named.

What do the meta-analyses and systematic reviews find?

The 2020 Cochrane review on saturated fat reduction found 17 percent fewer cardiovascular events. The 2019 Lancet fiber series found 15–30 percent lower mortality. Aune’s 2016 whole-grain dose-response and the nuts and legumes meta-analyses point the same way. And the IARC evaluation: in 2015 a Working Group of 22 experts from 10 countries classified processed meat as Group 1, carcinogenic to humans, and red meat as Group 2A, probably carcinogenic, primarily for colorectal cancer, concluding that each 50 gram portion of processed meat eaten daily increases colorectal cancer risk by 18 percent. Limitation: meta-analyses inherit the confounding of their input observational studies.

What do the randomized controlled trials show?

Ornish’s Lifestyle Heart Trial showed angiographic regression of coronary stenosis on a plant-based program without lipid-lowering drugs. Average stenosis regressed from 40.0 to 37.8 percent in the intervention group while progressing from 42.7 to 46.1 percent in controls, with 82 percent of intervention patients showing regression, and the five-year follow-up showed continued regression and more than twice as many cardiac events in controls. Barnard’s low-fat vegan trials in type 2 diabetes beat the standard ADA diet on glycemic control, with HbA1c falling 1.23 points in the vegan group versus 0.38 in the ADA group among those not changing medications. The Bergeron and Krauss APPROACH trial found red and white meat raise LDL equally, and non-meat plant protein beats both; independent of protein source, high saturated fat raised LDL and apoB.

The cleanest control for genetics is the Stanford identical-twin trial (Landry and colleagues, JAMA Network Open, 2023), which randomized one twin in each pair to a healthy vegan diet and the other to a healthy omnivorous diet. The vegan twins dropped LDL cholesterol by roughly 14 mg/dL in eight weeks, along with fasting insulin and body weight. Same genome, different plate, measurably different cardiovascular risk.

The Finnish Leg4Life trials are the newest and among the cleanest additions. Bäck and colleagues, a six-week randomized trial in 102 working-age men, found that partially replacing red and processed meat with non-soy legumes significantly lowered total and LDL cholesterol in the legume group while both rose in the meat group, and the legume group also lost more weight while maintaining adequate B12, iodine, and iron status. Pietilä and colleagues then showed the legume swap lowered plasma cholines and raised urinary dimethylamine while leaving urinary TMAO, high-sensitivity CRP, glycoprotein acetyls, Olink cytokines, and gut-barrier proxies unchanged. PREDIMED showed a plant-forward Mediterranean pattern cut major cardiovascular events by about 30 percent, and DiRECT showed type 2 diabetes remission is achievable via weight loss, 46 percent at one year and 36 percent at two. Limitation: diet trials are short and rely on surrogate endpoints.

What do the large population cohorts show?

The Adventist Health Study-2 is the strongest longevity cohort we have, and the subject of a dedicated post. Vegans had markedly lower type 2 diabetes: prevalence 2.9 percent versus 7.6 percent in non-vegetarians, with an adjusted odds ratio of 0.51 even after adjusting for BMI, and a prospective follow-up found incident diabetes in 0.54 percent of vegans versus 2.12 percent of non-vegetarians. Vegans also had far lower hypertension and were substantially less likely to be taking antihypertensive medication. For all-cause mortality, Orlich and colleagues followed 73,308 participants and found all vegetarians combined had an adjusted hazard ratio of 0.88 versus non-vegetarians, while the vegan-specific estimate of 0.85 trended favorably but did not reach significance on its own. EPIC-Oxford and the Harvard cohorts round it out: Satija’s 2017 healthful plant-based diet index predicted lower coronary heart disease, while an unhealthful plant-based index built on refined grains, sweets, and sugary drinks predicted higher risk, underscoring that quality matters. Limitation: residual confounding and healthy-user effects, which is why the trials and mechanism matter.

What is the mechanism behind the harm from animal foods?

TMAO from carnitine and choline is the best-worked example. Koeth and colleagues found omnivores produced more TMAO than vegans and vegetarians after an identical carnitine challenge, via a gut-microbiota-dependent pathway that accelerated atherosclerosis in mice. Add heme-iron oxidative pathways, fiber fermentation to short-chain fatty acids like butyrate, which is the primary fuel for colon cells, and, more cautiously, Neu5Gc, endotoxemia, and IGF-1. Limitation: mechanism establishes plausibility, not clinical magnitude.

The fashionable counterarguments, answered

“Seed oils are toxic.”

Refuted by the linoleic acid biomarker evidence above: higher intake associates with lower cardiovascular disease and diabetes. See the full breakdown and coconut oil versus animal fat.

“Saturated fat and cholesterol were exonerated.”

Refuted by LDL causality, established through genetics, feeding trials, and three drug classes, and covered in the cholesterol post. The null cohorts, Siri-Tarino, Chowdhury, and PURE, simply compared saturated fat against refined carbohydrate, which is a bad replacement, rather than against plant oils.

“Meat isn’t inflammatory, so it’s fine.”

Inflammation was never the main charge. Colorectal cancer, cardiovascular disease, type 2 diabetes, and mortality are. The Finnish trial even confirms that short trials in healthy people cannot detect the relevant harm: swapping meat for legumes changed lipids and choline metabolites while leaving inflammatory markers unmoved.

“You can’t get nutrient X on plants.”

B12 is microbial and livestock are supplemented for it; creatine, carnitine, and taurine are non-essential and endogenously made.

What animal products cost beyond your own plate

How much worse is animal agriculture for the climate?

Poore and Nemecek’s 2018 Science meta-analysis covers 570 studies, roughly 38,700 farms, 119 countries, and 40 products representing about 90 percent of global protein and calories, making it the definitive food life-cycle dataset. Producing 100 g of protein from beef emits roughly 50 kg of CO2-equivalent on average for dedicated beef herds, with the spread running from about 9 kg at the low end to 105 kg at the 90th percentile, and even the lowest-impact animal products typically exceed the highest-impact plant proteins. More than three-quarters of global agricultural land is used for livestock despite meat and dairy supplying a much smaller share of calories and protein, and the food supply chain drives about 26 percent of anthropogenic greenhouse gas emissions.

How much land would going plant-based free up?

The land-sparing opportunity is enormous. Hayek and colleagues found that shifting to plant-based diets could sequester 332–547 gigatonnes of CO2 through ecosystem recovery, equivalent to 99–163 percent of the CO2 budget for a 66 percent chance of staying under 1.5 degrees. Beef is the largest single driver of tropical deforestation, linked to roughly 41 percent of it, and most of the world’s soy is grown for animal feed, not tofu. The full case is in the rewilding piece.

Don’t livestock eat things humans can’t?

The FAO’s own “86 percent inedible” figure is a dry-weight tally in which a tonne of straw counts like a tonne of grain. Measured in the currency that feeds people, feed crops take 36 percent of global crop calories and 53 percent of plant protein, and roughly 67 percent of US crop calories go to feed. Livestock eat about a third of global cereal production, and producing 1 kg of boneless meat takes about 2.8 kg of human-edible feed in ruminant systems and 3.2 kg in monogastric systems. The upcycling claim does not survive the accounting, as covered in the dedicated post and in Two Acres.

What about crop deaths from plant agriculture?

The argument backfires, because most cropland runs through animals, so eating plants directly requires fewer total harvested acres. As counted in The Crop Deaths Argument, the original Davis proposal was corrected by Matheny in the same journal on a simple per-consumer rather than per-hectare accounting, inverting the conclusion five to one. Per million calories delivered, chicken tops the death ledger at roughly 251, then eggs at 92, beef at 29, pork at 18, and milk at 5, against roughly 2 for plant foods (Animal Visuals, 2009). See also the pesticide argument the carnivore movement cannot answer.

How does animal farming drive antibiotic resistance and pandemics?

Roughly 70 percent of the medically important antibiotics sold in the United States are used in food animals, and antibiotic use in livestock is a major global driver of antimicrobial resistance. The GRAM study found antimicrobial resistance was directly attributable to 1.27 million deaths in 2019 and associated with 4.95 million, exceeding the toll of HIV/AIDS or malaria. Intensive animal farming is also a pandemic incubator: the CDC estimates that three out of every four new or emerging infectious diseases in people come from animals, and both H5N1 avian influenza, which spread through US dairy herds in 2024 and produced the first US death in January 2025, and the 2009 swine flu pandemic emerged from livestock systems. Slaughterhouse work also inflicts documented physical harm and psychological harm, with a 2023 systematic review finding elevated depression, anxiety, and trauma symptoms among workers.

How many animals are killed for food each year?

Humans slaughtered about 83 billion land animals in 2022, the vast majority chickens, on top of an estimated 1.1–2.2 trillion wild finfish caught annually and a further 78–171 billion farmed fish killed in 2019. Sentience science increasingly recognizes the capacity to suffer across these species. When the products are not necessary for human health, and the nutrition section above establishes they are not, the ethical asymmetry is stark: we impose enormous suffering for something we do not need.

Is a vegan diet actually endorsed by health authorities?

Are vegans just claiming it’s healthy because they care about animals?

Start with the accusation, because it contains a hidden concession. The most widely used definition of veganism describes it as seeking to exclude, as far as is practicable, exploitation of and cruelty to animals. Here is what the accusation forgets: humans are animals too. A movement organized around not exploiting animals cannot coherently wave away the health of the human animal. Caring about animals includes the human one. And even granting the worst-faith version of the motive, it changes nothing about the facts. Motive does not move a meta-analysis.

Why is the healthiest way of eating also the most vegan way?

Whole plant foods happen to be simultaneously the healthiest choice and the most vegan choice, and it works on three levels. First, whole foods are the least invasive: a pot of lentils, a bag of oats, a head of kale use the least packaging, processing energy, and industrial infrastructure of anything you can eat. Second, heavily processed vegan junk, when it is genuinely unhealthy, sits in tension with veganism itself, because if humans are animals then a company selling vegans a product that quietly harms their health is exploiting an animal for profit. Not to overstate it: processed vegan foods are useful transition foods and are generally far better for the planet and for other animals than what they replace. The point is that whole foods are the fullest expression of the idea. Third, healthy vegans are better data and better ambassadors, improving the very cohort statistics this post leans on and making the movement more persuasive.

What does the Academy of Nutrition and Dietetics say about vegan diets?

The largest body of nutrition professionals on Earth. Its current position paper, approved January 2025 and in effect through December 31, 2032, states that in adults, appropriately planned vegetarian and vegan dietary patterns can be nutritionally adequate and can offer long-term health benefits such as improving several outcomes tied to cardiometabolic disease. It adds that moderate-quality evidence links vegetarian, including vegan, patterns to reduced cardiovascular disease incidence.

Is a vegan diet safe for children?

The American Academy of Pediatrics’ pediatric nutrition guidance holds that appropriately planned vegetarian and vegan diets can be nutritionally adequate during infancy, childhood, and adolescence, provided there is attention to key nutrients such as B12, vitamin D, iron, and calcium. Because the life-stage question deserves more than a paragraph, including what the 2025 ESPGHAN review actually concluded and why outcomes in children track supplementation rather than the label, it has a full guide in Raising Children Vegan. Pregnancy, breastfeeding, and early childhood are also the life stages where an omega-3 supplement has the strongest case, covered in the omega-3 guide.

What do cardiology and cancer organizations recommend?

The American Heart Association’s 2021 dietary guidance tells people to choose healthy protein sources mostly from plants and to use liquid plant oils rather than tropical or animal fats. The 2019 ACC/AHA primary prevention guideline carries a Class I recommendation for a diet emphasizing vegetables, fruits, legumes, nuts, whole grains, and fish, and describes healthy plant-based and Mediterranean patterns as achieving this. The American Cancer Society says a healthy eating pattern centers on vegetables, fiber-rich legumes, fruit, and whole grains and limits or excludes red and processed meats. The World Cancer Research Fund recommends making wholegrains, vegetables, fruit, and pulses a major part of the daily diet.

Can a plant-based diet improve type 2 diabetes?

The American Diabetes Association’s 2019 nutrition consensus report lists vegetarian and vegan among evidence-reviewed eating patterns and reports that such patterns can reduce A1C by an average of 0.3–0.4 percent in people with type 2 diabetes, alongside weight and LDL reductions.

What do national health services around the world say?

The NHS states that with good planning and an understanding of a balanced vegan diet, you can get all the nutrients your body needs. Health Canada says to choose protein foods that come from plants more often. The Australian Dietary Guidelines state that appropriately planned vegetarian diets, including total vegetarian or vegan diets, are healthy and nutritionally adequate and appropriate for all stages of the lifecycle. The Nordic Nutrition Recommendations 2023 recommend a predominantly plant-based diet. Portugal’s Directorate-General of Health issued official national guidelines for healthy vegetarian eating. And the more cautious voice, Germany’s DGE, holds that a vegan diet can be health-promoting for healthy adults provided B12 is supplemented and the diet is well planned. In the United States, Kaiser Permanente published clinical guidance that physicians should consider recommending a plant-based diet to all their patients.

Which way of eating has the most institutional support?

Step back and notice what just happened. Pediatricians, cardiologists, oncologists, diabetes specialists, dietitians, and national health ministries across at least four continents all say the same thing. No animal-centered way of eating enjoys anything remotely like this breadth of support: not the carnivore diet, not keto, not paleo, and not the ordinary meat-and-dairy-heavy Western plate most people actually eat. There is no position paper recommending that anyone build their diet around animal foods, and no pediatric, cardiology, oncology, or diabetes body endorses it. Where omnivorous patterns do earn a nod, like the Mediterranean or DASH diets, they earn it precisely for being plant-forward and for telling people to cut red and processed meat. The approved omnivorous diets are the ones that look most like this one.

Flip it around and the asymmetry is starker still. Plant-based eating has entire evidence-based medical organizations built around it. The American College of Lifestyle Medicine recommends an eating plan based predominantly on minimally processed vegetables, fruits, whole grains, legumes, nuts, and seeds. The Physicians Committee for Responsible Medicine, founded in 1985 and now more than 17,000 physician members strong, advocates plant-based nutrition as first-line care. The Ornish Program for Reversing Heart Disease was approved by Medicare as an Intensive Cardiac Rehabilitation program in 2010, among the first lifestyle-based programs covered under that benefit and the one shown in randomized trials to reverse coronary disease through lifestyle alone. Add NutritionFacts.org, the National Health Association, TrueNorth Health Center, Hippocrates Wellness, Rochester Lifestyle Medicine, Plant-Based Health Professionals UK, and Doctors For Nutrition. Now ask the obvious question: where is the equivalent for any animal-centered way of eating? There is no College of Carnivore Medicine, no professional body for keto or paleo, and no Medicare-covered program built on meat. The reason is not conspiracy; it is arithmetic. Position papers require evidence.

The twenty clichés, answered in one line each

“Where do you get your protein?”

From the same place the cow does: plants. All nine essential amino acids are in ordinary plant foods, protein combining was debunked by the person who popularized it, and substituting plant protein for animal protein tracks with lower mortality in the Harvard and NIH-AARP cohorts.

“Humans are omnivores, we evolved to eat meat.”

What we can digest is not the same question as what keeps us healthiest for eighty years. Evolution optimized for surviving to reproduce, not for avoiding coronary artery disease at sixty. The outcome data, not the ancestral menu, is what matters.

“But we have canine teeth.”

So do gorillas, who eat almost entirely plants. Tooth shape is a poor guide to optimal diet; digestive physiology and disease outcomes are better ones.

“It’s not natural.”

Neither is fortified salt, refrigeration, or the supplemented feed that gives farmed animals their B12. Naturalness is not a health claim.

“Everything in moderation.”

True for most foods, but the phrase quietly assumes every food belongs on the plate in the first place. Processed meat is an IARC Group 1 carcinogen, so “moderation” is doing a lot of unexamined work.

“Correlation isn’t causation.”

Correct, and that is why this case climbs the entire hierarchy of evidence: feeding trials, randomized controlled trials, Mendelian randomization, drug trials, and mechanism, not just cohorts. The cohorts are corroboration, not the foundation.

“Carbs make you fat” or “carbs cause diabetes.”

Refined flour and sugar are the problem; beans, oats, and whole fruit are protective. Whole fruit associates with lower type 2 diabetes risk while juice raises it, and the highest-carbohydrate traditional populations had the lowest rates of both diseases.

“Seed oils are inflammatory.”

Higher linoleic acid measured in human blood and fat tissue tracks with less cardiovascular disease and less diabetes across pooled cohorts covering thirteen countries.

“Saturated fat has been exonerated.”

LDL causality is settled by genetics, feeding trials, and three independent drug classes. The null studies compared saturated fat against refined carbohydrate, not against plant foods.

“Plants have antinutrients.”

Phytates and oxalates modestly reduce absorption of some minerals, and soaking, sprouting, fermenting, and cooking largely handle it. The populations eating the most of these foods are not the ones with the deficiency problems. Taken apart in full in Plant “Toxins” and “Anti-Nutrients”.

“Soy messes with your hormones.”

Reed and colleagues’ 2021 meta-analysis in Reproductive Toxicology, covering 41 studies with testosterone measured in 1,753 men, found no effect on total or free testosterone, and a 2025 dose-response meta-analysis of randomized trials agreed. Soy is also among the highest-quality plant proteins available.

“You have to take supplements, so it’s not complete.”

One vitamin, made by bacteria, that farmed animals are also supplemented with. Meanwhile roughly 12.5 percent of all American adults are B12 insufficient, and US policy already tells everyone over fifty to use fortified foods or pills.

“Vegans are weak, you can’t build muscle.”

Modeling studies of fully plant-based diets scaled to bodybuilder and professional athlete calorie needs supply enough protein and leucine to maximize hypertrophy and strength, and plant-based athletes compete at the elite level across strength and endurance sport.

“Plants can’t fuel athletes.”

The stronger point is rarely made: carbohydrate availability is what actually limits endurance performance, and plants are where carbohydrate lives. Even researchers arguing for lower intakes concede the mechanism, that falling blood glucose is what curtails muscle recruitment to protect the brain. Controlled work also finds ketogenic adaptation measurably worsens exercise economy at race intensities, so the athlete argument runs the opposite direction from how it is usually deployed.

“Grass-fed and regenerative is different.”

Better on some margins, and still constrained by soil carbon saturation, land arithmetic, and scalability. It does not close the gap with eating plants directly.

“What about crop deaths?”

Most cropland runs through animals, so eating plants directly requires fewer harvested acres and causes fewer animal deaths per calorie, not more.

“But livestock eat things humans can’t.”

The 86 percent inedible figure is a dry-weight tally where straw counts like grain. Measured in human food, feed crops take 36 percent of global crop calories and 53 percent of plant protein.

“Fiber is unnecessary, there’s no essential fiber.”

“Essential” means your body cannot synthesize it, not that it is not beneficial. By that standard exercise is unnecessary too. The fiber intakes that predict the lowest mortality are nowhere near zero.

“Plants don’t want to be eaten.”

Fruit is literally a seed-dispersal bribe. And plant defense chemistry is dosed for insects thousands to millions of times smaller than you, which is why the compounds behave as mild beneficial stressors at human doses rather than poisons.

“A plant-based diet is expensive.”

Beans, lentils, rice, oats, frozen vegetables, and seasonal produce are among the cheapest calories and nutrients in any grocery store. The expensive version is the one built on specialty mock meats.

“Vegan just means Oreos and french fries.”

It can, which is exactly why this post argues for whole foods specifically. The evidence favors healthful plant-based patterns, and the unhealthful ones perform worse.

How to actually do this

What is the simplest way to start eating plant-based?

The whole plan in one sentence: eat a variety across legumes, greens, grains, fruits, vegetables, mushrooms, nuts, seeds, seaweed, and herbs and spices, take a B12 supplement, and get some sun or a vitamin D pill. Everything below is detail on top of that.

1. Week 1, get the pattern right. Build every day around those categories, leaning on legumes, whole grains, greens, and fruit. Add a B12 supplement immediately, 50 mcg daily or 2,000 mcg weekly, chewed. This single step covers the only non-negotiable gap. The Go Vegan page has starter guides, meal ideas, and further reading.
2. Weeks 2 to 4, close the minor gaps. Add a daily ALA source such as ground flax or chia; use iodized salt or occasional moderate seaweed; get sensible sun or a vitamin D3 supplement, lichen-derived; include calcium-set tofu or fortified plant milk. An algae-oil EPA and DHA supplement at around 250 mg a day is optional insurance rather than a requirement, and it matters most in pregnancy, breastfeeding, childhood, and older age. See the full omega-3 guide. And include a small amount of whole-food fat with vegetable-heavy meals, roughly three to five grams from avocado, nuts, seeds, or tahini, because carotenoids, vitamin E, vitamin K, and provitamin A are poorly absorbed without it. This does not apply to vitamin C or the polyphenols, which are water-soluble and need no fat at all.
3. Then optimize if you are an athlete, older, or in a calorie deficit. Raise total protein, distribute it across meals at roughly 0.4–0.5 g/kg per meal, favor soy and other high-quality sources, and consider creatine if you train hard.

What if the transition is hard for me?

For most people this becomes second nature within a few weeks. But some come in with more to juggle: existing health conditions, a history of disordered eating, food access or budget constraints, a family to cook for, or simply a body that seems fussier than average. If that is you, there is no shame in it and no need to white-knuckle it alone. The best move is to work with a physician, registered dietitian, or nutritionist who practices lifestyle medicine, because that approach is explicitly evidence-based, is built around exactly this kind of whole-food plant-based pattern, and lets you get personalized, safe guidance while keeping your values intact rather than being told to just eat meat. You can find practitioners through the American College of Lifestyle Medicine. If your goal is specifically weight loss, the best diet to lose weight walks the same evidence from the satiety and calorie-density angle; if you are feeding a family, Raising Children Vegan covers pregnancy through the teenage years stage by stage.

Which blood markers should I actually check?

If a serum B12, confirmed with methylmalonic acid, comes back low, increase dosing and retest. This is the one marker everyone on plants should check periodically. If ferritin is genuinely low with symptoms, pair iron-rich plants with vitamin C and recheck. If you have kidney disease, use methyl- or hydroxocobalamin rather than high-dose cyanocobalamin under a physician’s care. These are the only routine markers worth watching once the pattern is in place.

Where I am keeping the claims honest

• Much of the cohort evidence is observational. Associations cannot establish causation, and the case rests on consilience across tiers, not any single study.
• The Adventist Health Study-2 vegan-specific mortality signal, while favorable at a hazard ratio of 0.85, was not statistically significant on its own. The robust, significant vegan signals in that cohort are for cardiometabolic outcomes, diabetes and hypertension, and for vegetarians combined on mortality.
• Diet trials are short and rely on surrogate markers like LDL and HbA1c. Long-term hard-endpoint trials on whole diets are logistically near-impossible, which is precisely why mechanism and genetics carry weight here.
• Not every institution is equally enthusiastic. The German Nutrition Society is more cautious and stresses supplementation and planning. The consensus is that well-planned plant-based eating is adequate and protective, not that any plant-based diet automatically is.
• A few figures are drawn from single studies or dataset compilations and are best read as estimates: the per-million-calorie animal-death numbers, the beef greenhouse-gas mean, which varies widely by production system, and the protein-quality scores, which vary by reference pattern and study.
• The Blue Zones have faced legitimate data-quality criticism. As detailed in the Saul Newman fact-check and Blue Zones, debunked or not, the specific attacks on the validated zones do not hold up against the primary registries, and the dietary-pattern conclusion survives, but they are supporting rather than load-bearing evidence.
• “Plant-based” is not automatically healthy. Refined grains, sugar, and fried foods are plants too. The evidence favors whole-food plant-based patterns specifically.

What a day looks like: oatmeal with berries, ground flax, and fortified soy milk; a big lentil or chickpea bowl over whole grains with greens and tahini; a tofu stir-fry with broccoli, mushrooms, and brown rice; fruit, nuts, and hummus to snack on; a dash of iodized salt; and your B12. Whole plants first, a few smart supplements for the genuine gaps, and no need to obsess over the rest. That is not a diet of deprivation. It is the single most evidence-backed way to eat for your own health, and the one that asks the least of the planet and the animals we share it with.

References

Nutrition, macronutrients, and outcomes

1. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian Dietary Patterns for Adults: A Position of the Academy of Nutrition and Dietetics. J Acad Nutr Diet. 2025;125(6):831–846. PubMed
2. Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. National Academies Press; 2002/2005.
3. Song M, Fung TT, Hu FB, et al. Association of Animal and Plant Protein Intake With All-Cause and Cause-Specific Mortality. JAMA Intern Med. 2016;176(10):1453–1463. DOI
4. Huang J, Liao LM, Weinstein SJ, et al. Association Between Plant and Animal Protein Intake and Overall and Cause-Specific Mortality. JAMA Intern Med. 2020;180(9):1173–1184. DOI
5. Reynolds A, Mann J, Cummings J, et al. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet. 2019;393(10170):434–445. Full text
6. Aune D, Keum N, Giovannucci E, et al. Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality. BMJ. 2016;353:i2716. Full text
7. Goldman DM, Warbeck CB, Karlsen MC. Completely Plant-Based Diets That Meet Energy Requirements for Resistance Training Can Supply Enough Protein and Leucine to Maximize Hypertrophy and Strength in Male Bodybuilders. Nutrients. 2024;16(8):1122.
8. Goldman DM, Warbeck CB, Karlsen MC. Plant-Based Diets and Professional American Football Players. Nutrients. 2024;16(12):1903.
9. Carlsen MH, Halvorsen BL, Holte K, et al. The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide. Nutr J. 2010;9:3. DOI
10. Maoka T. Carotenoids as natural functional pigments. J Nat Med. 2020;74(1):1–16. DOI
11. Serra V, Salvatori G, Pastorelli G. Dietary polyphenol supplementation in food producing animals: effects on the quality of derived products. Animals. 2021;11(2):401. DOI

Fats, cholesterol, and LDL causality

1. Mensink RP. Effects of saturated fatty acids on serum lipids and lipoproteins: a systematic review and regression analysis. World Health Organization; 2016.
2. Hooper L, Martin N, Jimoh OF, et al. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2020;CD011737. DOI
3. Sacks FM, Lichtenstein AH, Wu JHY, et al. Dietary Fats and Cardiovascular Disease: A Presidential Advisory From the American Heart Association. Circulation. 2017;136(3):e1–e23.
4. Ference BA, Ginsberg HN, Graham I, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease: evidence from genetic, epidemiologic, and clinical studies. Eur Heart J. 2017;38(32):2459–2472.
5. Marklund M, Wu JHY, Imamura F, et al. Biomarkers of Dietary Omega-6 Fatty Acids and Incident Cardiovascular Disease and Mortality. Circulation. 2019;139(21):2422–2436. Full text
6. Wu JHY, Marklund M, Imamura F, et al. Omega-6 fatty acid biomarkers and incident type 2 diabetes. Lancet Diabetes Endocrinol. 2017;5(12):965–974.
7. Mozaffarian D, Micha R, Wallace S. Effects on coronary heart disease of increasing polyunsaturated fat in place of saturated fat. PLoS Med. 2010;7(3):e1000252.

Randomized trials

1. Ornish D, Brown SE, Scherwitz LW, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet. 1990;336(8708):129–133.
2. Ornish D, Scherwitz LW, Billings JH, et al. Intensive lifestyle changes for reversal of coronary heart disease. JAMA. 1998;280(23):2001–2007.
3. Barnard ND, Cohen J, Jenkins DJA, et al. A low-fat vegan diet improves glycemic control and cardiovascular risk factors in a randomized clinical trial in individuals with type 2 diabetes. Diabetes Care. 2006;29(8):1777–1783. PubMed
4. Bergeron N, Chiu S, Williams PT, King SM, Krauss RM. Effects of red meat, white meat, and nonmeat protein sources on atherogenic lipoprotein measures. Am J Clin Nutr. 2019;110(1):24–33. Full text
5. Landry MJ, Ward CP, Cunanan KM, et al. Cardiometabolic effects of omnivorous vs vegan diets in identical twins: a randomized clinical trial. JAMA Netw Open. 2023;6(11):e2344457.
6. Bäck S, Päivärinta E, Pellinen T, et al. Nutritional and health benefits of a partial substitution of red and processed meat with non-soy legumes: a 6-week randomized controlled trial in healthy working-age men. Eur J Nutr. 2025;64(6):259. Full text
7. Pietilä TK, Cantini E, Itkonen ST, Salonen A, Pajari AM. Replacing red meat with non-soy legumes alters choline metabolites but not systemic inflammation or proxies of gut barrier function in healthy males in a 6-week RCT. J Nutr Biochem. 2026;154:110355. PubMed
8. Hall KD, Guo J, Courville AB, et al. Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake. Nat Med. 2021;27(2):344–353.
9. Lean MEJ, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT). Lancet. 2018;391(10120):541–551.

Cohorts and mechanism

1. Tonstad S, Butler T, Yan R, Fraser GE. Type of vegetarian diet, body weight, and prevalence of type 2 diabetes. Diabetes Care. 2009;32(5):791–796. PubMed
2. Pettersen BJ, Anousheh R, Fan J, Jaceldo-Siegl K, Fraser GE. Vegetarian diets and blood pressure among white subjects: results from the Adventist Health Study-2. Public Health Nutr. 2012;15(10):1909–1916. Full text
3. Orlich MJ, Singh PN, Sabaté J, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Intern Med. 2013;173(13):1230–1238. PubMed
4. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and Unhealthful Plant-Based Diets and the Risk of Coronary Heart Disease in U.S. Adults. J Am Coll Cardiol. 2017;70(4):411–422.
5. Koeth RA, Wang Z, Levison BS, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med. 2013;19(5):576–585. Full text
6. Bao W, Rong Y, Rong S, Liu L. Dietary iron intake, body iron stores, and the risk of type 2 diabetes: a systematic review and meta-analysis. BMC Med. 2012;10:119. Full text
7. Nsabimana S, Ismail T, Lazarte CE. Enhancing iron and zinc bioavailability in maize through phytate reduction. Front Nutr. 2024;11:1478155. Full text
8. Reed KE, Camargo J, Hamilton-Reeves J, Kurzer M, Messina M. Neither soy nor isoflavone intake affects male reproductive hormones: an expanded and updated meta-analysis of clinical studies. Reprod Toxicol. 2021;100:60–67.

Environment, public health, and animals

1. Poore J, Nemecek T. Reducing food’s environmental impacts through producers and consumers. Science. 2018;360(6392):987–992. Data explorer
2. Hayek MN, Harwatt H, Ripple WJ, Mueller ND. The carbon opportunity cost of animal-sourced food production on land. Nat Sustain. 2021;4:21–24. Full text
3. Cassidy ES, West PC, Gerber JS, Foley JA. Redefining agricultural yields: from tonnes to people nourished per hectare. Environ Res Lett. 2013;8:034015. Full text
4. Shepon A, Eshel G, Noor E, Milo R. Energy and protein feed-to-food conversion efficiencies in the US and potential food security gains from dietary changes. Environ Res Lett. 2016;11:105002. DOI
5. Mottet A, de Haan C, Falcucci A, et al. Livestock: On our plates or eating at our table? A new analysis of the feed/food debate. Glob Food Sec. 2017;14:1–8. Abstract
6. Matheny G. Least Harm: A Defense of Vegetarianism from Steven Davis’s Omnivorous Proposal. J Agric Environ Ethics. 2003;16(5):505–511. Record
7. Animal Visuals. The Number of Animals Killed to Produce One Million Calories in Eight Food Categories. 2009. Analysis
8. Murray CJL, Ikuta KS, Sharara F, et al. Global burden of bacterial antimicrobial resistance in 2019. Lancet. 2022;399(10325):629–655. Full text
9. Centers for Disease Control and Prevention. About Zoonotic Diseases. CDC
10. Slade J, Alleyne E. The Psychological Impact of Slaughterhouse Employment: A Systematic Literature Review. Trauma Violence Abuse. 2023;24(2):429–440. Full text
11. International Agency for Research on Cancer. IARC Monographs evaluate consumption of red meat and processed meat. Press Release N°240; 2015. Press release
12. Our World in Data. More than 80 billion land animals are slaughtered for meat every year. Data insight
13. Mood A, Brooke P. Estimating global numbers of fishes caught from the wild annually from 2000 to 2019. Anim Welf. 2024;33:e6. Full text

Position statements and institutional guidance

1. American Heart Association. 2021 Dietary Guidance to Improve Cardiovascular Health. Circulation. 2021;144:e472–e487. Full text
2. Arnett DK, Blumenthal RS, Albert MA, et al. 2019 ACC/AHA Guideline on the Primary Prevention of Cardiovascular Disease. Summary
3. Rock CL, Thomson C, Gansler T, et al. American Cancer Society Guideline for Diet and Physical Activity for Cancer Prevention. CA Cancer J Clin. 2020. Guideline
4. Evert AB, Dennison M, Gardner CD, et al. Nutrition Therapy for Adults With Diabetes or Prediabetes: A Consensus Report. Diabetes Care. 2019;42(5):731–754. Full text
5. World Cancer Research Fund. Wholegrains, vegetables, fruit and beans. Recommendation
6. British Dietetic Association and The Vegan Society. Memorandum of understanding. Statement
7. Health Canada. Canada’s Food Guide: healthy food choices. Guide
8. NHS. The vegan diet. Guidance
9. National Health and Medical Research Council. Australian Dietary Guidelines. Guidelines
10. Nordic Council of Ministers. Nordic Nutrition Recommendations 2023. Report
11. Direção-Geral da Saúde. Linhas de Orientação para uma Alimentação Vegetariana Saudável. 2015. DGS
12. Deutsche Gesellschaft für Ernährung. Update of the DGE position on vegan diet. 2024. Position
13. Tuso PJ, Ismail MH, Ha BP, Bartolotto C. Nutritional Update for Physicians: Plant-Based Diets. Perm J. 2013;17(2):61–66. Full text
14. American College of Lifestyle Medicine. Dietary position statement. ACLM
15. Physicians Committee for Responsible Medicine. About us. PCRM
16. European Food Safety Authority. Dietary reference values for vitamin K. EFSA Journal. 2017;15(5):4780. Full text
17. American Academy of Pediatrics. Pediatric Nutrition handbook, vegetarian and vegan diets chapter. AAP

Companion posts on VeganLinked

• Go Vegan, get started
• Raising Children Vegan
• The best diet to lose weight
• Plant “Toxins” and “Anti-Nutrients”
• The Great B12 Gamble
• Do vegans need DHA and EPA? The omega-3 guide
• Saturated fat, seed oils, and heart disease
• Cholesterol, from the ground up
• Animal fat, plant fat, and the journey to your arteries
• Coconut oil versus animal fat
• The Adventist Health Studies
• What the longest-living people eat
• Blue Zones, debunked or not
• Saul Newman versus the Blue Zones
• The crop deaths argument
• Two Acres, what beef produces versus what plants can
• Rewilding the plate
• Do livestock turn food we cannot eat into protein?
• The pesticide argument the carnivore movement cannot answer
• Raw and cooked
• No, a 40 percent number in one subgroup did not overturn the meat and bowel cancer link

A note on the evidence: this post is built on primary research and the positions of major health bodies, and it aims to be honest about the strength of that evidence in both directions. Where findings are strong, it says so; where the evidence is younger, thinner, or mixed, it says that too. It is for general information and education, not personalized medical advice. For guidance specific to you, consult a physician, registered dietitian, or nutritionist, ideally one who practices lifestyle medicine.

Read more...

Ingredients

2 pounds fresh beets, peeled and sliced

1 cup white vinegar

1/2 cup water

1/2 cup granulated sugar

1 teaspoon salt

2 teaspoons black peppercorns

1 teaspoon mustard seeds

1 teaspoon whole cloves

2 whole bay leaves

 

INSTRUCTIONS.

1)  In a large pot, bring the sliced beets to a boil in water. Reduce the heat and simmer for 20-30 minutes, or until tender. Drain and set aside.

2)  In a separate saucepan, combine the white vinegar, water, granulated sugar, salt, black peppercorns, mustard seeds, whole cloves, and bay leaves. Bring the mixture to a boil, stirring to dissolve the sugar and salt.

3)  Place the cooked beets into sterilized jars.

4)   Pour the hot vinegar mixture over the beets in the jars, ensuring the beets are fully submerged. Leave about 1/2 inch of headspace at the top of each jar 

5)  Seal the jars with sterilized lids.

6)  Allow the jars to cool to room temperature, then refrigerate for at least 24 hours before serving to allow the flavors to meld.

❗️NOTE! If desired after step three you can omit sealing the jar/s and cool at room temp then completely then refrigerate for 2 days before using.

Read more...

Somewhere along the way, “raw” stopped being a preparation method and became a philosophy. And like most food philosophies, it gets something deeply right: plants, in abundance, close to their whole form. And then it overcorrects into treating heat as the enemy.

Heat isn’t the enemy. Heat is a tool. It builds some nutrients up, tears others down, disarms plant defenses, and unlocks entire food categories. The question was never raw versus cooked. The question is which foods win raw and which foods win cooked, and the science has answered it in remarkable detail.

What the stove unlocks

Does cooking tomatoes increase lycopene absorption?

Some of the most striking findings in nutrition science are about cooking increasing what your body actually receives. Lycopene absorption from tomato paste runs about two and a half times higher than from fresh tomatoes (Gärtner et al., 1997), and heating tomatoes doesn’t just rearrange lycopene into its more absorbable form; it measurably raises total antioxidant activity (Dewanto & Liu, 2002). Cooking, in that study, literally increased the food’s nutritional value.

Are cooked carrots better than raw carrots?

The same story holds for beta-carotene: the plasma response from cooked carrots and spinach ran roughly three times that of raw (Rock et al., 1998). That particular comparison landed as a strong trend rather than a slam-dunk (P=0.09), though total beta-carotene in the blood was significantly higher on the cooked diet (P<0.04); the mechanism is well established either way, because heat ruptures the plant cell walls and protein complexes that otherwise trap carotenoids.

Do you need fat to absorb carotenoids?

Yes, and this is the missing half of every claim above. Lycopene, beta-carotene, lutein, zeaxanthin, and vitamin E are all fat-soluble, which means they need dietary fat and the bile it triggers in order to be absorbed at all. Heat breaks the carotenoid out of the plant cell; fat carries it across. Skip the fat and you undo much of what the cooking just bought you. In vitro work on carrots makes the size of this vivid: carotene accessibility sits at only a few percent in raw carrot and climbs several-fold once the carrot is cooked, with fat present (Hedrén et al., 2002). Note that the landmark tomato study above used tomato paste served with oil, not naked paste.

So the practical rule has two halves, not one. Cook the tomatoes and carrots, and eat them with something fatty: a spoon of tahini, some avocado, a handful of walnuts, a drizzle of olive oil on the roasting pan. A naked salad and a dressed salad are not the same meal nutritionally, and the dressed one wins by a wide margin.

Do you get more calories from cooked food than raw food?

Cooking also gelatinizes starch, and controlled feeding work shows organisms extract significantly more net energy from cooked starch than raw (Carmody, Weintraub & Wrangham, 2011), which is exactly why long-term raw eaters run chronically under-fueled.

Does cooking destroy fiber?

No, and this matters more than most of the nutrients people argue about. Fiber survives cooking essentially intact, and the foods cooking unlocks (beans, lentils, whole grains, tubers) are the densest fiber sources on the planet. That is not a small footnote. The synthesis commissioned for the World Health Organization, pooling 185 prospective studies and 58 trials, found the highest fiber consumers had 15 to 30 percent lower all-cause and cardiovascular mortality, with the benefit greatest around 25 to 29 grams a day and still climbing beyond that (Reynolds et al., Lancet 2019). A pot of beans is one of the most efficient ways to hit that number, and it requires a stove.

What foods are unsafe to eat raw?

Then there’s what cooking makes safe. Raw kidney beans contain phytohemagglutinin; the FDA notes as few as four or five raw beans can cause violent illness. Cassava carries cyanide-generating compounds. Heat neutralizes both, kills Salmonella, E. coli, and Listeria, and, most importantly for a vegan eating pattern, opens up legumes, whole grains, and tubers: the protein, iron, zinc, and calorie backbone of every nutritionally adequate vegan diet ever documented.

What the cutting board protects

Does cooking destroy vitamin C?

Raw wins its own categories, and they matter. Vitamin C and folate are heat-fragile, and the cooking method is decisive: boiling can strip roughly half the vitamin C from a vegetable (in one careful comparison, losses ran about 55% for broccoli and 50% for spinach), while steaming the same vegetables lost only around 10–15% (Lee et al., 2018). That’s the whole argument for a mixture in one data point: keep fresh fruit and salads raw, and when you do cook, favor gentler methods over a hard boil.

There’s a bonus here that only a mixed plate can collect. Vitamin C sharply increases absorption of non-heme iron, the kind in plants, so the raw half of the meal actively improves what you get out of the cooked half. Squeeze lemon over the lentils. Put raw peppers or tomatoes in the salad beside the bean stew. Raw and cooked aren’t competing for space on the plate; they’re working on each other. More on that in the full rundown of essential nutrients on a vegan diet.

Do you have to eat broccoli raw to get sulforaphane?

Cruciferous vegetables are the elegant case, because the answer is both at once. Sulforaphane, one of the most studied protective compounds in the plant kingdom, requires the enzyme myrosinase, which cooking destroys. But the precursor, glucoraphanin, is heat-stable. So cooked broccoli is a loaded spring waiting for an enzyme, and any raw crucifer supplies it. In the lab, adding heat-stable mustard-seed myrosinase to boiled broccoli reignites sulforaphane formation (Ghawi et al., 2013); in actual humans, pairing cooked broccoli with a teaspoon of raw mustard powder raised absorbed sulforaphane more than four-fold (Okunade et al., 2018). Gentle steaming to around 60°C can actually beat raw, because it knocks out the protein that diverts glucoraphanin into inactive byproducts while sparing myrosinase (Matusheski et al., 2004).

Should you chop broccoli before cooking it?

There’s a zero-cost kitchen trick that solves the whole dilemma: chop your crucifers and let them sit before cooking. Chopping activates the myrosinase, the wait lets it finish converting glucoraphanin to sulforaphane, and sulforaphane itself, once formed, is heat-stable. Estimates of the ideal wait range from about 40 minutes (a practical rule of thumb from Elizabeth Jeffery’s lab) up to the 90 minutes the broccoli study above found optimal; anywhere in that window, the work is largely done by the time the pan gets hot.

Are raw cruciferous vegetables bad for your thyroid?

Cooking most of your crucifers reduces their goitrogenic compounds, which matters specifically for vegans, the group most likely to run low on iodine (Leung et al., 2011). Boiling is especially effective here: 30 minutes in water leaches well over half the total glucosinolates (studies report roughly 58% from Brussels sprouts up to 77% from broccoli), with shorter boils removing proportionally less. The same leaching is double-edged, since it also carries off some of the beneficial glucosinolates you’re cooking crucifers to keep, which is one more reason the sweet spot is gentle steaming or a quick sauté plus a little raw, rather than a long hard boil. The contrast in the literature is stark: a woman eating 1–1.5 kg of raw bok choy daily landed in myxedema coma (Chu & Seltzer, 2010), while 150 g of cooked Brussels sprouts daily for a month showed no thyroid effect at all (McMillan et al., 1986). Cook most, add a little raw, and secure your iodine. On that last point, keep it simple and match what this site recommends everywhere else: iodized salt is the easiest reliable source, seaweed works in moderation, and kelp in particular can overshoot the requirement badly, so treat it as a garnish rather than a staple. Every base covered.

Is cooking ever bad for you?

None of this means “cook everything as hard as possible,” and honesty requires naming cooking’s own downsides. Very high, dry heat has costs: acrylamide, a probable carcinogen, forms when starchy foods like potatoes are fried, roasted, or baked to a deep brown (it doesn’t form in boiling or steaming), and advanced glycation end-products build up under intense dry heat. Those AGEs are overwhelmingly a browned-animal-food problem and stay low in a whole-food plant diet, so they’re a minor issue here, but the principle still points the same way. The target isn’t maximal cooking; it’s gentle cooking, steaming, simmering, a quick sauté, alongside plenty of raw. Char and deep-fry are where cooking stops helping.

Greens: the perfect case study

Leafy greens might be the single best illustration of why preparation method matters as much as food choice.

Does boiling spinach reduce oxalates?

Spinach, chard, and beet greens are mineral-rich on paper and mineral-stingy in practice, because their oxalate binds calcium in the gut before you can absorb it. Only about 5% of spinach’s calcium is absorbable, versus roughly 40–60% from low-oxalate greens like kale and bok choy (Heaney & Weaver, 1988; 1990). Boiling solves most of this: soluble oxalate drops 30–87% into the cooking water you discard (Chai & Liebman, 2005). And the same heat breaks down the leaf’s cell-wall matrix so carotenoids absorb far better (Rock et al., 1998; Castenmiller et al., 1999), while shrinking the greens down so an honest portion actually fits on a plate. Worth knowing where that cargo is headed: lutein and zeaxanthin, the carotenoids concentrated in leafy greens, get pulled selectively into the retina, where they form a filter against damaging blue light, and lutein is also the dominant carotenoid in human brain tissue. Softening the leaf and adding a little fat is how those pigments actually reach your eyes and your brain rather than passing through.

Do green smoothies remove oxalates?

Blending accomplishes half the job. Mechanically shredding the leaf matrix improves beta-carotene bioavailability; Castenmiller’s team showed liquefied spinach outperformed whole leaf in humans, so a blender is, in effect, a thousand chews you didn’t have to do. But blending removes nothing. Every milligram of oxalate is still in the glass, now fully liberated. In a varied diet that’s fine; at cleanse-level extremes it isn’t. There are published cases of acute oxalate kidney injury from green-smoothie cleanses (Makkapati et al., 2018) and intensive juicing regimens (Getting et al., 2013).

Are oxalates actually dangerous?

Here’s the part that deserves to be said just as loudly: for a healthy person eating a variety of greens, oxalate is simply not a problem. Rotate spinach and chard with kale, collards, arugula, romaine, and bok choy. Keep calcium in the diet: calcium eaten alongside oxalate binds it in the gut, and the landmark data show higher dietary calcium predicts fewer kidney stones, not more (Curhan et al., 1993; Borghi et al., 2002). The narrow exceptions, people with calcium-oxalate stones or kidney disease, should lean on boiled greens and adequate calcium. Variety solves what restriction never could.

One clarification on method, since this section recommends boiling and the rest of the post recommends against it. Boiling is a targeted tool for the high-oxalate greens specifically: spinach, chard, beet greens. For kale, collards, bok choy, and the rest of the low-oxalate crowd there is no oxalate problem to solve, so the general rule applies and a quick steam or sauté keeps more of the vitamin C, folate, and chlorophyll that a hard boil would pour down the drain.

Sprouting: raw’s best tool, graded honestly

What does sprouting do to beans and grains?

Raw eating deserves full credit for its most powerful technique. Germination flips a seed’s biology on: phytate falls by roughly 25–75%, freeing up minerals; folate multiplies three- to four-fold; vitamin C appears where the dry seed had almost none; gas-producing oligosaccharides decline; protein digestibility improves. One honest caveat on that first item, because this site has argued it at length elsewhere: phytate is not simply a villain to be eliminated. It is also an iron-chelating antioxidant with its own benefits, and the anti-nutrient panic is largely overblown. Soaking, sprouting, fermenting, and leavening all reduce it comfortably, which is why nobody eating a varied diet needs to think about it. The full case is in plant “toxins” and anti-nutrients. Sprouted lentils, mung beans, chickpeas, quinoa, and buckwheat genuinely open the legume and grain categories in raw form, which is exactly why a sprouting raw diet is categorically more adequate than fruitarianism, a pattern so unsustainable it exists in the literature only as case reports of malnutrition.

Are broccoli sprouts better than broccoli?

The crown jewel belongs to raw: three-day broccoli sprouts can carry 10 to 100 times the glucoraphanin of mature broccoli (Fahey, Zhang & Talalay, 1997). That eye-popping range came from selected cultivars under lab conditions, so retail sprouts vary a lot, but even the low end of it is remarkable, and human trials back the payoff: a broccoli-sprout beverage measurably increased excretion of air-pollutant toxins in a randomized study in Qidong, China (Egner et al., 2014).

Are raw sprouts safe to eat?

One clarification before the limits, because this post talks about lectins, phytate, oxalate, and goitrogens more than most things I write: none of that is an argument against legumes, greens, or crucifers. It is an argument about preparation. Properly cooked beans are among the healthiest foods on earth, and the compounds below are a raw-preparation problem, not a plant problem.

The honest limits: sprouting does not neutralize kidney-bean phytohemagglutinin or soy’s trypsin inhibitors; those need a hard boil. It doesn’t gelatinize starch, so cooking still wins on energy. And the warm, humid sprouting environment incubates pathogens that often live inside the seed. The 2011 German fenugreek-sprout outbreak infected more than 4,000 people, caused over 900 cases of acute kidney complications, and killed 50, and the CDC advises children, pregnant people, elders, and the immunocompromised to avoid raw sprouts entirely. Sprout-then-cook captures nearly everything at once: faster cooking, better mineral access, zero outbreak risk.

What actually happens to people on the extremes

What does research say about long-term raw food diets?

This is where the debate should end, because we have outcome data. The Giessen Raw Food Study followed over 500 long-term raw foodists, people who soaked, sprouted, and did everything right within the raw framework, and found a dose-response relationship: the higher the raw fraction, the more underweight the person, with about 30% of women under 45 experiencing partial or complete loss of their menstrual cycle at the highest raw intakes (Koebnick et al., 1999). Follow-up found widespread B12 deficiency (Koebnick et al., 2005), and that finding keeps replicating: a 2022 study of strict raw food eaters found only about half were supplementing B12, with elevated homocysteine to show for it (Abraham et al., 2022). The same pattern shows up in a separate community: a dietary-record study of followers of the mostly-raw “Hallelujah” diet found intakes falling short on energy, vitamin D, calcium, zinc, and selenium (Donaldson, 2001), and a companion study in that community that actually measured blood found most participants had low serum B12 and nearly half had elevated methylmalonic acid, the metabolic fingerprint of true B12 deficiency (Donaldson, 2000). Long-term raw vegans showed clinically low bone mass (Fontana et al., 2005) and severe dental erosion (Ganss et al., 1999). And if your fruit intake is heavy and acidic, at minimum rinse with water afterward rather than brushing immediately, because enamel doesn’t grow back.

Do raw foodists absorb more nutrients?

Then there’s the most instructive dataset in this entire debate, from the same German research program. Three groups, compared head to head: average eaters at about 4 daily servings of fruits and vegetables; a mixed raw-and-cooked “wholesome nutrition” group at 7; and raw foodists at a heroic 17 servings a day, taking in roughly triple the beta-carotene. So who had the most beta-carotene actually circulating in their blood? Not the raw group eating triple the intake. The mixed raw-and-cooked group, at the middle level of consumption, had the highest plasma beta-carotene of all (Garcia et al., 2010), because cooking pried the carotenoid loose from the plant matrix that the raw eaters’ bodies couldn’t fully crack. And there’s an even sharper wrinkle worth stating honestly. Beta-carotene itself actually reads fine in raw foodists’ blood; the carotenoid that collapses is lycopene, which comes almost entirely from cooked tomato and runs low in raw eaters despite their mountains of produce (Garcia et al., 2008). That’s the bioavailability problem made visible: the one major carotenoid most dependent on cooking is precisely the one an all-raw diet can’t deliver. And it isn’t trivia; researchers treat blood carotenoid levels as a meaningful marker of overall health risk (Hercberg et al., 2009). More intake, less in the blood, is the whole raw-versus-mixed argument in a single panel.

Meanwhile, the Academy of Nutrition and Dietetics’ position stands: appropriately planned vegan diets are healthful and nutritionally adequate at every life stage, and every “appropriately planned” version in that literature is built on cooked legumes, grains, and starches as its backbone.

To my raw friends, before the comments section catches fire

Let me say this first, directly and with love: if you eat fully raw and you are genuinely thriving (labs beautiful, cycle regular, energy high, dentist bored), I believe you, and I’m glad. This piece is not an attack on you. Honestly, the raw community gets a lot right: whole foods, zero junk, real reverence for plants. Same team. Same destination.

Why don’t anecdotes count as evidence?

But this channel runs on a different fuel than testimony, and here’s why that matters. Anecdotes, even thousands of them, cannot answer a population question, because of one unavoidable distortion: survivorship bias. The people thriving on raw make the videos. The people whose weight kept sliding, whose periods stopped, whose enamel eroded, quietly add cooked food back and exit the community, and nobody films a testimonial about the diet they had to leave. That’s exactly why the Giessen study is so valuable: it didn’t sample the highlight reel, it sampled the community of more than 500 experienced, committed raw foodists, and the problems scaled with the raw fraction in a clean dose-response pattern. Dose-response is the fingerprint of a real effect, not bad luck in recruitment.

Read more...

The short version: the claim that people eating a varied whole-food plant-based diet need to fear “plant toxins” or “anti-nutrients” does not survive contact with the evidence. These compounds are dosed for insects and microbes, humans encounter them at a tiny fraction of the relevant per-kilogram exposure, and at real dietary intakes many are actively health-promoting, while the highest plant intakes predict lower mortality at every tier of evidence. The legitimate caveats are narrow and conditional. The fix is variety, a few simple kitchen habits, and individualized support for the minority who need it.

Two different arguments wearing one costume

Spend ten minutes in the carnivore corner of the internet and you’ll meet a recurring performance: broccoli is “trying to kill you,” spinach is “loaded with oxalates,” beans are “full of lectins,” grains are “anti-nutrient bombs.” It sounds like one coherent indictment. It is actually two separate claims mashed together, and keeping them apart is the first step to seeing why neither works.

“Plant toxins” refers to defense chemicals, compounds a plant synthesizes to deter things that eat it: alkaloids, glucosinolates (and their breakdown products like sulforaphane and goitrin), polyphenols, salicylates, and caffeine. The argument here is “these are literally pesticides, so they must be poisoning you too.”

“Anti-nutrients” refers to compounds that interfere with the absorption of nutrients: phytate (which binds minerals), lectins (which bind carbohydrate structures), oxalates (which bind calcium), tannins (which bind iron and protein), and goitrogens (which interfere with iodine uptake). The argument here is “these steal your minerals, so plant foods leave you deficient.”

Different mechanisms, different evidence, different rebuttals. Blurring them is rhetorically convenient because it lets a scary word from one bucket (“toxin”) borrow menace for a compound in the other. Let’s refuse the blur.

Part 1. “Plant toxins”: dosed for bugs, not for you

The size-and-scale argument

Plants can’t run. Their defense is chemistry, and that chemistry was shaped by an evolutionary arms race with the organisms that actually threaten a plant: insects, mites, nematodes, fungi, and bacteria. The dose that matters, biologically, is the dose per unit body mass of the thing being deterred, and here the numbers are decisive.

An aphid weighs on the order of 1 milligram. A caterpillar, around 1 gram. A human, about 70,000 grams. That’s the plant’s target audience versus you: a difference of four to nearly eight orders of magnitude in body mass. Now factor in how much each eats. A caterpillar routinely consumes multiples of its own body weight in leaf tissue every day; its entire existence is eating the very tissue where defense compounds concentrate. A human eating a generous 80-gram serving of broccoli is consuming roughly one one-thousandth of body weight of that plant, once, as part of a mixed meal. Stack the mass difference on top of the intake difference and the per-kilogram chemical exposure a human experiences from a serving of vegetables is thousands to millions of times lower than what the compound was “designed” to deliver to a chewing insect. “The plant is trying to poison bugs” is true and completely beside the point at human scale.

It gets better. Many of the most potent botanical insecticides act on targets humans don’t have or barely share:

• Insect-specific ion channels and receptors. Caffeine and pyrethrin-type compounds disrupt invertebrate neural signaling. Insect sodium channels and receptor subtypes differ enough from ours that the selectivity is real, and mammals carry esterases that hydrolyze pyrethrins before they can do much. It is why pyrethroid insecticides are relatively low-toxicity to us.
• Molting and developmental hormones. Numerous plant compounds mimic or antagonize insect molting hormones (ecdysteroids) and juvenile hormone. Azadirachtin, the active principle in neem, is the classic case: it disrupts molting, and humans do not molt. These are endocrine systems we simply do not possess.
• Alkaline insect midguts. Many plant protease inhibitors and lectins are tuned to the alkaline gut of insects. The sharpest illustration is Bt toxin, which needs both an alkaline midgut and specific receptors to work, and humans have an acidic stomach and lack the receptors entirely. A number of these compounds are degraded by that acidity or by cooking, though some, like raw kidney-bean lectin, are heat-stable and have to be destroyed by thorough boiling (more on that below).

And humans are not passive. We evolved as omnivores eating a chemically diverse plant diet, and we carry the toolkit to prove it: a large family of cytochrome P450 enzymes and glutathione-S-transferases that metabolize and conjugate xenobiotics for excretion; 25 functional bitter taste receptors (the TAS2R family) that let us detect and modulate intake of concentrated bitter compounds, and a liver-and-kidney clearance system that handles the trickle of plant chemistry a normal diet delivers.

On top of biology sits history. Ten thousand years of crop domestication has been, in large part, an unbroken program of selecting against toxicity and bitterness. Wild almonds carry enough amygdalin to kill; domesticated sweet almonds don’t. Wild potatoes and their glycoalkaloids, wild cucurbits and their cucurbitacins, wild lettuce and its bitter latex, all bred down. The vegetables in your kitchen are the tamed descendants of far nastier ancestors.

Hormesis: the dose makes the medicine

Here is where the carnivore argument doesn’t just fail. It inverts. At the low doses humans actually consume, many plant defense compounds act as hormetins: they impose a brief, mild, fully recoverable stress that trips the cell’s adaptive machinery, and the adaptation leaves you better off than baseline. This is not fringe theory; it’s mainstream cell biology. Dietary hormetins such as sulforaphane, quercetin, and green-tea catechins react with sensor proteins like Keap1, releasing the transcription factor Nrf2 to switch on antioxidant response elements that induce hundreds of cytoprotective genes, including superoxide dismutase and glutathione peroxidase. The same low-dose stress recruits glutathione synthesis, heat-shock proteins, autophagy, and the AMPK and sirtuin pathways.

In other words: the “toxin” gives your cells a small workout, and your cells respond by upgrading their defenses. The net effect at dietary doses is protective. This is the identical logic behind exercise, sauna, and fasting, a point we’ll return to.

Sulforaphane is the strongest worked example, and it’s human data. In Qidong, China (a region with high aflatoxin and air-pollution exposure and correspondingly high liver-cancer rates), Johns Hopkins-led teams ran randomized trials of broccoli-sprout beverages. In the glucosinolate arm, urinary excretion of detoxified aflatoxin and of the combustion carcinogen phenanthrene rose in step with sulforaphane metabolites. A later randomized trial of 291 participants (Egner et al., Cancer Prevention Research 2014) found the broccoli-sprout beverage rapidly and durably raised excretion of the benzene detoxification product, by about 61%, and of acrolein by about 23%. Mechanistically, sulforaphane is a potent inducer of Nrf2-driven phase II detoxification enzymes (NQO1, glutathione-S-transferases), inhibits histone deacetylase, and, in a randomized human and mouse study (Yanaka et al., Cancer Prevention Research 2009), suppresses Helicobacter pylori colonization and gastritis, though it reduces rather than eradicates the infection. It’s worth being precise about what these trials measured: detoxification biomarkers and bacterial colonization, not cancer incidence or cure. That’s still a defense compound demonstrably helping human physiology, which is the opposite of the scare.

Other defense compounds have suggestive-to-strong evidence: curcumin, resveratrol, EGCG, and quercetin are all documented Nrf2 activators, though much of their disease-outcome evidence remains preclinical and is complicated by bioavailability problems. The honest framing: sulforaphane is the best-substantiated, and it is reasonable to expect (though not yet proven at the same tier for each) that other dietary hormetins share in these benefits.

And most of the benefit isn’t hormesis at all

Here is the move the “plants are low-dose poison” framing depends on you not noticing. It collapses several unrelated mechanisms into one. Hormesis is real, and it explains sulforaphane and the polyphenols well. It explains almost nothing else on your plate.

• Nitrate from leafy greens and beets is reduced to nitrite and then to nitric oxide, a vasodilator. Pooled across 22 randomized trials, beetroot juice lowered systolic blood pressure by about 3.5 mmHg, and in adults with diagnosed hypertension the pooled effect runs closer to 5 mmHg. That is substrate provision. Your body is being handed a raw material, not stressed into adapting.
• Beta-glucan, the soluble fiber in oats and barley, forms a gel that traps bile acids so they are excreted rather than reabsorbed, and the liver then pulls cholesterol out of circulation to build replacement bile acids. Around 3 g a day lowers LDL by roughly 5 to 7%, an effect solid enough that both the FDA in 1997 and EFSA in 2010 authorized the health claim. That is physical chemistry happening in your intestine.
• Folate. After the United States mandated folic acid fortification of enriched grains in 1998, CDC surveillance recorded a 27% decline in neural-tube-defect-affected pregnancies, with spina bifida down 31%. Nobody got tougher from the deficiency. Babies got spina bifida.
• Vitamin C. Our lineage carries a broken GULO gene and cannot synthesize it at all. That is not a plant attacking us. It is evidence that our ancestors outsourced a metabolic function because the dietary supply was so dependable that maintaining the enzyme stopped being worth its cost. That is the exact inverse of chemical warfare.
• Fiber feeds the bacteria that produce butyrate, which the next section takes up in full.

Bile acid sequestration, nitric oxide synthesis, bacterial fermentation, and enzyme cofactors are not variations on a single theme. They are four different mechanisms, and not one of them is a stress response. The argument only works if you pretend there is only one.

Caffeine: the insecticide you happily drink

If you want a single example that detonates the “it’s a pesticide, therefore it’s bad for you” syllogism, it’s caffeine. Caffeine is unambiguously a plant insecticide. James Nathanson demonstrated in Science in 1984 that caffeine and related methylxanthines disrupt feeding, development, and reproduction in insects, and at higher doses kill them. Coffee plants ramp up caffeine production under insect attack, and researchers are actively developing caffeine-based agricultural insecticides.

And yet coffee, a warm caffeine-and-polyphenol extraction of a defended seed, is among the most reliably health-associated foods studied. The definitive synthesis is Poole et al.’s 2017 BMJ umbrella review of 201 observational meta-analyses covering 67 outcomes, which found the largest reductions at three-to-four cups a day versus none: about 17% lower all-cause mortality (relative risk 0.83, 95% CI 0.79–0.88) and roughly 19% lower cardiovascular mortality, while high-versus-low consumption was associated with an 18% lower risk of incident cancer (0.82, 0.74–0.89), plus lower risk of type 2 diabetes, Parkinson’s disease, and liver disease.

A genuine insecticide, consumed by choice by billions, tracks with living longer. A compound’s status as a “plant toxin” tells you essentially nothing about its effect on a 70-kilogram omnivore with a functioning liver.

Part 2. The microbiome: your 100-trillion-cell answer to “anti-nutrients”

The carnivore framing treats the human gut as a naive tube that either absorbs a nutrient or gets robbed of it. Real digestion is a three-way negotiation between food, host, and roughly 100 trillion microbes, and that third party changes everything.

Fiber becomes short-chain fatty acids. Fiber and resistant starch that humans can’t digest are fermented by colonic bacteria into acetate, propionate, and butyrate. Butyrate is the primary energy source for colonocytes, the cells lining your colon; it also inhibits histone deacetylase, strengthens the gut barrier, and exerts anti-inflammatory and anti-carcinogenic effects. A meat-only diet delivers essentially none of this substrate.

Polyphenols become postbiotics. Much of a polyphenol’s benefit is actually delivered by bacteria that transform it. The cleanest example is urolithin A: gut microbes convert ellagitannins from pomegranates, walnuts, and berries into urolithin A, and human trials show it induces mitophagy signatures in skeletal muscle. A randomized, placebo-controlled trial in middle-aged adults (Singh et al., Cell Reports Medicine 2022) reported roughly 12% improvements in muscle strength alongside meaningful gains in aerobic endurance and six-minute walk performance; though in fairness, the study’s pre-specified primary endpoint, peak power output, did not reach statistical significance, and the trial was funded by the company that sells the supplement. The mechanism is sound and the direction is encouraging; the “plant chemical” isn’t the finished product, your microbiome finishes it.

“Anti-nutrients” are microbial food. Phytate and resistant starch aren’t just mineral-binders, they’re fermentable substrates that feed beneficial microbes. And a healthy microbiome directly neutralizes the compounds the carnivore camp fears: Oxalobacter formigenes and other taxa degrade dietary oxalate in the gut, lowering urinary oxalate and kidney-stone risk. Tellingly, one of the biggest risk factors for losing this protection is antibiotic exposure, not eating plants.

Diversity beats labels. In the observational, self-selected American Gut Project (McDonald et al., mSystems 2018), spanning 15,096 samples across the US, UK, and Australia, people eating more than 30 different plant types per week had more diverse gut microbiomes and fewer antibiotic-resistance genes than those eating 10 or fewer. Strikingly, plant diversity tracked microbiome diversity more closely than the coarse vegan-versus-omnivore label did. It’s association rather than proof, but it’s the single best real-world signal for the variety prescription this article keeps returning to.

The contrast case. Low-fiber, animal-heavy patterns push the microbiome the other direction, toward bile-tolerant organisms and away from fiber fermenters, and load the diet with the precursors (carnitine from red meat, choline from eggs and meat) that gut microbes convert to trimethylamine, which the liver oxidizes to TMAO. TMAO is a biomarker associated with cardiovascular risk in humans and accelerated atherosclerosis in mice, from work by Stanley Hazen’s group (Nature Medicine 2013 and after); whether it is causal in people is still debated, but the diet–microbiome link is robust; that same group found vegans and vegetarians produce markedly less TMAO from a carnitine challenge because they lack the bacterial machinery. The microbiome you build determines the metabolites you make.

Part 3. “Anti-nutrients” are often protective

Phytate (IP6): the mineral-binder that behaves like a medicine

Phytate (inositol hexaphosphate, or IP6) is the marquee “anti-nutrient.” Yes, it chelates iron, zinc, and calcium and can reduce their absorption from a single meal. But that same iron-chelating property makes it a potent antioxidant: by locking up free iron, IP6 blocks the iron-catalyzed formation of hydroxyl radicals. And the anti-cancer literature is genuinely striking, though only at the preclinical level. IP6 reduces proliferation, induces apoptosis and differentiation in cancer cell lines, inhibits colorectal cancer in carcinogen-treated rodents, and reduces liver metastasis in mouse models; epidemiologically, IP6-rich diets track with lower colon-cancer risk.

Here I’ll be more careful than most write-ups, because the evidence deserves it. The robust IP6 anti-cancer evidence is preclinical. The human clinical evidence is a single small pilot, Bačić et al. (Journal of Experimental & Clinical Cancer Research, 2010), 14 breast-cancer patients randomized to IP6 plus inositol versus placebo alongside chemotherapy. It found significantly better quality of life and functional status and protection against chemotherapy-induced drops in white cells and platelets, but tumor markers were explicitly unchanged in both groups. So the honest claim is: IP6 has broad, consistent preclinical anti-cancer activity and one small human trial showing it eased chemotherapy toxicity. That’s a compound plausibly working for you, the opposite of the scare, while stopping short of proven human anti-tumor efficacy.

There’s also a phytate paradox worth noting: despite binding calcium, higher phytate intake is associated in observational work with greater bone mineral density and lower markers of vascular calcification, likely because the same crystallization-inhibiting chemistry that matters for stones also inhibits pathological calcification. The mineral-absorption concern, meanwhile, is real only for people whose diets are already marginal in iron or zinc and who don’t use the trivially simple mitigations below.

Lectins: a scare built almost entirely on raw kidney beans

The lectin panic leans hard on one true fact and then overgeneralizes wildly. The true fact: raw and undercooked red kidney beans contain phytohaemagglutinin, and eating as few as four or five raw beans can cause acute vomiting and diarrhea. Per the FDA’s own accounting, raw kidney beans carry 20,000–70,000 hemagglutinating units; proper cooking drops that to 200–400 units, a reduction of roughly 99%. Boiling for at least 10 minutes destroys the toxin (slow-cooking alone can stay too cool). The entire dietary-lectin hazard is a preparation problem that human cooking solved millennia ago.

What about lectins as anti-cancer agents? Again, honesty matters. Purified lectins are legitimate laboratory anti-cancer candidates: mushroom (Agaricus bisporus) lectin reversibly inhibits colon-cancer cell proliferation in vitro (Yu et al., Cancer Research 1993), and many plant lectins show antiproliferative activity in cells and mice. But this is preclinical, and it involves purified, often injected lectins, not dietary ones. The one plant lectin ever developed into a standardized injectable cancer drug, mistletoe lectin, actually failed its pivotal randomized trial (Steuer-Vogt et al., European Journal of Cancer 2001, 477 head-and-neck cancer patients: no reduction in recurrence, metastasis, survival, or quality of life). If a purified, dose-controlled lectin drug can’t move the needle, dietary lectins in cooked beans are not the cancer engine the scare implies. And the robust epidemiology showing legumes lower colorectal-cancer risk (Zhu et al., Scientific Reports 2015, 14 cohorts) credits fiber and isoflavones, not lectins.

So the fair conclusion is a two-parter that still defeats the carnivore claim: dietary lectins in properly prepared food are not a meaningful hazard, and legumes are among the most consistently life-extending foods studied, even if the specific “lectins cure cancer” version is not established in humans.

And legumes are life-extending. Higher legume intake is associated with lower all-cause mortality: a 2023 dose–response meta-analysis of 32 cohorts and over 1.1 million people (Zargarzadeh et al., Advances in Nutrition) found about 6% lower all-cause mortality per 50 g/day. Worth being precise about what that particular analysis did and didn’t find: it looked at mortality outcomes, and its cancer-mortality estimate did not reach statistical significance. The cancer signal rests on other work. The Zhu meta-analysis cited above found lower colorectal cancer risk with higher legume intake, and a PREDIMED sub-analysis (Papandreou et al., Clinical Nutrition 2019) found the highest total-legume intake associated with 49% lower cancer mortality and the highest lentil intake with 37% lower cancer mortality. That same PREDIMED paper also observed higher cardiovascular mortality with high legume and dry-bean intake, a reminder that single-cohort subgroup findings are hypothesis-generating and best read alongside the larger meta-analysis. Beans were also a shared staple among the world’s longest-lived populations. If lectins were the menace advertised, the most lectin-rich staple on the planet would not keep showing up in the diets of people who outlive everyone else.

Oxalates, tannins, goitrogens, saponins: real but narrow

• Oxalates matter for one group (calcium-oxalate stone-formers and people with specific predisposing conditions) and at one behavior: extreme intake. The literature carnivore influencers cite is almost entirely case reports of mono-eating: a 65-year-old woman who developed oxalate nephropathy after a green-smoothie cleanse (she also had prior gastric bypass and recent antibiotics, both predisposing); juice-diet and high-dose-vitamin-C nephropathy cases. These are stories about kilograms of a single high-oxalate food, usually in someone already predisposed.
• Tannins and polyphenols inhibit non-heme iron absorption when consumed with a meal, which is exactly why the practical fix is to keep tea and coffee away from meals. Otherwise they’re antioxidants associated with benefit.
• Goitrogens from raw brassicas only threaten the thyroid when iodine is deficient and raw-brassica intake is enormous. The famous case: an 88-year-old woman who developed myxedema coma after eating 1–1.5 kg of raw bok choy daily for months (New England Journal of Medicine, 2010). Cooking deactivates the myrosinase enzyme responsible, and adequate iodine removes the risk. This is not an argument against kale.
• Saponins, the compounds you rinse off quinoa, are bitter surfactants with largely neutral-to-beneficial profiles at dietary levels, including cholesterol-binding and some anti-inflammatory activity.

The pattern is unmistakable: every genuine harm is a story of extreme mono-consumption or a specific predisposition, and every one is solved by variety, cooking, and, rarely, individualized adjustment.

Part 4. Mono-mealing versus variety: the real dividing line

The honest concession, the one that makes this argument credible rather than cheerleading, is that a few compounds can cause trouble if you build your diet around a single high-compound food. Liters of raw spinach-and-chard smoothies every day can overload oxalate. Star fruit can harm people with kidney disease. Kilograms of raw brassicas with an iodine-poor diet can suppress the thyroid. These are real.

But look at what they have in common. They are all failures of mono-eating, and the solution is precisely what a whole-food plant-based diet is supposed to be: variety across legumes, greens, grains, fruits, vegetables, mushrooms, nuts, seeds, seaweed, herbs and spices. Spread your intake across that spectrum and no single compound accumulates to a problematic dose. No oxalate overload. No goitrogen problem. No mineral-binding crisis.

Meanwhile diversity maximizes the upside: more distinct fibers feeding more microbial species, more hormetins hitting more adaptive pathways, more of the phytochemical spectrum the American Gut Project linked to a healthier gut. Variety isn’t just the safety mechanism; it’s the benefit multiplier.

And notice the irony: the carnivore diet is the actual mono-diet. It is the eating pattern most vulnerable to the “too much of one thing” failure mode it projects onto plants.

Part 5. The absorption hacks (this part is genuinely easy)

The mineral-absorption objection is the most technically legitimate part of the anti-nutrient argument, and it is fully defused by a handful of habits. None of this requires supplements or effort beyond ordinary cooking.

• Pair vitamin C with meals. Ascorbic acid reduces ferric to ferrous iron and forms a soluble complex that resists binding by phytate and tannins, dramatically boosting non-heme iron absorption. Classic isotope work by Hallberg and colleagues found adding vitamin C to a meal increased non-heme iron absorption several-fold in single-meal studies. Squeeze lemon on your lentils, have fruit with your oatmeal, add bell peppers to the bean chili. (In fairness: the effect is larger in single-meal studies than across a whole mixed diet, but the direction is robust and the habit is free.)
• Add a little acid. Acetic and other organic acids improve mineral solubility and absorption. A splash of vinegar (with the mother, if you like) or a squeeze of citrus on greens and grains helps.
• Keep tea and coffee away from meals. Their tannins inhibit non-heme iron uptake, so enjoy them an hour or so on either side of iron-rich meals rather than with them.
• Soak, sprout, ferment, and cook. These traditional techniques slash anti-nutrient loads. Soaking and sprouting reduce phytate substantially, commonly on the order of 25–75% depending on the grain or legume and the conditions. Fermentation is more powerful still: one maize study combining soaking, germination, and lactic fermentation (Nsabimana et al., Frontiers in Nutrition 2024) cut phytate by about 86% and lowered the phytate-to-iron and phytate-to-zinc ratios by roughly 85% and 81%, substantially improving estimated iron and zinc bioavailability. Cooking destroys lectins and reduces oxalate, and boiling greens and discarding the water lowers soluble oxalate further.

For accessible, well-referenced walkthroughs of these exact hacks, NutritionFacts.org is an excellent free resource. Search it for phytates, lectins, oxalates, iron absorption, and anti-nutrients. Dr. Greger translates the primary literature into practical kitchen steps better than almost anyone.

Part 6. When people need individualized help

A subset of people run into real friction when they shift toward a plant-rich diet, and it’s worth naming honestly, because the carnivore camp weaponizes these experiences as proof plants are toxic. Usually the truth is the reverse: the change reveals a pre-existing condition that was previously masked.

• Undiagnosed SIBO or IBS can flare when fiber increases suddenly. The fix is usually gradual ramp-up, cooking, and sometimes temporary low-FODMAP structuring, not fiber elimination.
• A pre-existing kidney-stone tendency calls for oxalate-aware choices, hydration, adequate dietary calcium (which binds oxalate in the gut), and rebuilding oxalate-degrading gut flora.
• Thyroid and iodine status should be squared away with iodized salt or seaweed before anyone eats large amounts of raw brassicas.
• Histamine intolerance can react to fermented foods specifically.

These are conditions to manage, not reasons to abandon the healthiest dietary pattern. People with them may simply need help learning to diversify their plates for their individual needs, so they can acclimate and make this way of eating simple and second nature.

If that’s you, be encouraged: all great things take a little effort up front, and more for some people than others. Your best bet is a physician, registered dietitian, or nutritionist who practices lifestyle medicine, someone who can give evidence-based guidance that respects both your values and your health goals. The American College of Lifestyle Medicine and similar directories are a good place to start looking.

Part 7. The epidemiological reality check

Here’s the test that settles it. If plant foods were net-toxic, higher plant intake should predict worse health outcomes. It predicts the opposite, consistently, at every level of the evidence hierarchy:

• Fiber and whole grains. Reynolds et al.’s 2019 Lancet series pooled 185 prospective studies (roughly 135 million person-years) and 58 clinical trials, finding a 15–30% decrease in all-cause and cardiovascular mortality, plus 16–24% reductions in coronary heart disease, stroke, type 2 diabetes, and colorectal cancer, comparing the highest fiber intakes to the lowest. Benefit was greatest at 25–29 g/day, with dose-response curves suggesting more benefit above that.
• Dietary patterns. PREDIMED, DASH, and the Adventist Health Study-2 all converge on plant-forward eating lowering cardiometabolic risk and mortality.
• Legumes and longevity. Higher legume intake tracks with lower mortality across large meta-analyses, and beans are a shared feature of the world’s longest-lived populations.
• Fruits and vegetables. Dose-response meta-analyses show mortality declining as intake rises.

Note the asymmetry, because it’s the whole ballgame. The carnivore case against plants rests on mechanistic speculation (“this compound is a pesticide,” “this binds that mineral”) extrapolated from insects and petri dishes. The case for plant-rich eating rests on outcome data: actual humans, actually living longer and getting sick less, at every tier from mechanism to cohort to randomized trial. When mechanism and outcomes disagree, outcomes win.

Part 8. The argument that eats itself

The deepest problem with “these compounds activate stress pathways, therefore they’re poisons” is that the people making it don’t believe it. You can tell, because they endorse a long list of other things that work the exact same way.

• Exercise works through oxidative stress, mechanical muscle damage, and transient inflammation, followed by adaptation. It’s hormesis, activating the same Nrf2 and AMPK pathways as dietary hormetins.
• Sauna, cold plunge, and fasting, staples of the same influencer ecosystem, are textbook hormetic stressors, deliberately imposed for the adaptive rebound.
• Meanwhile the carnivore plate isn’t chemically pristine. Cooked and charred meat generates heterocyclic amines and polycyclic aromatic hydrocarbons (established mutagens) and advanced glycation end-products, and heme iron is a pro-oxidant implicated in colorectal cancer risk.

So the standard is applied to exactly one food group. If “it activates a stress response” condemns broccoli, it condemns your workout, your sauna, and your fast. And if, as the evidence actually shows, a mild and recoverable stress that triggers adaptation makes you stronger, then that is a point in favor of eating a varied plant diet, not against it.

The argument doesn’t just fail. It reverses on the person making it.

Part 9. The clichés, answered in one place

If you’ve argued about this online, you’ve met these lines. They recycle endlessly because they’re short, confident, and sound mechanistic. Here’s each one against what the evidence actually shows.

“Plants don’t want to be eaten.”

Some don’t. Many actively do. Fruit is a seed-dispersal bribe, a plant paying an animal in sugar to move its offspring. Capsaicin is the giveaway: it deters mammals, which grind seeds, but birds lack the receptor, and birds disperse seeds intact. That’s not warfare, it’s negotiation. And “want” is doing a lot of work in a sentence about an organism with no intentions.

“Broccoli is trying to kill you.”

Broccoli’s defense chemistry is dosed for a caterpillar that eats multiples of its body weight in leaf tissue daily. You are tens of thousands of times heavier and eat roughly a thousandth of your body weight in it, occasionally. The main compound in question, sulforaphane, has randomized human trials showing increased clearance of benzene and acrolein and suppression of H. pylori, a Group 1 carcinogen.

“Lectins are the new gluten.”

The entire hazard is raw or undercooked red kidney beans, which carry 20,000–70,000 hemagglutinating units. Proper boiling drops that to 200–400, about a 99% reduction. Meanwhile legumes track with roughly 6% lower all-cause mortality per 50 g/day and lower colorectal cancer risk. This is a cooking instruction masquerading as a food-group indictment.

“Spinach will give you kidney stones.”

The cited cases are almost uniformly people drinking liters of raw greens daily, often with prior gastric bypass or recent antibiotics. Boiling reduces soluble oxalate, dietary calcium binds oxalate in the gut before it reaches your kidneys, and rotating greens keeps any single load low. If you’re a known stone-former, that’s a reason for oxalate-aware choices, not for deleting vegetables.

“Phytates are mineral thieves.”

Soaking, sprouting, and fermenting cut phytate dramatically (fermentation studies show reductions up to roughly 86%), and vitamin C at meals overrides the binding effect on iron. Phytate is also a potent iron-chelating antioxidant with broad preclinical anti-cancer activity and observational links to higher bone density and less vascular calcification. The “thief” may be working for you.

“Kale and soy will wreck your thyroid.”

The signature case in the literature is an 88-year-old woman eating 1–1.5 kg of raw bok choy daily while iodine-deficient. Cooking deactivates the enzyme responsible, and iodine sufficiency removes the risk entirely. For soy specifically, meta-analyses in iodine-replete people find no clinically meaningful thyroid effect.

“Fiber is unnecessary; there’s no essential fiber.”

“Essential” is a technical term meaning your body can’t synthesize it, not a synonym for “beneficial.” By that standard exercise is unnecessary too. Reynolds et al.’s 2019 Lancet analysis of 185 prospective studies and 58 trials found 15–30% lower all-cause and cardiovascular mortality at the highest fiber intakes, with dose-response continuing above 30 g/day. Butyrate, the primary fuel for your colon cells, is made by bacteria fermenting the fiber you were told you don’t need.

“Plant nutrients aren’t bioavailable.”

Bioavailability is a solvable engineering problem, and the solutions are a lemon wedge, a splash of vinegar, an overnight soak, and drinking your tea between meals. Vitamin C raises non-heme iron absorption several-fold. Note also that non-heme iron’s regulated absorption is arguably a feature; heme iron bypasses that regulation and is a pro-oxidant implicated in colorectal cancer.

“Plant protein is incomplete.”

Every whole plant food contains all nine essential amino acids, in varying ratios. The protein-combining rule was popularized by Frances Moore Lappé in 1971 and retracted by Lappé herself in the 1981 revision, because the liver maintains an amino acid pool across meals. More on this in our nutrition deep dive.

“Humans are apex predators.”

Our lineage deleted the GULO gene and lost the ability to synthesize vitamin C, an outsourcing that only makes sense with a reliably plant-rich diet. We also carry expanded salivary amylase gene copies for starch digestion, a colon adapted to fermenting plant fiber, and 25 bitter taste receptors. Obligate carnivores have none of that profile.

“Carnivore cured my autoimmune condition.”

Elimination diets frequently produce short-term improvement; they remove ultra-processed food, alcohol, most additives, and usually a lot of calories, all at once. That tells you something was removed that mattered; it doesn’t identify which thing, and it isn’t outcome data. And there are, as far as we can find, no long-term controlled studies of morbidity or mortality on carnivore diets at any evidence tier; the widely cited 2021 survey was self-reported.

“Nutritional epidemiology is junk science.”

This argument disarms the person making it. The plant-forward case has randomized controlled trials (PREDIMED, DASH, the Lyon Diet Heart Study, and the 58 trials inside the Reynolds meta-analysis) sitting on top of the cohort data. The carnivore case has mechanistic speculation and testimonials. If observational evidence is worthless, the carnivore position has nothing left at all.

“Antioxidants were debunked.”

What failed were isolated, high-dose supplements. In the ATBC trial of 29,133 Finnish male smokers, beta-carotene supplementation produced an 18% excess in lung cancer incidence and an 8% excess in overall mortality; CARET was stopped early in 1996 after showing 28% more lung cancer and 17% higher overall mortality in the active arm. That result supports the hormesis framing rather than undermining it: mega-dose antioxidants can blunt the very adaptive signaling that makes whole foods and exercise work, an effect Ristow and colleagues suggested in 2009 (PNAS) when vitamin C and E supplementation appeared to blunt exercise-induced insulin sensitivity gains. Whole foods deliver a low-dose, matrixed signal. Pills deliver a sledgehammer.

“Nightshades cause inflammation.”

There is no human trial evidence for this. Glycoalkaloid levels in ripe tomatoes, peppers, eggplant, and potatoes are far below any threshold of concern, and these foods consistently track with lower cardiovascular and cancer risk in cohorts.

“Sugar is a drug, so fruit is poison.”

Whole fruit arrives packaged in fiber, water, and polyphenols, which is why it behaves nothing like refined sugar metabolically. Muraki et al. (BMJ 2013) found whole fruit intake associated with lower type 2 diabetes risk while fruit juice was associated with higher risk. Same sugar, different matrix.

“Soy feminizes men.”

Testosterone is the claim, and testosterone is where the evidence is cleanest. Reed et al.’s 2021 meta-analysis in Reproductive Toxicology (41 studies, with testosterone measured in 1,753 men) found no effect on total or free testosterone, and an independent 2025 dose–response meta-analysis of randomized trials (Rajaie et al., Food Frontiers) found no significant effect on testosterone, free testosterone, or SHBG overall. That newer analysis did find a nonlinear estradiol signal at high isoflavone doses, around 72 mg/day, well above the 30–50 mg/day typical of Japanese intake, and mostly in men not exercising. Worth knowing. But “a small estradiol shift at supplemental doses” is a different sentence from “soy feminizes men.”

“Liver is nature’s multivitamin.”

It’s genuinely nutrient-dense, and dense enough in preformed vitamin A (roughly 5,000 µg RAE per 100 g raw, about 550% of the Daily Value) that a single modest serving can exceed the adult tolerable upper limit of 3,000 µg, so frequent large servings risk hypervitaminosis A. More to the point, “dense in some nutrients” doesn’t address the fiber, resistant starch, and phytochemical spectrum that are simply absent from it: the substrates your microbiome runs on.

“Oxalate dumping.”

This is not a recognized clinical phenomenon. It has no diagnostic criteria, no measurable signature, and no presence in the medical literature. It exists as a post-hoc explanation for why people feel unwell after eliminating plants.

“Plants are sprayed with pesticides anyway.”

A separate argument, and one that goes badly for the carnivore position once you account for the plants fed to livestock. We handled it in the pesticide argument the carnivore movement can’t answer.

The through-line: every one of these takes a real fact (a real compound, a real case report, a real limitation of epidemiology) and inflates it past what it can carry. None of them survive contact with dose, preparation, or outcome data.

What to actually do

If you’re a healthy person worried by carnivore content: stop worrying and eat the rainbow. Concretely:

1. Aim for variety. Work toward roughly 30 different plants a week across legumes, greens, grains, fruits, vegetables, mushrooms, nuts, seeds, seaweed, herbs and spices. Diversity is both your safety margin against any single compound and your biggest lever on microbiome health.
2. Cook your beans properly: boil rather than just slow-cooking raw kidney beans, and lean on soaking, sprouting, and fermenting for grains and legumes.
3. Deploy the iron hacks: a vitamin C source at meals, a splash of acid on greens and grains, and tea or coffee kept about an hour away from iron-rich meals.
4. Cover the genuine gaps that have nothing to do with anti-nutrients: reliable B12, adequate iodine from iodized salt or seaweed, vitamin D as needed, and an omega-3 source.

If you have a specific condition (stone history, thyroid or iodine issues, IBS or SIBO, histamine intolerance), these are conditional adjustments, not elimination orders. Ramp fiber gradually, hydrate, get iodine status right before heavy raw-brassica intake, be oxalate-aware if you form stones, and get individualized help from a lifestyle-medicine clinician.

The thresholds that would change this advice: if you are eating kilograms of a single raw high-oxalate or high-goitrogen food daily, or building your entire diet around one plant, stop. That’s mono-eating, and it’s the one failure mode that’s real. If you have primary hyperoxaluria, advanced kidney disease, or a diagnosed absorptive disorder, your intake genuinely needs clinical tailoring. Short of those, the plant-toxin and anti-nutrient arguments give you no reason to remove a single food group.

Where I’m keeping the claims honest

• Evidence tiers, stated plainly. Sulforaphane’s chemoprotective effects have randomized human trial support. IP6’s anti-cancer activity is strong in cells and rodents but rests on a single small human pilot that improved quality of life and reduced chemo toxicity without changing tumor markers. The “lectins fight cancer” idea is preclinical and largely involves purified or injected lectins; the one lectin with a large human trial failed it. The population-level benefit of legumes is real but is credited to fiber and isoflavones, not specifically to lectins. None of these caveats rescue the carnivore argument; they just keep our claims calibrated.
• The legitimate conditional risks are narrow and named: oxalate for stone-formers and extreme intakes; goitrogens with iodine deficiency plus massive raw-brassica intake; phytohaemagglutinin in improperly cooked kidney beans; single-meal mineral binding, which the hacks address. These call for adjustment, not elimination.
• Mechanism labels. Grouping nitrate, beta-glucan, folate, and fiber as “not hormesis” is a claim about mechanism, not a claim that hormesis is unimportant. Sulforaphane and the polyphenols really do work through adaptive stress pathways. The point is only that they are one mechanism among several, not the master explanation.
• Observational versus causal. Much of the outcome evidence (coffee, legumes, fiber and mortality) is observational and carries residual confounding. The strength of the plant case comes from consilience across mechanism, cohorts, and randomized trials all pointing the same way, plus the striking asymmetry versus the purely mechanistic carnivore claims.
• Individual variation is real. Microbiome composition (whether you make urolithin A, whether you carry Oxalobacter) and genetics both modulate these effects, which is exactly why some people transitioning need personalized support.

References

Every study named in this post, with links to the primary source. Where a claim rests on consensus science rather than a single paper, no citation is listed.

Sulforaphane, hormesis, and caffeine

1. Egner PA, Chen JG, Zarth AT, et al. Rapid and sustainable detoxication of airborne pollutants by broccoli sprout beverage: results of a randomized clinical trial in China. Cancer Prevention Research. 2014;7(8):813–823. View source
2. Yanaka A, Fahey JW, Fukumoto A, et al. Dietary sulforaphane-rich broccoli sprouts reduce colonization and attenuate gastritis in Helicobacter pylori-infected mice and humans. Cancer Prevention Research. 2009;2(4):353–360. View source
3. Nathanson JA. Caffeine and related methylxanthines: possible naturally occurring pesticides. Science. 1984;226(4671):184–187. View source
4. Poole R, Kennedy OJ, Roderick P, et al. Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ. 2017;359:j5024. View source

Microbiome

1. Singh A, D’Amico D, Andreux PA, et al. Urolithin A improves muscle strength, exercise performance, and biomarkers of mitochondrial health in a randomized trial in middle-aged adults. Cell Reports Medicine. 2022;3(5):100633. View source
2. McDonald D, Hyde E, Debelius JW, et al. American Gut: an open platform for citizen science microbiome research. mSystems. 2018;3(3):e00031-18. View source
3. Koeth RA, Wang Z, Levison BS, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nature Medicine. 2013;19(5):576–585. View source

Phytate, lectins, and legumes

1. Bačić I, Družijanić N, Karlo R, et al. Efficacy of IP6 + inositol in the treatment of breast cancer patients receiving chemotherapy: prospective, randomized, pilot clinical study. Journal of Experimental & Clinical Cancer Research. 2010;29:12. View source
2. Yu L, Fernig DG, Smith JA, Milton JD, Rhodes JM. Reversible inhibition of proliferation of epithelial cell lines by Agaricus bisporus (edible mushroom) lectin. Cancer Research. 1993;53(19):4627–4632. View source
3. Steuer-Vogt MK, Bonkowsky V, Ambrosch P, et al. The effect of an adjuvant mistletoe treatment programme in resected head and neck cancer patients: a randomised controlled clinical trial. European Journal of Cancer. 2001;37(1):23–31. View source
4. Zhu B, Sun Y, Qi L, Zhong R, Miao X. Dietary legume consumption reduces risk of colorectal cancer: evidence from a meta-analysis of cohort studies. Scientific Reports. 2015;5:8797. View source
5. Zargarzadeh N, Mousavi SM, Santos HO, et al. Legume consumption and risk of all-cause and cause-specific mortality: a systematic review and dose-response meta-analysis of prospective studies. Advances in Nutrition. 2023;14(1):64–76. View source
6. Papandreou C, Becerra-Tomás N, Bulló M, et al. Legume consumption and risk of all-cause, cardiovascular, and cancer mortality in the PREDIMED study. Clinical Nutrition. 2019;38(1):348–356. View source
7. US Food and Drug Administration. Phytohaemagglutinin. Bad Bug Book: Foodborne Pathogenic Microorganisms and Natural Toxins Handbook, 2nd ed. View source

Oxalate and goitrogen case reports

1. Chu M, Seltzer TF. Myxedema coma induced by ingestion of raw bok choy. New England Journal of Medicine. 2010;362(20):1945–1946. View source
2. Makkapati S, D’Agati VD, Balsam L. “Green smoothie cleanse” causing acute oxalate nephropathy. American Journal of Kidney Diseases. 2018;71(2):281–286. View source

Absorption and food preparation

1. Nsabimana S, Ismail T, Lazarte CE. Enhancing iron and zinc bioavailability in maize (Zea mays) through phytate reduction: the impact of fermentation alone and in combination with soaking and germination. Frontiers in Nutrition. 2024;11:1478155. View source
2. Hallberg L, Brune M, Rossander L. The role of vitamin C in iron absorption. International Journal for Vitamin and Nutrition Research Supplement. 1989;30:103–108. View source

Fiber, nitrate, beta-glucan, and folate

1. Reynolds A, Mann J, Cummings J, Winter N, Mete E, Te Morenga L. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet. 2019;393(10170):434–445. View source
2. Bahadoran Z, Mirmiran P, Kabir A, Azizi F, Ghasemi A. The nitrate-independent blood pressure-lowering effect of beetroot juice: a systematic review and meta-analysis. Advances in Nutrition. 2017;8(6):830–838. View source
3. Whitehead A, Beck EJ, Tosh S, Wolever TM. Cholesterol-lowering effects of oat β-glucan: a meta-analysis of randomized controlled trials. American Journal of Clinical Nutrition. 2014;100(6):1413–1421. View source
4. Centers for Disease Control and Prevention. Spina bifida and anencephaly before and after folic acid mandate, United States, 1995–1996 and 1999–2000. MMWR. 2004;53(17):362–365. View source

Antioxidant supplement trials and hormones

1. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. New England Journal of Medicine. 1994;330(15):1029–1035. View source
2. Omenn GS, Goodman GE, Thornquist MD, et al. Risk factors for lung cancer and for intervention effects in CARET, the Beta-Carotene and Retinol Efficacy Trial. Journal of the National Cancer Institute. 1996;88(21):1550–1559. View source
3. Ristow M, Zarse K, Oberbach A, et al. Antioxidants prevent health-promoting effects of physical exercise in humans. PNAS. 2009;106(21):8665–8670. View source
4. Reed KE, Camargo J, Hamilton-Reeves J, Kurzer M, Messina M. Neither soy nor isoflavone intake affects male reproductive hormones: an expanded and updated meta-analysis of clinical studies. Reproductive Toxicology. 2021;100:60–67. View source
5. Rajaie S, et al. Effects of soy isoflavones on male reproductive hormones: a dose-response meta-analysis of randomized controlled trials. Food Frontiers. 2025;6(6):3166–3179. View source
6. Otun J, Sahebkar A, Östlundh L, Atkin SL, Sathyapalan T. Systematic review and meta-analysis on the effect of soy on thyroid function. Scientific Reports. 2019;9:3964. View source
7. Muraki I, Imamura F, Manson JE, et al. Fruit consumption and risk of type 2 diabetes: results from three prospective longitudinal cohort studies. BMJ. 2013;347:f5001. View source

Books and primary documents

1. Lappé FM. Diet for a Small Planet, 10th anniversary edition. New York: Ballantine Books; 1981:162.

Further reading

• Do we need animals for any nutrient?
• Raw versus cooked: why cooking is a tool, not the enemy
• What the longest-living people actually eat
• The Adventist Health Study
• The seed oil question
• The pesticide argument the carnivore movement can’t answer
• Cholesterol
• B12

Read more...

A deep dive into the science of diet and longevity, climbing the hierarchy of evidence from population patterns to identical-twin trials, and why every rung points the same direction.

What do the longest-living people actually eat?

Ask people what the world’s longest-living humans eat and you’ll get confident answers pointing in opposite directions. Some will tell you it’s fish. Some will swear it’s pork, citing Hong Kong. Others will say it doesn’t matter at all: it’s genes, or wine, or walking.

The actual answer is more interesting, and more consistent, than the debate suggests. But to get there, we have to be honest about how we know what we know. Nutrition science gets criticized, sometimes fairly, for leaning on observational studies that can’t prove cause and effect. So this post is organized the way evidence-based medicine itself is: as a climb up the hierarchy of evidence. We’ll start on the bottom rung, population-level patterns and their famous failure modes, and ascend through prospective cohorts and meta-analyses to the tools that dodge confounding by design: randomized controlled trials, identical-twin experiments, and Mendelian randomization. The case will rest its weight only on rungs built to hold it. (A note for readers who’ve followed the recent controversies over Blue Zones demography: so have we. The Blue Zones appear near the end of this post, as illustration rather than foundation, nothing here stands on them.) When every independent rung points the same direction, that convergence is how science builds confidence about anything it can’t test in an 80-year randomized trial.

Spoiler: the direction is plants. The question worth exploring is how far that direction goes.

The ladder: how do you even study what makes people live to 100?

How do scientists study diet and longevity?

No one can randomize newborns to eat one way for a lifetime and count who’s still standing at 100. So researchers climb a ladder of evidence, and each rung has known strengths and known failure modes.

Rung one: ecological studies compare whole populations, country A eats X and lives to Y. They’re cheap, vivid, and the weakest rung on the ladder, because populations differ in a thousand ways besides diet. Drawing individual-level conclusions from group-level data even has a name: the ecological fallacy. Keep that term handy; we’ll need it shortly, and note that it disqualifies rung-one arguments no matter which side they seem to favor.

What is a prospective cohort study?

Rung two: prospective cohort studies: the workhorse of nutritional epidemiology, follow tens of thousands of individuals for decades, record what they eat, and track real endpoints: heart attacks, cancers, deaths. Studies like the Adventist Health Studies, EPIC-Oxford, the Nurses’ Health Study, and the Health Professionals Follow-up Study have collectively tracked millions of person-years [1–5]. Their strengths are enormous sample sizes, hard outcomes, and long time horizons. Their weaknesses are just as well documented: confounding (health-conscious people differ in many ways beyond diet), the healthy-user bias, measurement error in food-frequency questionnaires, and reverse causation (sick people sometimes change their diets). Good studies adjust statistically for smoking, exercise, weight, education, alcohol, and more, but adjustment is never perfect. Systematic reviews and meta-analyses pool these cohorts into steadier estimates, inheriting both their power and their blind spots.

What is the hierarchy of evidence in nutrition science?

The top rungs are the tools that dodge confounding by design: randomized controlled trials that change diets and watch disease itself, identical-twin experiments that hold genetics constant, and Mendelian randomization studies that use gene variants, dealt at conception, immune to lifestyle, as nature’s own randomizer. And the capstone is triangulation: does the association show a dose-response gradient? Does it replicate across populations with different confounding structures? Do the trials, the twins, and the genes move in the predicted direction? Do the biological mechanisms make sense? When the answer to all of those is yes, the case becomes hard to dismiss: this is essentially the framework epidemiologists have used since Austin Bradford Hill laid out his causal criteria in 1965, and it’s how we became confident smoking causes lung cancer without ever running a smoking RCT [6].

Hold onto that ladder. The rest of this post climbs it in order.

A tale of two questions: average lifespan vs. exceptional longevity

Doesn’t Hong Kong eat the most meat and live the longest?

Here’s where a popular counterargument usually enters the chat: “Hong Kong has the longest life expectancy in the world and eats enormous amounts of pork and meat. Checkmate, plant people.”

It’s a genuinely useful objection, because unpacking it teaches us exactly which questions population data can and can’t answer.

Average life expectancy at birth is a whole-society statistic. It is powerfully shaped by infant and child mortality, healthcare access, wealth, sanitation, road safety, and, above all, smoking rates. When researchers at the University of Hong Kong analyzed 263 million deaths across 18 high-income regions to figure out why Hong Kong tops the tables, their answer, published in The Lancet Public Health, wasn’t pork. It was the rare combination of rapid economic development with world-leading tobacco control, producing, in the authors’ words, “fewer diseases of poverty while suppressing the diseases of affluence” [7]. Hong Kong has pushed daily smoking below 9%, 8.5% as of 2024, among the lowest rates of any developed economy, and smoking alone accounted for roughly a third to a half of Hong Kong’s longevity edge over other rich regions in the study’s models [7, 8]. Add a public hospital system that’s essentially free at the point of care, among the world’s lowest infant mortality rates, a dense walkable city where daily movement is built into life, and strong family integration of elders, and you’ve explained the ranking without ever mentioning a barbecue pork bun [7, 9].

Why life expectancy doesn’t tell you what a country eats

There’s also a cohort effect hiding in plain sight. Life expectancy tables are pulled upward by today’s oldest residents, people in their 80s and 90s who spent their childhoods and young adulthoods in an era of postwar scarcity, eating predominantly rice and vegetables with meat as a garnish. Hong Kong’s famously high per-capita meat consumption is a phenomenon of recent decades of affluence. In other words, the generation responsible for the longevity statistics did not eat the modern Hong Kong diet for most of their lives, and the generations who have are the ones now driving Hong Kong’s substantial burden of colorectal cancer, which ranks among the city’s most common cancers, consistent with everything we know about processed and red meat [10]. Tellingly, Hong Kong’s own public health researchers point out that the city’s gains in lifespan have not been matched by gains in healthspan, years lived in good health, leaving a growing burden of late-life disability [9].

So the pork argument commits the ecological fallacy twice over: it attributes a whole-society statistic to one food, and it attributes today’s diet to yesterday’s survivors. Average life expectancy is a referendum on a society’s healthcare, wealth, and cigarettes far more than on its dinner plates.

And intellectual honesty demands that this blade cut both ways. If country-level pork statistics can’t indict meat, then region-level bean statistics can’t acquit it, which is why this post refuses to build on the most famous pro-plant geography story of all, the Blue Zones. They’re real places with a real dietary pattern, and we’ll visit them near the end as corroborating color; but rung one can’t bear weight in either direction, no matter whose argument it happens to flatter. The evidence that can bear weight starts one rung up: following individuals.

Yes, longevity has many parents, let’s name them all

Before going deeper on food, let’s be upfront about everything else that shapes a long life, because pretending diet is the only variable would be exactly the kind of sloppy reasoning we just criticized.

Genetics matters, but far less than most people assume. The famous Danish twin studies estimated that only about 20–25% of the variation in human lifespan is heritable [16], and a 2018 analysis of 400 million historical family records concluded that true heritability may be well under 10% once you account for the fact that people tend to marry others with similar lifestyles [17]. Not smoking is enormous: it’s most of Hong Kong’s secret. Sleep, stress management, and physical activity all independently predict mortality. Socioeconomic status and access to healthcare shape everything from infant survival to cancer detection. Avoiding excess alcohol, drugs, and environmental toxins shows up in every longevity population.

What does the Harvard longevity study say actually matters?

And then there’s the one most longevity content skips, which may be the largest of all: relationships. The Harvard Study of Adult Development has tracked the same men since 1938, originally 724 of them, split between Harvard sophomores and boys from Boston’s poorest tenement neighborhoods, following them through work, marriage, illness, and old age for more than eight decades, with the study now extending into their children [105]. Its current director, psychiatrist Robert Waldinger, summarizes eighty-plus years of data in one sentence: good relationships keep us happier and healthier. Not wealth, not fame, not achievement. The study’s most arresting finding is a prediction test: gathering everything they knew about the men at age 50, the variable that best forecast who would be thriving at 80 wasn’t their midlife cholesterol: it was how satisfied they were in their relationships [105]. Loneliness, by contrast, behaves like a clinical risk factor; pooling 148 studies and more than 300,000 people, Holt-Lunstad’s team found strong social ties carried an odds ratio for survival of 1.50, a 50% greater likelihood of being alive at follow-up, an effect they rank alongside quitting smoking and ahead of obesity or inactivity [106]. And the quality matters more than the count, high-conflict marriages predict worse health outcomes than divorce, while feeling you can genuinely count on someone in your 80s tracks with sharper memory for longer [105].

Two things follow, and they cut in opposite directions from what you might expect. The first is that anyone building a longevity plan around supplements and biomarkers while neglecting the people in their life has the priorities backwards. The second is a caution about reading that cholesterol finding too fast: a single cohort of 724 men showing relationship satisfaction outpredicting cholesterol does not overturn the genetic and trial evidence that LDL causes heart disease: those are different questions on different rungs. Both are true. Relationships are a massive and underrated lever; LDL still clogs arteries.

Notice, too, that the world’s celebrated longevity hotspots, which we’ll visit near the end, are saturated with exactly these features: natural daily movement, strong community and purpose, low smoking, low-stress cultures [12]. So is Loma Linda. Which is precisely what makes the Adventist cohort such a useful natural experiment: it holds community, faith, purpose, and low smoking roughly constant across its members, then lets diet vary from meat-eating to vegan. The dietary gradient we’re about to examine appears on top of strong social connection, not instead of it.

Is diet really the biggest factor in longevity?

So no, diet is not the whole story, and this post has no interest in pretending otherwise. But here is why it deserves the spotlight rather than a footnote. First, it’s the single largest modifiable risk factor for death on Earth: the Global Burden of Disease collaborators pinned 11 million deaths in 2017 on dietary risks, about one in five deaths worldwide, more than tobacco [18]. Zoom into the United States and the picture sharpens further: Micha and colleagues traced 318,656 cardiometabolic deaths in a single year, 45.4% of all US deaths from heart disease, stroke, and type 2 diabetes, to suboptimal intake of just ten foods and nutrients, too much processed meat, red meat, and sodium; too few vegetables, fruits, nuts, and whole grains [74]. Second, unlike your genes, your childhood socioeconomic status, or your luck, what goes on your plate is under your direct control three times a day, starting today. And third, critically for this post, diet is the variable where the research lets us see a clean dose-response gradient. Which brings us back to the food.

The gradient: every step toward plants, measured

Do vegans have better health outcomes than meat eaters?

If plant-predominant eating drives the longevity signal, we’d predict something specific: within a single population, health outcomes should improve stepwise as diets move from meat-heavy to fully plant-based. Not a binary “vegetarians vs. everyone,” but a staircase. That is exactly what the best cohorts show.

What is the Adventist Health Study?

The Adventist Health Study-2 (AHS-2) is the crown jewel of rung two. It enrolled about 96,000 Seventh-day Adventists across North America, a famously long-lived community, centered on Loma Linda, California, whose members outlive their fellow Americans by up to a decade [15], where almost nobody smokes, almost nobody drinks, and everyone shares similar religious culture, community support, and health consciousness, yet whose members happen to span the full dietary spectrum from regular meat-eaters to vegans [1]. That internal variation is a gift to science: comparing vegan Adventists to meat-eating Adventists controls for the lifestyle confounders that plague comparisons between, say, vegans and the general public. It’s about as close to isolating the diet variable as observational research gets, and it’s the single most-argued-over dataset in nutrition, which is why it has its own companion piece: What the Adventist Health Studies Actually Tell Us About Vegan Diets, And What They Don’t.

The staircase appears on essentially every outcome measured. Average BMI climbs in orderly steps from 23.6 in vegans to 25.7 in lacto-ovo vegetarians to 26.3 in pescatarians to 28.8 in nonvegetarians [19]. Type 2 diabetes prevalence runs 2.9% in vegans versus 7.6% in nonvegetarians, with lacto-ovo, pesco, and semi-vegetarians falling neatly in between [19]. Hypertension and high cholesterol show the same gradient [20]. Vegan Adventists had the lowest overall cancer incidence of any diet group, including meaningfully lower rates of female-specific cancers [21]. And on the outcome that matters most, the landmark 2013 mortality analysis in JAMA Internal Medicine found all vegetarian patterns combined had 12% lower all-cause mortality than nonvegetarians (HR 0.88), with pescatarians at 0.81, vegans at 0.85, lacto-ovo vegetarians at 0.91, and semi-vegetarians at 0.92, and vegan men doing best of all at 0.72 [1]. The 2024 update, following the cohort through 2015, with 12,515 deaths, sharpened rather than simply confirmed this picture, and honesty requires reporting what it found. In the minimally adjusted model, vegans still died at about 0.88 times the rate of their meat-eating co-religionists; but once the researchers layered in BMI, exercise, and prevalent disease, the all-vegan advantage shrank toward statistical noise, a hazard ratio around 0.97 to 0.99 [22]. The clearest surviving signal was in men: vegan men at 65 carried a hazard ratio of 0.72, yet by 85 that edge had vanished, the point estimate drifting just above 1.0 [22]. Two things are worth saying plainly about this. First, adjusting for BMI is arguably over-adjustment, since a lower body weight is one of the mechanisms by which a vegan diet extends life, controlling it away partly controls away the diet’s own benefit. Second, the authors flag a plausible culprit for higher stroke and neurodegenerative signals in some vegans: under-supplementation of B12 and omega-3s, not the plants themselves [22]. This is the honest shape of the data, a real advantage, strongest in midlife men, sensitive to how much you adjust for diet’s own downstream effects, and contingent on doing veganism competently. (We’ll dissect the vegan-vs-pescatarian nuance shortly: it’s important, and it cuts differently than people assume.)

Do vegetarians have a higher risk of stroke?

EPIC-Oxford, the UK’s counterpart cohort of about 48,000 meat-eaters, fish-eaters, and vegetarians/vegans followed for 18 years, found vegetarians and vegans had a 22% lower rate of ischemic heart disease than meat-eaters, translating to about 10 fewer cases of heart disease per 1,000 people per decade [2]. That echoes an earlier pooled analysis of five prospective studies finding 24% lower ischemic heart disease mortality in vegetarians [23]. EPIC-Oxford also handed us a finding any honest write-up must include: vegetarians showed a higher rate of hemorrhagic stroke, about 3 extra cases per 1,000 people per decade, against the 10 fewer heart disease cases [2]. Researchers suspect very low B12 status (this cohort largely predates routine supplementation) and possibly very low cholesterol in a subset may play roles; the paper itself notes the vegetarians’ fiber and saturated fat intakes weren’t even that different from the meat-eaters’, and the net cardiovascular math still favored the plant-based groups [2, 24]. The takeaway isn’t “eat meat”: it’s “supplement B12 and eat enough of the right plants,” which modern evidence-based vegan guidance already insists on. We’ll see this theme again.

What is the plant-based diet index?

The Harvard cohorts, Nurses’ Health Study and Health Professionals Follow-up Study, together tracking over 200,000 people, added a crucial refinement: the plant-based diet index, which scores everyone (not just self-identified vegetarians) on how plant-predominant and how healthfully plant-predominant their diet is. Higher healthful plant-based scores predicted 25% lower coronary heart disease and 34% lower type 2 diabetes; higher unhealthful plant scores (refined grains, fries, soda) predicted increased risk [3, 4]. This kills two birds: it shows the plant gradient holds in mainstream American populations, and it shows the claim was never “anything without meat is magic”: it’s whole plant foods doing the work.

Does red meat increase mortality risk?

And the mirror image holds for animal foods. In the same Harvard cohorts, each additional daily serving of unprocessed red meat was associated with 13% higher mortality, and processed meat 20% higher [5]; people who increased red meat intake over an eight-year span had elevated death rates in the years that followed [25]; and statistically swapping a few percent of calories from animal protein to plant protein predicted lower mortality [26]. The same cohorts tie red meat to sharply higher type 2 diabetes risk, with processed meat worst of all [76]; a meta-analysis in Circulation found each daily serving of processed meat associated with 42% higher coronary heart disease and 19% higher diabetes incidence [75]; and Europe’s own EPIC cohort of nearly half a million people replicated the mortality signal, estimating that over 3% of all deaths could be prevented if processed meat intake fell below 20 grams a day [77]. Even within AHS-2’s health-conscious, low-meat population, unprocessed red meat intake was associated with 18% higher all-cause mortality [27]. In 2015, the WHO’s International Agency for Research on Cancer, after reviewing over 800 studies, classified processed meat as a Group 1 carcinogen (the same certainty category as tobacco, though not the same magnitude) and red meat as Group 2A, probably carcinogenic [28].

Are dairy and eggs linked to cancer?

And it isn’t only meat: the evidence reaches dairy and eggs, which is what makes this a vegan question rather than merely a vegetarian one. A meta-analysis of cohort studies in the American Journal of Clinical Nutrition found higher intakes of total dairy, milk, cheese, and dairy-derived calcium each associated with increased prostate cancer risk [78], a link the World Cancer Research Fund independently grades as limited-but-suggestive [54]. Within AHS-2 itself, the community’s parallel milk habits set up a natural comparison with a striking result: the highest dairy-milk drinkers carried about a 50% higher breast-cancer hazard than the lowest, while soy milk showed no such association [81]. For eggs, a dose-response meta-analysis of prospective studies found no link with total prostate cancer but a 47% higher risk of fatal prostate cancer per five eggs a week [80], echoing the Harvard physicians’ cohort in which men eating 2.5 or more eggs weekly had an 81% higher risk of lethal prostate cancer than those eating fewer than half an egg [79], plausibly via choline, the same TMAO precursor we’ll meet again in the mechanisms section. None of these signals is as airtight as the processed-meat literature, and honest writing should say so; but they all lean the same direction, and they’re part of why the fully plant-exclusive pattern, not just the meatless one, keeps earning the top of the staircase.

Which foods are most strongly linked to a longer life?

Now flip to the affirmative half of the ledger, because every food on the winning side has a literature of its own. Legumes may be the single most consistent dietary predictor of survival ever measured: a multi-country study of older adults across four cultures found roughly 7–8% lower mortality risk for every 20-gram daily increment, in every population studied [89]. Daily nut eaters ran about 20% lower total mortality across three decades of Harvard follow-up [90], and adding nuts or extra-virgin olive oil to a plant-forward Mediterranean pattern cut cardiovascular events by roughly 30% in the randomized PREDIMED trial, a study worth citing carefully, because critics raise its history first: the 2013 original was retracted and republished in 2018 after a statistician showed roughly a fifth of participants hadn’t been properly randomized, yet the reanalysis reran the numbers honestly and the ~30% reduction held [99]. A trial that survives its own retraction is stronger evidence, not weaker. Whole grains show a clean dose-response, with about 90 grams a day associated with roughly one-fifth lower cardiovascular risk [91]; fruit and vegetable benefits scale all the way up to ten servings a day, an intake estimated to avert nearly eight million premature deaths worldwide each year [92]; and the fiber threading through all of them, per a Lancet analysis commissioned for WHO guidelines, delivers 15–30% reductions in all-cause mortality, heart disease, type 2 diabetes, and colorectal cancer at higher intakes [93]. Daily leafy-green eaters declined slowly enough that, on the study’s own cognitive slope, they tested like people some eleven years younger, an observational association, not a fountain of youth, but a large one [94]. Even the quieter corners of the plant kingdom hold up: a meta-analysis spanning over 600,000 people tied mushroom consumption to lower all-cause mortality [95]; in Japan’s large cohorts, daily seaweed eaters had substantially lower ischemic heart disease rates [96]; and a randomized controlled-feeding trial found that seasoning an ordinary American diet with about 1.3 teaspoons of mixed herbs and spices a day modestly lowered 24-hour blood pressure [97]. When researchers bundled plant components into a single head-to-head intervention: the “portfolio” diet of nuts, soy protein, viscous fiber, and plant sterols: it lowered LDL cholesterol nearly as much as a starting dose of statin in a randomized JAMA trial [98]. There is no counterpart literature on the other side: no cohort or trial showing red meat, processed meat, or eggs add protection that a whole-plant pattern lacks: their best showings are null results, and their worst are cataloged above. Dairy’s one genuine bright spot, a probable reduction in colorectal cancer risk largely attributed to calcium [54], is achievable from plant calcium sources without accepting the prostate and breast signals we just covered.

Step back and look at the shape of all this evidence. More plants, better outcomes; more animal foods, worse outcomes; the middle positions land in the middle; and the pattern replicates in California Adventists, British vegetarians, and Boston nurses alike, populations with completely different confounding structures. Pooled meta-analyses agree: vegetarians run roughly 25% lower ischemic heart disease incidence and mortality, and vegans about 15% lower total cancer incidence [82]. And the top of the hierarchy has only firmed up since: a 2024 umbrella review commissioned by the Academy of Nutrition and Dietetics pooled 21 systematic reviews and tied vegetarian and vegan patterns to lower cardiovascular disease incidence and mortality, while a second 2024 umbrella review in PLOS ONE, spanning dozens of meta-analyses, reached the same verdict on lipids, glycemic control, and cancer risk [109, 110]. And the global INTERHEART study, spanning 52 countries, found that a short list of modifiable factors, daily fruit and vegetable intake protective among them, accounted for roughly 90% of first-heart-attack risk on every continent studied [72]. A monotonic dose-response gradient replicated across cohorts is one of Bradford Hill’s strongest signatures of causation [6]. So the natural question becomes: what happens at the top of the staircase?

Where do vegans fit? Reading the evidence like a scientist

Is veganism a diet or an ethical position?

Before reading the vegan data, we have to fix a category error that quietly distorts it. Veganism is not, at its root, a dietary pattern. As the movement has defined it since 1944, it’s an ethical commitment to avoiding animal exploitation “as far as is possible and practicable”, in food, clothing, and everything else [83]. The diet is a consequence of the ethic, not the point of it. Epidemiology, though, needs diet categories, so it borrows the word “vegan” to label everyone who excludes animal products, regardless of what they eat instead. Every other diet label in these studies describes what people eat; vegan alone describes what people refuse. Notice the measurement mismatch that creates: “pescatarian” and “Mediterranean” describe patterns people typically adopt for their health, so those labels arrive pre-bundled with health-seeking food choices. “Vegan” describes an exclusion people most often adopt for the animals, bundled with whatever plant foods they happen to like. Oreos, white bread, and french fries are all vegan.

Now hold that beside what nutrition science actually crowns. Ask which foods carry the strongest evidence for preventing chronic disease and extending healthy life, and the answer is strikingly consistent across the plant-based diet indices, the position papers, the cancer reports, and the global commissions: legumes, leafy greens and other vegetables, whole grains, fruits, mushrooms, nuts, seeds, seaweeds, herbs, and spices [3, 4, 13, 48, 54, 67]. Every single item on that list is a vegan food. The vegan pantry contains the entire evidence-backed roster, but nothing about the ethic guarantees an individual vegan eats from it. And in the early decades of any social movement, ethics recruits before health does. The vegans captured by EPIC-Oxford (enrolled 1993–2001) and AHS-2 (enrolled 2002–2007) largely predate mainstream whole-food plant-based medicine, universal B12 guidance, and even decent vegan options at the grocery store [1, 2]. This is why researchers studying “vegans” have to work so hard to control for diet quality, and why, as we’re about to see, the older cohorts mostly couldn’t.

Why is there so little research on vegans?

Here’s the awkward truth the epidemiology has to work with: there is no vegan country. No nation, no large population anywhere, has eaten fully plant-based for generations, so vegans enter these cohorts as small subgroups, roughly 5,500 in AHS-2, about 2,600 in EPIC-Oxford [1, 2]. Small numbers mean wide confidence intervals, and wide confidence intervals mean statistics can’t cleanly separate the groups at the top of the staircase. In the AHS-2 mortality data, the pescatarian point estimate (0.81) nosed ahead of the vegan one (0.85), but their confidence intervals overlapped heavily: the study cannot tell you pescatarians outlived vegans, only that both handily beat meat-eaters [1]. Statistically, a tie at the top.

But even that tie deserves scrutiny, because several confounders plausibly ran against the vegans:

Do pescatarians live longer than vegans?

Motivation shapes diet quality, and the psychology literature says the sorting runs opposite to what skeptics assume. The standard objection to favorable vegan findings is healthy-user bias: vegans, the story goes, are health-obsessed self-selectors whose numbers are inflated by unmeasured conscientiousness. The actual research on dietary motivation points the other way. Across studies, the less restrictive a form of animal-product avoidance, the more likely it is health-driven [84]. Pescatarians skew heavily health-motivated, theirs is the least restrictive pattern and the one mainstream guidelines actively bless with recommendations to eat fish regularly [85]. Vegans, by contrast, are distinguished by animal and environmental motives more than health motives [86], with the commitment running through a rejection of speciesism itself [87]. In other words, health-optimizers preferentially sort into pescatarianism, while ethics sorts people into veganism, where the only requirement is that animal products be absent, a bar Oreos famously clear. That supplies a concrete mechanism for why the vegan category harbors more nutritionally careless members despite owning the healthiest food base, why B12 gaps concentrate there, and, critically, why any residual healthy-user bias in these cohorts plausibly flatters the fish-eaters, not the vegans. The Harvard PDI data proved unhealthful plant-based eating carries real risk [4], yet AHS-2’s categories couldn’t distinguish a whole-food vegan from a junk-food vegan; nobody controlled for diet quality within groups. (One honest footnote: this motivation research comes from general populations. AHS-2’s participants all sit inside a church that promotes plant-forward eating on religious and health grounds, so the same sorting is plausible there but not directly demonstrated.)

Nobody measured B12. Vitamin B12 deficiency, entirely preventable with a supplement costing pennies, raises homocysteine and plausibly harms vessels and nerves. Many vegans in these older cohorts weren’t reliably supplementing (recall EPIC-Oxford’s stroke signal) [2, 24]. Unsupplemented veganism isn’t the diet any modern authority recommends, but it’s part of what these studies measured.

Nobody controlled for duration or consistency. A “vegan” at baseline may have adopted the diet six months ago, and food-frequency questionnaires misclassify people in both directions. Both problems dilute real effects.

Despite all that headwind, vegans still posted the lowest BMI, the lowest diabetes prevalence, the lowest hypertension, the lowest overall cancer incidence, and a mortality point estimate 15% below meat-eaters, with vegan men showing the largest survival advantage of any group in the study [1, 19–21]. If anything, the fair reading is that the epidemiology understates the potential of a well-planned vegan diet, because it mostly measured unplanned ones. And one symmetry to concede in the same breath: if the vegan category is too internally varied to lean on when a result looks unfavorable, then it is equally varied when a result looks favorable. Group averages over a category defined by refusal are blunt instruments in both directions, which is exactly why this case doesn’t rest on them.

And this is where the logic of inference matters. When a dose-response curve improves monotonically at every measured step, one serving of red meat to none, meat to fish, fish to dairy-and-eggs, most-plants to all-plants on biomarkers: the burden of proof falls on anyone claiming the curve suddenly reverses at the final step. Nothing in the data suggests it does, provided the final step is taken with whole foods and basic supplementation.

Is eating a little meat okay for your health?

Which raises the question that actually divides plant-based clinicians: not whether the curve reverses at the end, but whether it flattens. Many doctors who recommend plant-forward eating tell patients that a small amount of animal food, often pegged around 5% of calories, is unlikely to matter much. That advice has a real basis. The longest-lived populations on record ate roughly 95% plants, not 100%, with meat appearing as a rarity: Blue Zones research puts it around five times a month in small portions, and Dan Buettner’s own recommendation is framed as a range, 95% to 100% plant-based [12, 13]. Notice what that range does. It isn’t a prescription for 5% animal food; it’s an upper bound with zero explicitly inside it. The plants are credited with the work, and the small animal share is treated as tolerated rather than required.

And the honest position is that at that intake, the health difference between 5% and 0% is probably small, possibly too small for any study we have to detect. But be precise about why we can’t detect it. It isn’t that researchers measured the endpoint and found the curve flat. It’s that the endpoint was never populated: with no vegan country and no large vegan population, the lowest-animal group anyone could observe is the 5% group. “Five percent is fine” is a statement about where the data stops, not a finding about where the benefit stops. Three practical problems follow. First, 5% of what, processed meat and occasional fish carry very different risk profiles, and a percentage target flattens a distinction the evidence says matters enormously. Second, a threshold is harder to hold than a rule; in practice “a little” drifts, and clinicians often end up recommending zero precisely because it’s the more stable instruction. Third, and most telling: the case for keeping that 5% is about palatability and adherence, not physiology. Nobody argues the animal share is doing nutritional work that plants plus a B12 supplement can’t do. So even granting the generous reading, that the last 5% is a wash for your health, the health argument has simply gone quiet at that margin. It hasn’t come out in favor of anything. And when the health ledger is neutral, the remaining question isn’t medical: if you don’t need it, why eat it?

To move beyond inference, though, we need the tools that don’t share epidemiology’s weaknesses. That’s next.

Lifespan vs. healthspan: aiming at the target you can actually hit

What is the difference between lifespan and healthspan?

A quick but essential reframe before the heavy evidence. Lifespan, your final number, is partly hostage to genetics and plain luck: the drunk driver, the rare mutation, the pandemic. As we saw, genes explain perhaps a fifth of lifespan variation, maybe far less [16, 17]. Healthspan: the years you live free of chronic disease, disability, and a pharmacy’s worth of prescriptions, is a different animal. It is overwhelmingly determined by the modifiable factors, diet chief among them, and it is where the gap between eating patterns yawns widest.

Stanford physician James Fries called the goal “compression of morbidity”: pushing the onset of chronic disease so late that the sick, dependent phase of life shrinks to a brief window at the very end [29]. This is precisely what the plant-based data show. Adventist vegetarians don’t just die later; along the way they take fewer medications, undergo fewer surgeries, and carry less diabetes, hypertension, and heart disease [15, 19, 20]. Contrast that with Hong Kong, where researchers openly worry that lifespan has outrun healthspan, leaving more years of late-life dependency [9]. Quality of life is not a consolation prize for the years column, and conveniently, the same choices that expand healthspan (preventing the heart attacks, strokes, cancers, and diabetes that kill most people) are the ones that feed back into lifespan anyway. Aim at healthspan; lifespan largely comes along for the ride, minus the luck you never controlled.

So: which diet maximizes healthspan? Epidemiology gave us the gradient. Now let’s stress-test it with every tool that doesn’t depend on observational data.

Triangulation: the evidence that doesn’t share epidemiology’s weaknesses

Can a plant-based diet reverse heart disease?

Randomized controlled trials. You can’t randomize lifespans, but you can randomize diets and watch disease itself. Dean Ornish’s Lifestyle Heart Trial randomized patients with coronary artery disease to a whole-food, very-low-fat vegetarian diet plus lifestyle changes or usual care: at one year, 82% of the intervention group showed measurable regression of their arterial blockages on angiography, and at five years the diet group had continued regression while controls progressed and suffered more than twice the cardiac events [30, 31]. The trial was small, a few dozen patients, and bundled diet with exercise and stress management, so it proves the package works rather than isolating the plants; but the direction is unambiguous. Caldwell Esselstyn’s case series at the Cleveland Clinic, not a randomized trial, and reliant on self-selected adherents, put 198 patients with established cardiovascular disease on whole-food plant-based nutrition: among the 89% who stuck with it, recurrent cardiac events hit 0.6%, versus 62% among those who abandoned the diet [32]. Take the exact numbers with the grain of salt any uncontrolled series deserves; the effect size is nonetheless extraordinary. The BROAD trial in New Zealand randomized patients to a whole-food plant-based diet with no calorie counting and recorded the largest BMI reduction of any comparable trial at 6 and 12 months [33]. Neal Barnard’s randomized trials found a low-fat vegan diet outperformed the American Diabetes Association’s own recommended diet for glycemic control in type 2 diabetes [34]. Even DASH: the diet that topped the AHA’s rankings, earned its reputation in randomized feeding trials where a pattern centered on fruits and vegetables lowered blood pressure within weeks [73]; it’s a plant-forward design that can be executed fully plant-based, pointing the same direction while stopping short of the destination. These are causal demonstrations that moving all the way to plants changes disease trajectories, not just biomarkers.

What did the Stanford identical twin vegan study find?

Twin studies, genetics, controlled. In 2023, Stanford ran the study skeptics kept asking for: 22 pairs of identical twins, one twin randomized to a healthy vegan diet, the other to a healthy omnivorous diet, for eight weeks. Same genes, same upbringing, both arms eating well, isolating the animal-food variable about as cleanly as human research allows. The vegan twins ended with significantly lower LDL cholesterol (~14 mg/dL lower, from an already-healthy baseline), roughly 20% lower fasting insulin, and 4.2 pounds more weight loss than their genetically identical omnivore siblings [35]. A follow-up analysis of the same twins found the vegan arm reduced biological age acceleration across multiple epigenetic clocks: they got younger by molecular measures of aging [36]. Senior author Christopher Gardner’s conclusion was that, thinking about longevity, most people would benefit from shifting further toward plants [35].

Does cholesterol actually cause heart disease?

Mendelian randomization, nature’s own RCT. MR studies use gene variants, randomly assigned at conception and immune to lifestyle confounding, to test causality. MR has established beyond reasonable doubt that LDL cholesterol and ApoB-containing lipoproteins cause atherosclerotic cardiovascular disease, cumulatively, over a lifetime [37]. That matters here because randomized trials consistently show whole-food plant-based and vegan diets produce the largest dietary LDL reductions, while the primary dietary drivers of LDL are saturated fat and cholesterol, found overwhelmingly in animal foods [30, 34, 35]. The causal chain is welded shut at both ends: diet moves LDL (trials), and LDL moves heart disease (genetics). Similarly, MR studies implicate higher IGF-1 levels in causing breast, prostate, and colorectal cancers [38], and animal protein intake raises IGF-1, with studies finding vegans have the lowest circulating levels [39, 40]. One cohort analysis found that adults 50–65 reporting high protein intake, predominantly animal-based, had a fourfold increase in cancer death risk while plant protein showed no such association: the authors themselves reached for a comparison to smoking, though the honest framing is narrower: it’s a striking signal from a modest NHANES sample, it reversed entirely after 65 (where higher protein turned protective), and it’s worth citing for the mechanism rather than the shock value [40].

How do plant-based diets work biologically?

Mechanisms: the biology all points the same way. Fiber, found only in plants, feeds gut bacteria that produce short-chain fatty acids like butyrate, which nourish the colon, tame inflammation, and improve metabolic health; a Harvard experiment showed that even a few days on an animal-based diet rapidly shifts the microbiome toward bile-tolerant, inflammation-associated species [41]. Carnitine and choline, concentrated in meat, eggs, and dairy, are converted by gut microbes into TMAO, a metabolite that promotes atherosclerosis, and vegans, lacking the microbes and the substrate, barely produce it even when fed carnitine [42, 43]. A meta-analysis found vegetarians and vegans have significantly lower C-reactive protein, the blood’s cardinal inflammation marker [44]. Ornish’s team showed comprehensive plant-based lifestyle change increased telomerase activity and, over five years, actually lengthened telomeres: the protective chromosome caps whose erosion tracks cellular aging, while controls’ telomeres shortened [45, 46]. Add heme iron’s oxidative and nitroso-compound chemistry, advanced glycation end-products, and the inflammatory sugar molecule Neu5Gc from red meat, and you have half a dozen independent biological pathways, each separately predicting exactly the gradient the cohorts observe [28, 47]. When mechanism, genetics, randomized trials, twins, and epidemiology all converge, that’s not a coincidence. That’s consilience: the strongest form of proof available for a question no ethics board would ever let us test directly.

So what about the Blue Zones?

Are the Blue Zones real or debunked?

You may have noticed that this post has now made its entire case, gradient, trials, twins, genetics, mechanisms, without once leaning on the most famous longevity story in popular culture. That was deliberate. With the load-bearing walls in place, the Blue Zones can be enjoyed for what they are: corroborating color.

The story: demographers Michel Poulain and Gianni Pes, later joined by author Dan Buettner, identified regions with unusual concentrations of very old people, Okinawa (Japan), Sardinia (Italy), Nicoya (Costa Rica), Ikaria (Greece), and Loma Linda, California [11, 12]. These places differ in geography, religion, and cuisine, but their plates rhyme: an analysis of 154 dietary surveys conducted across the five zones found that roughly 95% of what the longest-lived residents ate came from plants, beans above all, plus whole grains, vegetables, greens, tubers, fruits, and nuts, with meat treated as a rarity or celebratory food [13]. Traditional Okinawans are the starkest case: mid-century surveys show they derived about 69% of their calories from sweet potatoes alone, with well over 90% of the diet from plants, and Okinawa long held the world’s highest concentration of centenarians and the longest disability-free life expectancy [14]. Sardinia’s hotspot runs on minestrone, sourdough, and fava beans; Nicoya on the “three sisters” of beans, squash, and corn; Ikaria on greens, legumes, and olive oil [12, 13].

Now, the heat. Demographic critics, most prominently Saul Justin Newman, whose analysis of age-record errors and pension fraud in extreme-age data earned an Ig Nobel Prize, argue that some celebrated old-age clusters partly reflect bad paperwork rather than good habits [88]. The skeptic’s case deserves its strongest form, and Newman is right that most raw age claims at the extreme are unreliable. Two things temper him. First, his central paper remains a preprint, not peer-reviewed, while in 2025 the gerontologists Austad and Pes answered him in The Gerontologist, walking through the multi-source cross-validation (birth ledgers, church archives, censuses) behind each validated zone [107]. Second, his load-bearing examples don’t hold up claim by claim against the primary sources: Japan’s notorious “230,000 missing centenarians” came off the koseki family register, a genealogical record with no purge mechanism, while the life-expectancy figures are built from the resident registry, which was untouched; Greece’s 9,000 figure was a pension payment roll, while Ikaria was validated years earlier on separate records; and Costa Rica’s age exaggeration had already been caught, measured, and published by Luis Rosero-Bixby himself, the very researcher Newman accuses, who then rebuilt his estimates from birth-ledger dates to route around it. The full accounting is its own post, Saul Newman vs. the Blue Zones: A Fact-Check, with the broader ten-pillar version, covering Sardinia’s self-auditing verification and the Loma Linda zip-code fallacy, in Blue Zones: Debunked or Not?. The fairest reading leaves one honest casualty: Nicoya, whose longevity advantage Rosero-Bixby has since documented fading in later-born cohorts [107]. Individual supercentenarian claims warrant deep suspicion; the diet-longevity signal survives at the population level. And here’s the thing, anyway: this post doesn’t need to referee that fight, and its structure is the reason. Strike every Blue Zone from the record tomorrow and not one preceding section changes. The cohorts still show their gradient, the twins still show their LDL, the genes still show their causality.

And notice the neat closing of the loop. The one Blue Zone built on modern vital records with a formal prospective cohort attached is Loma Linda, and that cohort is AHS-2, the individually tracked, statistically adjusted dataset this post already spent its middle chapters mining [1, 15]. It’s also the one zone without the confounding problem the others carry: Adventists live in the same country as their neighbors, shop at the same supermarkets, use the same medical system, and sit in the same unremarkable California county, so what differs is mainly what they do. And the gradient shows up here too, Adventist men outlived other Californians by 7.3 years, while vegetarian Adventist men outlived them by 9.5 [15]. Same religion, same town, more of the behavior, more of the benefit. Critics who dismiss Loma Linda as the freeway-exit Blue Zone are attacking the geographic packaging; underneath it sits the cleanest natural experiment in the field. The single Blue Zone that was converted into rigorous rung-two science delivered the plant-based gradient in full. The Blue Zones are the illustration on the cover; the evidence above is the book.

The steelman corner: the best studies from the other side

What are the best arguments against a plant-based diet?

A case that only rehearses friendly evidence isn’t a case; it’s a pep rally. So here are the exhibits that do the heaviest lifting in arguments against plant-based eating, each taken seriously, and each placed honestly on the ladder we built at the start.

What did the PURE study really show?

Exhibit one: the PURE study. In 2017, the Prospective Urban Rural Epidemiology study of over 135,000 people across 18 countries reported that higher carbohydrate intake was associated with higher mortality, while higher fat intake, including saturated fat, was associated with lower mortality [100]. Cue a thousand butter headlines. But look at who was actually being compared. PURE’s high-carbohydrate participants were overwhelmingly low-income populations eating the bulk of their calories as white rice and refined grains, poverty diets, not whole-plant diets, and in those settings, meat and fat intake are markers of affluence, which travels with healthcare access, sanitation, and lower infectious-disease mortality; much of the excess death in the high-carb strata wasn’t cardiovascular at all. The study contained essentially no whole-food plant-based eaters to test. And here’s the part the butter headlines skipped: PURE’s own companion paper, published in the very same issue of The Lancet, found that higher intake of fruits, vegetables, and legumes predicted lower mortality, with legumes protective in their own right [101]. Where its saturated-fat association conflicts with hundreds of controlled feeding trials and the Mendelian randomization evidence on LDL [37], it is a single confounded multi-country food questionnaire arguing against rungs above it. Where PURE is methodologically strongest, whole plant foods: it agrees with this post.

Why did some scientists say to keep eating red meat?

Exhibit two: the 2019 “keep eating red meat” papers. The NutriRECS consortium’s guidelines in the Annals of Internal Medicine made global news by advising adults to simply continue their current red and processed meat consumption [102]. Read past the headline and something odd appears: the panel’s own meta-analyses found the same directional risk reductions everyone else finds for eating less meat. What changed was the grading, not the data. The authors applied GRADE, a certainty framework built for drug trials, which rates virtually all nutritional epidemiology “low certainty” because thirty-year randomized feeding trials don’t exist. By that yardstick, the evidence that cigarettes cause lung cancer is also “low certainty”, there were never any smoking RCTs either [6]. Major health bodies, including the AHA, the American Cancer Society, and Harvard’s nutrition department, publicly rejected the advice and kept their recommendations; subsequent reporting revealed undisclosed industry-linked funding ties for the lead author, prompting a correction to the paper’s disclosure statement. An older genre of null meta-analyses on meat and cancer, meanwhile, came from consultancy scientists with meat-industry groups among the funders [104]. None of this is new evidence for meat. It’s a re-grading exercise on the same evidence, and re-grading is not refutation.

Do studies show vegetarians are less healthy?

Exhibit three: the Austrian “unhealthy vegetarians” study. A 2014 cross-sectional survey found Austrian vegetarians self-reporting poorer health, more cancer, and more anxiety than meat-eaters [103]. Critics wave it constantly, and its ladder position settles it. Cross-sectional studies photograph a single moment, so they cannot say which came first, the diet or the disease, and reverse causation is the obvious reading, one the authors themselves flagged: people frequently adopt vegetarian diets after a cancer or digestive diagnosis, stacking the vegetarian category with the already-ill. Add a few hundred self-reporting participants and no diet-quality data, and you have a study sitting below even the cohorts this post spent pages stress-testing, a rung this post’s own side isn’t permitted to stand on either.

One recurring exhibit deserves a final sentence: the U-shaped curves showing low cholesterol associated with higher mortality in some observational data. Same diagnosis, serious illness lowers cholesterol before it kills, and Mendelian randomization, which is immune to that trick, shows lifelong lower LDL means lifelong lower cardiovascular risk [37].

Notice the pattern. Every counter-exhibit either sits on a lower rung than the evidence it’s deployed against, re-grades rather than refutes, or collapses under exactly the scrutiny this post applied to its own side’s Blue Zones. That’s the value of climbing the ladder in public: it flattens weak arguments no matter whose they are.

What the world’s leading health authorities actually say

What do major health organizations say about vegan diets?

If the science above were fringe, you’d expect the institutions charged with reading all of it to disagree. They don’t.

The Academy of Nutrition and Dietetics: the world’s largest organization of food and nutrition professionals, with over 100,000 credentialed practitioners, published its updated position paper in 2025, concluding that appropriately planned vegetarian and vegan dietary patterns in adults can be nutritionally adequate and can offer long-term health benefits, particularly for cardiometabolic disease, and directing dietitians to actively support people adopting them [48]. (Its landmark 2016 position went further across the lifespan, deeming well-planned vegan diets appropriate for all life stages including pregnancy, infancy, and athletics; the 2025 update simply scoped itself to adults, explicitly leaving other life stages to dedicated future papers, not, as some meat advocates spun it, reversing anything [48, 49].)

The American Heart Association’s 2021 dietary guidance is built around plant-forward principles: plenty of vegetables, fruits, whole grains, and legumes, protein preferentially from plants, and minimal processed meat [50]. Even more telling, the AHA’s 2023 scientific statement graded ten popular diets against its criteria: the plant-forward patterns swept the top tier, DASH a perfect 100, pescatarian 92, Mediterranean 89, and vegetarian 86; the vegan pattern scored 78 in the second tier, docked not for disease outcomes but for practical adherence challenges and the need to manage B12; and the meat-centric paleo (53) and ketogenic (31) patterns landed dead last, flagged as contradicting heart-healthy guidance outright [51, 52].

The American Cancer Society’s guideline calls for eating patterns high in vegetables, fruits, legumes, and whole grains while limiting or eliminating red and processed meats [53], echoing the World Cancer Research Fund/AICR, whose recommendations are to eat a diet rich in whole plant foods, limit red meat, and eat little if any processed meat [54], and the WHO’s IARC carcinogen classifications noted earlier [28]. The American Diabetes Association’s consensus report lists vegetarian and vegan eating patterns among evidence-based options for diabetes management, with trials showing improved glycemia and cardiovascular risk factors [55]. The American Academy of Pediatrics’ own nutrition handbook recognizes that well-planned vegetarian and vegan diets can support normal growth and development in children, with appropriate attention to nutrients like B12, vitamin D, iron, and zinc [56].

What do official nutrition guidelines recommend?

Even the bodies that set the numerical rules tilt the same way. The National Academies’ Dietary Reference Intakes put the Acceptable Macronutrient Distribution Range for carbohydrate at 45–65% of calories: the highest range of any macronutrient, and a range that whole-plant diets naturally satisfy, versus 10–35% for protein and 20–35% for fat [57]. The same DRIs set the fiber recommendation at 14 grams per 1,000 calories (fiber exists only in plants, and roughly 19 in 20 Americans fall short) and concluded that trans fat and dietary cholesterol have no safe intake level, recommending consumption as low as possible within a nutritionally adequate diet, a bar only plant foods clear by default, since cholesterol comes exclusively from animal foods [57].

Why did the 2025 Dietary Guidelines ignore its own science panel?

Then there’s the revealing saga of the U.S. Dietary Guidelines. The independent 2025 Dietary Guidelines Advisory Committee, twenty of the country’s leading nutrition scientists, spent two years reviewing the evidence and delivered its most plant-forward scientific report ever: move beans, peas, and lentils into the protein group and list them first, ahead of nuts, seeds, and soy, then seafood, and meat, poultry, and eggs last; increase legume intake; reduce meat, poultry, and egg quantities; and replace saturated fat specifically with plant sources, while stating there is compelling evidence tying higher red and processed meat intake to harm [58, 59]. What happened next says more about politics than science: the final 2025–2030 guidelines, released January 7, 2026 under new HHS leadership, largely set that report aside, publishing a streamlined “real food” guide built around an inverted pyramid that foregrounds protein, full-fat dairy, and “healthy fats,” explicitly naming beef tallow and butter, while retaining the decades-old cap on saturated fat at under 10% of calories, a retention that sits awkwardly beside its promotion of the very foods richest in it. Harvard’s nutrition department, the American Heart Association, the Physicians Committee, and sidelined committee member Christopher Gardner all objected that the guidelines contradict the evidence their own advisory panel had assembled [60–62]. When the science panel and the political document diverge, remember which one read the evidence.

Which countries recommend plant-based diets?

The international trend needs no such asterisk. Canada’s 2019 Food Guide tells citizens to choose plant protein more often and eliminated the dairy food group entirely [63]. Denmark’s official dietary guidelines urge eating plant-rich, with legumes daily and far less meat [64]. The Nordic Nutrition Recommendations 2023, the science basis for five nations’ guidelines, prescribe a predominantly plant-based dietary pattern [65]. Germany’s nutrition society now recommends a diet of roughly three-quarters plant foods [66]. And the EAT-Lancet Commission, synthesizing global evidence on diets that maximize human health, first in 2019, updated as EAT-Lancet 2.0 in October 2025, landed on a planetary health plate that is overwhelmingly plant-based, with animal foods optional and minimal; the 2025 update estimates that global adoption is associated with roughly 27% lower risk of premature death, on the order of 15 million lives a year [67, 108]. Kaiser Permanente, one of the largest healthcare organizations in the US, published guidance in its own journal advising physicians to consider recommending plant-based diets to all their patients [68].

An entire plant-based medical ecosystem, with no counterpart on the other side

Which medical organizations are fully plant-based?

Beyond position papers, an institutional infrastructure has grown up around plant-predominant and fully plant-based nutrition, and its scale is worth appreciating. The American College of Lifestyle Medicine, among the fastest-growing medical professional societies in the country, now board-certifying physicians in lifestyle medicine, states as its official dietary position that an eating plan based predominantly on a variety of minimally processed, whole plant foods should be the foundation for treating and reversing chronic disease [69]. The Physicians Committee for Responsible Medicine counts more than 17,000 physicians among nearly a million members and supporters advocating plant-based prevention [13]. The Ornish Program for Reversing Heart Disease, built on a whole-food plant-based diet, became the first program ever covered by Medicare under its Intensive Cardiac Rehabilitation benefit in 2010, after CMS reviewed the peer-reviewed evidence that it improves and reverses coronary disease; major private insurers followed [70, 71]. (The other originally approved ICR program, Pritikin, is likewise plant-centered [71].) TrueNorth Health Center runs the largest medically supervised fasting facility in the world on an exclusively whole-food plant-based protocol; the National Health Association has promoted plant-exclusive eating since 1948; NutritionFacts.org provides a free, non-commercial video library translating the primary literature; Hippocrates Wellness has run plant-exclusive residential programs for decades. Whatever one thinks of any single organization, together they represent physicians, dietitians, researchers, insurers, and federal payers converging on the same dietary direction.

Is there a carnivore-diet equivalent of plant-based medicine?

Now run the comparison that ends the debate: where is the equivalent ecosystem for the opposite diet? There is no “American College of Carnivore Medicine” board-certifying physicians. No position paper from any national medical, cardiology, oncology, diabetes, dietetic, or pediatric organization anywhere on Earth endorses meat-maximal eating. No Medicare-covered carnivore program exists, because no body of peer-reviewed trials demonstrates disease reversal to submit to CMS. No cohort of 96,000 meat-maximizers shows a longevity gradient favoring more animal foods: the gradients all run the other way. The closest mainstream evaluation such diets have received is the AHA scoring keto and paleo at the bottom of its rankings as contradicting heart-healthy guidance [51, 52]. In science, when one hypothesis accumulates cohorts, trials, mechanisms, genetics, institutions, and insurers, and the rival hypothesis accumulates podcasts: the asymmetry is the answer.

The bottom line

What does all the evidence add up to?

Strip away the noise and the picture is coherent from every rung of the ladder. Within the best-studied longevity population on Earth, health improves stepwise with every rung toward fully plant-based, and vegans sit at or near the top on weight, diabetes, blood pressure, cancer, and survival despite every methodological headwind [1, 19–22]. Randomized trials show whole-food plant-based eating doesn’t just correlate with less disease: it reverses it [30–34]. Identical twins randomized to vegan eating beat their own genetic duplicates on the biomarkers that cause heart disease and even on molecular aging itself [35, 36]. The genetics, the microbiome, the inflammation data, and the telomeres all point the same way [37–46]. Even the world’s centenarian hotspots, taken as color, not foundation, rhyme with the same roughly 95%-plant, bean-anchored pattern [13]. And the institutions whose job is to weigh everything, from the Academy of Nutrition and Dietetics to the AHA to the world’s national food guides to Medicare’s own actuaries, keep landing on the same side [48–71].

Lifespan will always keep some cards hidden: genes, accidents, luck. Healthspan is the hand you play yourself, and the strongest cards in it are the people you keep close and what you put on your plate. The first of those this post can only point at; the second it can measure, and no eating pattern has more converging evidence behind it than one built entirely from whole plant foods, vegetables, fruits, legumes, whole grains, nuts, seeds, mushrooms, herbs, and spices, done properly, which means a reliable B12 supplement (and attention to vitamin D and omega-3s) as non-negotiable basics rather than afterthoughts. One distinction worth carrying out the door: veganism names the ethic; whole-food plant-based names the diet. The quiet convenience of our moment is that they meet on the same plate, every food group at the core of the evidence-backed pattern is vegan, so living the ethic and optimizing the diet call for the same grocery list. And all of that is tallying only the health ledger: the ethical and environmental columns are taken up in the companion post, Rewilding the Plate. You don’t need a vegan country to exist before acting on this; the staircase is visible, every step up improves the view, and nothing at the top suggests the pattern breaks. The longest-living people already showed us the direction. The science just keeps confirming how far it goes.

A companion document, an evidence table sorting all 106 references by their rung on the ladder, accompanies this post.

Related reading: What the Adventist Health Studies Actually Tell Us About Vegan Diets, And What They Don’t is the full deep-dive on the cohort at the heart of this post’s gradient. Rewilding the Plate tallies the environmental ledger, and The Burden of Proof Has Shifted asks what follows once the evidence lands this way.

References

1. Orlich MJ, Singh PN, Sabaté J, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Internal Medicine. 2013;173(13):1230–1238.
2. Tong TYN, Appleby PN, Bradbury KE, et al. Risks of ischaemic heart disease and stroke in meat eaters, fish eaters, and vegetarians over 18 years of follow-up: results from the prospective EPIC-Oxford study. BMJ. 2019;366:l4897. Link
3. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and unhealthful plant-based diets and the risk of coronary heart disease in US adults. Journal of the American College of Cardiology. 2017;70(4):411–422.
4. Satija A, Bhupathiraju SN, Rimm EB, et al. Plant-based dietary patterns and incidence of type 2 diabetes in US men and women: results from three prospective cohort studies. PLoS Medicine. 2016;13(6):e1002039.
5. Pan A, Sun Q, Bernstein AM, et al. Red meat consumption and mortality: results from 2 prospective cohort studies. Archives of Internal Medicine. 2012;172(7):555–563.
6. Hill AB. The environment and disease: association or causation? Proceedings of the Royal Society of Medicine. 1965;58(5):295–300.
7. Ni MY, Canudas-Romo V, Shi J, et al. Understanding longevity in Hong Kong: a comparative study with long-living, high-income countries. The Lancet Public Health. 2021;6(12):e919–e931.
8. The University of Hong Kong. Tobacco control shown to be the key driver for Hong Kong’s world-leading longevity. HKUMed press release, 2021.
9. Woo J, et al. Non-communicable diseases, longevity, and health span: a Hong Kong perspective. Review, 2025 (PMC11942087).
10. Hong Kong Cancer Registry, Hospital Authority. Colorectal cancer statistics (leading cancer sites in Hong Kong).
11. Poulain M, Pes GM, Grasland C, et al. Identification of a geographic area characterized by extreme longevity in the Sardinia island: the AKEA study. Experimental Gerontology. 2004;39(9):1423–1429.
12. Buettner D, Skemp S. Blue Zones: lessons from the world’s longest lived. American Journal of Lifestyle Medicine. 2016;10(5):318–321.
13. Buettner D. The Blue Zones (analysis of 154 dietary surveys across the five zones); summarized in Institute of Medicine, Business Engagement in Building Healthy Communities. National Academies Press; 2015.
14. Willcox BJ, Willcox DC, Todoriki H, et al. Caloric restriction, the traditional Okinawan diet, and healthy aging. Annals of the New York Academy of Sciences. 2007;1114:434–455.
15. Fraser GE, Shavlik DJ. Ten years of life: is it a matter of choice? Archives of Internal Medicine. 2001;161(13):1645–1652.
16. Herskind AM, McGue M, Holm NV, et al. The heritability of human longevity: a population-based study of 2872 Danish twin pairs born 1870–1900. Human Genetics. 1996;97(3):319–323.
17. Ruby JG, Wright KM, Rand KA, et al. Estimates of the heritability of human longevity are substantially inflated due to assortative mating. Genetics. 2018;210(3):1109–1124.
18. GBD 2017 Diet Collaborators. Health effects of dietary risks in 195 countries, 1990–2017. The Lancet. 2019;393(10184):1958–1972.
19. Tonstad S, Butler T, Yan R, Fraser GE. Type of vegetarian diet, body weight, and prevalence of type 2 diabetes. Diabetes Care. 2009;32(5):791–796.
20. Matsumoto S, et al. Association between vegetarian diets and cardiovascular risk factors in non-Hispanic white participants of the Adventist Health Study-2. Journal of Nutritional Science. 2019;8:e6.
21. Tantamango-Bartley Y, Jaceldo-Siegl K, Fan J, Fraser G. Vegetarian diets and the incidence of cancer in a low-risk population. Cancer Epidemiology, Biomarkers & Prevention. 2013;22(2):286–294.
22. Fraser GE, Cosgrove CM, Mashchak AD, et al. Cause-specific and all-cause mortalities in vegetarian compared with those in nonvegetarian participants from the Adventist Health Study-2 cohort. The American Journal of Clinical Nutrition. 2024. Link
23. Key TJ, Fraser GE, Thorogood M, et al. Mortality in vegetarians and nonvegetarians: detailed findings from a collaborative analysis of 5 prospective studies. The American Journal of Clinical Nutrition. 1999;70(3 Suppl):516S–524S.
24. Vegan Health (Norris J). Mortality and cardiovascular disease rates of vegetarians: discussion of EPIC-Oxford stroke findings, B12, and cholesterol. veganhealth.org.
25. Zheng Y, Li Y, Satija A, et al. Association of changes in red meat consumption with total and cause specific mortality among US women and men. BMJ. 2019;365:l2110.
26. Song M, Fung TT, Hu FB, et al. Association of animal and plant protein intake with all-cause and cause-specific mortality. JAMA Internal Medicine. 2016;176(10):1453–1463.
27. Alshahrani SM, Fraser GE, Sabaté J, et al. Red and processed meat and mortality in a low meat intake population. Nutrients. 2019;11(3):622.
28. Bouvard V, Loomis D, Guyton KZ, et al. Carcinogenicity of consumption of red and processed meat. The Lancet Oncology. 2015;16(16):1599–1600.
29. Fries JF. Aging, natural death, and the compression of morbidity. New England Journal of Medicine. 1980;303(3):130–135.
30. Ornish D, Brown SE, Scherwitz LW, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. The Lancet. 1990;336(8708):129–133.
31. Ornish D, Scherwitz LW, Billings JH, et al. Intensive lifestyle changes for reversal of coronary heart disease. JAMA. 1998;280(23):2001–2007.
32. Esselstyn CB Jr, Gendy G, Doyle J, et al. A way to reverse CAD? Journal of Family Practice. 2014;63(7):356–364b.
33. Wright N, Wilson L, Smith M, et al. The BROAD study: a randomised controlled trial using a whole food plant-based diet in the community for obesity, ischaemic heart disease or diabetes. Nutrition & Diabetes. 2017;7(3):e256.
34. Barnard ND, Cohen J, Jenkins DJA, et al. A low-fat vegan diet improves glycemic control and cardiovascular risk factors in a randomized clinical trial in individuals with type 2 diabetes. Diabetes Care. 2006;29(8):1777–1783.
35. Landry MJ, Ward CP, Cunanan KM, et al. Cardiometabolic effects of omnivorous vs vegan diets in identical twins: a randomized clinical trial. JAMA Network Open. 2023;6(11):e2344457. Link
36. Dwaraka VB, Aronica L, Carreras-Gallo N, et al. Unveiling the epigenetic impact of vegan vs. omnivorous diets on aging: insights from the Twins Nutrition Study (TwiNS). BMC Medicine. 2024;22:301.
37. Ference BA, Ginsberg HN, Graham I, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. Evidence from genetic, epidemiologic, and clinical studies: a consensus statement from the European Atherosclerosis Society Consensus Panel. European Heart Journal. 2017;38(32):2459–2472.
38. Larsson SC, Carter P, Vithayathil S, et al. Insulin-like growth factor-1 and site-specific cancers: a Mendelian randomization study. Cancer Medicine. 2020;9(18):6836–6842.
39. Allen NE, Appleby PN, Davey GK, et al. The associations of diet with serum insulin-like growth factor I and its main binding proteins in 292 women meat-eaters, vegetarians, and vegans. Cancer Epidemiology, Biomarkers & Prevention. 2002;11(11):1441–1448.
40. Levine ME, Suarez JA, Brandhorst S, et al. Low protein intake is associated with a major reduction in IGF-1, cancer, and overall mortality in the 65 and younger but not older population. Cell Metabolism. 2014;19(3):407–417.
41. David LA, Maurice CF, Carmody RN, et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014;505(7484):559–563.
42. Koeth RA, Wang Z, Levison BS, et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nature Medicine. 2013;19(5):576–585.
43. Tang WHW, Wang Z, Levison BS, et al. Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk. New England Journal of Medicine. 2013;368(17):1575–1584.
44. Menzel J, Jabakhanji A, Biemann R, et al. Systematic review and meta-analysis of the associations of vegan and vegetarian diets with inflammatory biomarkers. Scientific Reports. 2020;10:21736.
45. Ornish D, Lin J, Daubenmier J, et al. Increased telomerase activity and comprehensive lifestyle changes: a pilot study. The Lancet Oncology. 2008;9(11):1048–1057.
46. Ornish D, Lin J, Chan JM, et al. Effect of comprehensive lifestyle changes on telomerase activity and telomere length in men with biopsy-proven low-risk prostate cancer: 5-year follow-up of a descriptive pilot study. The Lancet Oncology. 2013;14(11):1112–1120.
47. Samraj AN, Pearce OMT, Läubli H, et al. A red meat-derived glycan promotes inflammation and cancer progression. PNAS. 2015;112(2):542–547.
48. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian dietary patterns for adults: a position paper of the Academy of Nutrition and Dietetics. Journal of the Academy of Nutrition and Dietetics. 2025;125(6):831–846. Link
49. Melina V, Craig W, Levin S. Position of the Academy of Nutrition and Dietetics: vegetarian diets. Journal of the Academy of Nutrition and Dietetics. 2016;116(12):1970–1980.
50. Lichtenstein AH, Appel LJ, Vadiveloo M, et al. 2021 dietary guidance to improve cardiovascular health: a scientific statement from the American Heart Association. Circulation. 2021;144(23):e472–e487.
51. Gardner CD, Vadiveloo MK, Petersen KS, et al. Popular dietary patterns: alignment with American Heart Association 2021 dietary guidance: a scientific statement from the American Heart Association. Circulation. 2023;147(22):1715–1730.
52. American Heart Association. 10 popular diets scored for heart-healthy elements. AHA Newsroom, April 27, 2023. Link
53. Rock CL, Thomson C, Gansler T, et al. American Cancer Society guideline for diet and physical activity for cancer prevention. CA: A Cancer Journal for Clinicians. 2020;70(4):245–271.
54. World Cancer Research Fund/American Institute for Cancer Research. Diet, Nutrition, Physical Activity and Cancer: A Global Perspective. Third Expert Report. 2018.
55. Evert AB, Dennison M, Gardner CD, et al. Nutrition therapy for adults with diabetes or prediabetes: a consensus report. Diabetes Care. 2019;42(5):731–754.
56. Kleinman RE, Greer FR, eds. Pediatric Nutrition, 8th ed. American Academy of Pediatrics; 2019 (chapter on vegetarian and vegan diets in children).
57. Institute of Medicine (National Academies). Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. National Academies Press; 2005.
58. 2025 Dietary Guidelines Advisory Committee. Scientific Report of the 2025 Dietary Guidelines Advisory Committee. USDA/HHS; December 2024.
59. Physicians Committee for Responsible Medicine. Prioritizing plant-based protein in the Scientific Report of the 2025–2030 Dietary Guidelines is a step forward. December 2024.
60. U.S. Department of Health and Human Services and U.S. Department of Agriculture. Dietary Guidelines for Americans, 2025–2030. January 2026.
61. Harvard T.H. Chan School of Public Health, The Nutrition Source. Dietary Guidelines for Americans 2025–2030: progress on added sugar, protein hype, saturated fat contradictions. January 2026.
62. U.S. News & World Report. New Dietary Guidelines for Americans 2025–2030: reimagining the food pyramid (comments of DGAC member Christopher Gardner). January 2026.
63. Health Canada. Canada’s Food Guide. 2019.
64. Danish Veterinary and Food Administration. The Official Dietary Guidelines, Good for Health and Climate. 2021.
65. Blomhoff R, Andersen R, Arnesen EK, et al. Nordic Nutrition Recommendations 2023. Nordic Council of Ministers; 2023.
66. German Nutrition Society (DGE). Position statement on a more plant-based diet. 2024.
67. Willett W, Rockström J, Loken B, et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. The Lancet. 2019;393(10170):447–492. Link
68. Tuso PJ, Ismail MH, Ha BP, Bartolotto C. Nutritional update for physicians: plant-based diets. The Permanente Journal. 2013;17(2):61–66.
69. American College of Lifestyle Medicine. Dietary lifestyle position statement (whole-food, predominantly plant-based nutrition for treatment and reversal of chronic disease). 2022.
70. Centers for Medicare & Medicaid Services. National Coverage Determination 20.31.2: Ornish Program for Reversing Heart Disease (Intensive Cardiac Rehabilitation). Effective August 12, 2010.
71. Ornish Lifestyle Medicine. The Ornish Reversal Program: first program covered by Medicare under Intensive Cardiac Rehabilitation.
72. Yusuf S, Hawken S, Ôunpuu S, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. The Lancet. 2004;364(9438):937–952.
73. Appel LJ, Moore TJ, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. New England Journal of Medicine. 1997;336(16):1117–1124.
74. Micha R, Peñalvo JL, Cudhea F, et al. Association between dietary factors and mortality from heart disease, stroke, and type 2 diabetes in the United States. JAMA. 2017;317(9):912–924.
75. Micha R, Wallace SK, Mozaffarian D. Red and processed meat consumption and risk of incident coronary heart disease, stroke, and diabetes mellitus: a systematic review and meta-analysis. Circulation. 2010;121(21):2271–2283.
76. Pan A, Sun Q, Bernstein AM, et al. Red meat consumption and risk of type 2 diabetes: 3 cohorts of US adults and an updated meta-analysis. The American Journal of Clinical Nutrition. 2011;94(4):1088–1096.
77. Rohrmann S, Overvad K, Bueno-de-Mesquita HB, et al. Meat consumption and mortality, results from the European Prospective Investigation into Cancer and Nutrition. BMC Medicine. 2013;11:63.
78. Aune D, Navarro Rosenblatt DA, Chan DSM, et al. Dairy products, calcium, and prostate cancer risk: a systematic review and meta-analysis of cohort studies. The American Journal of Clinical Nutrition. 2015;101(1):87–117.
79. Richman EL, Kenfield SA, Stampfer MJ, et al. Egg, red meat, and poultry intake and risk of lethal prostate cancer in the prostate-specific antigen-era: incidence and survival. Cancer Prevention Research. 2011;4(12):2110–2121.
80. Keum N, Lee DH, Marchand N, et al. Egg intake and cancers of the breast, ovary and prostate: a dose-response meta-analysis of prospective observational studies. British Journal of Nutrition. 2015;114(7):1099–1107.
81. Fraser GE, Jaceldo-Siegl K, Orlich M, et al. Dairy, soy, and risk of breast cancer: those confounded milks. International Journal of Epidemiology. 2020;49(5):1526–1537. Link
82. Dinu M, Abbate R, Gensini GF, Casini A, Sofi F. Vegetarian, vegan diets and multiple health outcomes: a systematic review with meta-analysis of observational studies. Critical Reviews in Food Science and Nutrition. 2017;57(17):3640–3649.
83. The Vegan Society. Definition of veganism (established 1944). vegansociety.com.
84. Rosenfeld DL. The psychology of vegetarianism: recent advances and future directions. Appetite. 2018;131:125–138.
85. Rosenfeld DL, Tomiyama AJ. How proximal are pescatarians to vegetarians? An investigation of dietary identity, motivation, and attitudes toward animals. Journal of Health Psychology. 2021;26:713–727.
86. Hopwood CJ, Bleidorn W, Schwaba T, Chen S. Health, environmental, and animal rights motives for vegetarian eating. PLOS ONE. 2020;15(4):e0230609.
87. Rosenfeld DL. Ethical motivation and vegetarian dieting: the underlying role of anti-speciesist attitudes. Anthrozoös. 2019;32(6):785–796.
88. Newman SJ. Supercentenarian and remarkable age records exhibit patterns indicative of clerical errors and pension fraud. bioRxiv preprint (first posted 2019; awarded the 2024 Ig Nobel Prize in Demography).
89. Darmadi-Blackberry I, Wahlqvist ML, Kouris-Blazos A, et al. Legumes: the most important dietary predictor of survival in older people of different ethnicities. Asia Pacific Journal of Clinical Nutrition. 2004;13(2):217–220.
90. Bao Y, Han J, Hu FB, et al. Association of nut consumption with total and cause-specific mortality. New England Journal of Medicine. 2013;369(21):2001–2011.
91. Aune D, Keum N, Giovannucci E, et al. Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: systematic review and dose-response meta-analysis of prospective studies. BMJ. 2016;353:i2716.
92. Aune D, Giovannucci E, Boffetta P, et al. Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality, a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology. 2017;46(3):1029–1056.
93. Reynolds A, Mann J, Cummings J, et al. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. The Lancet. 2019;393(10170):434–445.
94. Morris MC, Wang Y, Barnes LL, et al. Nutrients and bioactives in green leafy vegetables and cognitive decline. Neurology. 2018;90(3):e214–e222.
95. Ba DM, Gao X, Al-Shaar L, et al. Prospective study of dietary mushroom intake and risk of mortality: results from continuous NHANES 2003–2014 and a meta-analysis. Nutrition Journal. 2021;20:80.
96. Murai U, Yamagishi K, Sata M, et al. Seaweed intake and risk of cardiovascular disease: the Japan Public Health Center–based Prospective (JPHC) Study. The American Journal of Clinical Nutrition. 2019;110(6):1449–1455.
97. Petersen KS, Davis KM, Rogers CJ, Proctor DN, West SG, Kris-Etherton PM. Herbs and spices at a relatively high culinary dosage improves 24-hour ambulatory blood pressure in adults at risk of cardiometabolic diseases: a randomized, crossover, controlled-feeding study. The American Journal of Clinical Nutrition. 2021;114(6):1936–1948.
98. Jenkins DJA, Kendall CWC, Marchie A, et al. Effects of a dietary portfolio of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive protein. JAMA. 2003;290(4):502–510.
99. Estruch R, Ros E, Salas-Salvadó J, et al. Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. New England Journal of Medicine. 2018;378(25):e34. Link
100. Dehghan M, Mente A, Zhang X, et al. Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries from five continents (PURE): a prospective cohort study. The Lancet. 2017;390(10107):2050–2062.
101. Miller V, Mente A, Dehghan M, et al. Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study. The Lancet. 2017;390(10107):2037–2049.
102. Johnston BC, Zeraatkar D, Han MA, et al. Unprocessed red meat and processed meat consumption: dietary guideline recommendations from the Nutritional Recommendations (NutriRECS) Consortium. Annals of Internal Medicine. 2019;171(10):756–764.
103. Burkert NT, Muckenhuber J, Großschädl F, Rásky É, Freidl W. Nutrition and health: the association between eating behavior and various health parameters: a matched sample study. PLOS ONE. 2014;9(2):e88278.
104. Alexander DD, Mink PJ, Cushing CA, Sceurman B. A review and meta-analysis of prospective studies of red and processed meat intake and prostate cancer. Nutrition Journal. 2010;9:50.
105. Waldinger RJ. “What makes a good life? Lessons from the longest study on happiness.” TED, 2015. https://www.youtube.com/watch?v=8KkKuTCFvzI, reporting the Harvard Study of Adult Development, ongoing since 1938. See also Waldinger RJ, Schulz MS, The Good Life: Lessons from the World’s Longest Scientific Study of Happiness (Simon & Schuster, 2023). Link
106. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Medicine. 2010;7(7):e1000316.
107. Austad SN, Pes GM. The validity of Blue Zones demography: a response to critiques. The Gerontologist. 2025;65(12):gnaf246. See also Rosero-Bixby L. The exceptional rural longevity of Nicoya, Costa Rica, and its later-cohort attenuation. Demographic Research. 2023;49:723–736; Saito Y, Yong V, Robine J-M. The mystery of Japan’s missing centenarians explained. Demographic Research. 2012;26:239–252. Link
108. EAT-Lancet Commission 2.0. Food, planet, health: the EAT-Lancet Commission on healthy, sustainable, and just food systems. The Lancet. 2025 (estimated ~27% lower premature mortality with adoption of the planetary health diet).
109. Landry MJ, Ward CP, Koh LM, et al. Vegetarian and vegan dietary patterns and cardiovascular disease and mortality: an umbrella review of systematic reviews (commissioned by the Academy of Nutrition and Dietetics). 2024.
110. Capodici A, Fantini MP, et al. Cardiovascular and metabolic benefits of plant-based diets: an umbrella review of 48 meta-analyses. PLOS ONE. 2024;19(5):e0300711.

Read more...

Short version: almost any diet that takes weight off will make you feel better and improve your lab numbers for a while, because weight loss itself flatters nearly every biomarker. That is exactly why weight loss alone is a terrible test of a diet. A whole-food, plant-based (WFPB) way of eating is the rare approach where the same mechanism that removes the weight, swapping calorie-dense, fiber-free animal and processed foods for fiber- and water-rich whole plants, is also the mechanism that lowers cholesterol, improves insulin sensitivity, and tracks with less heart disease, diabetes, and premature death. It lets most people eat until full without counting a single calorie, it has real evidence at every rung of the scientific pyramid, and it is endorsed by mainstream health bodies. That is the opposite of a fad.

This post makes that case in full. It walks the entire hierarchy of evidence, from mechanism up through randomized trials and decades-long cohorts, compares WFPB honestly against Ozempic, keto, surgery, and even the famous Twinkie diet, sets out what the major health organizations conclude, and ends with a concrete plan and an honest accounting of where the evidence is softer than the headline. Every study named below was checked against its primary source.

Part 1. A nation that is both heavy and sick

How many American adults are overweight or obese?

Start with the uncomfortable numbers. During August 2021 to August 2023, an estimated 40.3 percent of U.S. adults age 20 and older had obesity and 9.4 percent had severe obesity (a BMI of 40 or higher), per the CDC’s National Center for Health Statistics (Emmerich, Fryar, Stierman, Ogden; NCHS Data Brief No. 508, 2024). Another 31.7 percent were overweight (Fryar, Afful, Saif; NCHS Health E-Stat, 2024). Add those together and a clear majority of American adults now carry more weight than is healthy.

Does losing weight actually improve your health?

Yes, and this is the part almost everyone gets right. Shedding excess fat tends to lower blood pressure, steady blood sugar, improve cholesterol, ease joints, and lift energy. The part almost everyone gets wrong is the assumption that because weight loss is good, the method does not much matter. It matters enormously, because most methods are either hard to sustain, carry their own risks, or improve your weight while doing little (or worse) for the disease processes you cannot feel. The rest of this article is about that gap.

Part 2. Why the weight comes off almost by itself

Why is fat more fattening, gram for gram, than carbs or protein?

Because of simple arithmetic your body cannot argue with. Fat carries 9 calories per gram, while protein and carbohydrate carry about 4, and water carries zero. Every whole animal food contains fat, including saturated fat, and contains no fiber at all. (One precision note, because getting it exactly right matters: it is a myth that animal foods contain zero carbohydrate, since dairy carries the sugar lactose, but they contain no fiber whatsoever. Fiber comes only from plants.) Whole plant foods flip the ratio: they are built largely from fiber, water, and starch rather than concentrated fat. How the two food kingdoms package their fat so differently is the whole subject of a companion piece, Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries.

What is calorie density, and why does it matter for weight loss?

Calorie density is the number of calories packed into each gram of food, and it may be the single most useful lever for eating fewer calories without feeling deprived. Barbara Rolls and colleagues at Penn State have spent decades documenting it: because water and fiber add weight and volume with few or no calories, water-rich, high-fiber foods let you eat larger, more satisfying portions for fewer calories. In her controlled studies, lowering the energy density of the diet caused people to spontaneously eat less, and in one trial, women counseled simply to add water-rich foods lost more weight while eating a greater volume of food than women told to restrict fat and portions (Rolls, Nutrition Bulletin, 2017).

Can adding food to your diet make you lose weight?

Strange as it sounds, yes, if the food is dilute enough. In one trial, overweight women told to add three apples or three pears a day to their usual diet lost weight, about two pounds, while a comparison group given oat cookies with similar fiber did not; the fruit’s advantage was its low calorie density (de Oliveira, Sichieri, Mozzer; Appetite, 2008). The classic demonstration is even starker: served diets low versus high in energy density and told to eat to satiety, 20 obese and nonobese subjects reached fullness at a mean of 1,570 calories a day on the low-density diet versus 3,000 on the high-density diet, with equal acceptance ratings for both (Duncan, Bacon, Weinsier; American Journal of Clinical Nutrition, 1983). Half the calories, same satisfaction. And in Hawaii, 20 Native Hawaiians placed on a traditional pre-Western-contact diet (7 percent fat, 78 percent complex carbohydrate) and encouraged to eat to satiety dropped their intake from about 2,594 to 1,569 calories a day and lost an average of 7.8 kg, about 17 pounds, in just 21 days, with cholesterol and blood pressure falling too; it was a single-arm study without a control group, so hold the exact number gently, but the direction matches everything else on this page (Shintani, Hughes, Beckham, O’Connor; American Journal of Clinical Nutrition, 1991). A systematic review commissioned for U.S. dietary guidance concluded that the evidence supports a relationship between energy density and body weight in adults and in children and adolescents (Pérez-Escamilla and colleagues; Journal of the Academy of Nutrition and Dietetics, 2012), and the CDC’s own practitioner guidance recommends low-energy-density fruits and vegetables for exactly this reason.

Can you lose weight without counting calories on a plant-based diet?

For most people, yes, and that is the whole point. Put the arithmetic and the calorie-density research together and you can eat ad libitum, meaning until comfortably full, and still run a calorie deficit, because the food does the accounting for you. That is the difference between white-knuckle dieting and a way of eating that feels liberating. There is even a fourth, subtler mechanism: whole plant foods appear to slightly raise your calorie burn after meals, which we will meet in the trials below.

Two classic lines of evidence round out the picture. In the satiety-index experiments that fed people dozens of common foods and tracked hunger for hours afterward, plain boiled potatoes topped the chart and croissants came in last, and fullness per calorie tracked the physical weight and water of the serving while running opposite to its fat content (Holt, Miller, Petocz, Farmakalidis; European Journal of Clinical Nutrition, 1995). Fiber quietly stacks the deck further: a classic review estimated that adding about 14 grams of fiber a day was associated with roughly a 10 percent drop in calorie intake and around four pounds of weight loss over a few months (Howarth, Saltzman, Roberts; Nutrition Reviews, 2001), partly because fiber fills you up and partly because a share of fiber-trapped calories is never absorbed at all.

Part of that fullness is hormonal, which is the honest version of the “nature’s Ozempic” idea now in the air. In a randomized crossover trial, a tofu burger matched calorie-for-calorie and gram-for-gram to a processed-meat-and-cheese burger triggered substantially more of the satiety hormone GLP-1: about 30 percent more in men with type 2 diabetes and 16 percent more in healthy men (Klementova and colleagues; Nutrients, 2019). GLP-1 is the same pathway the weight-loss drugs act on. The honest caveat is that this trial measured hormones and fullness for a few hours, not pounds lost over months, so read it as a mechanism, not a miracle: fiber-rich whole foods nudge the satiety system in the same direction the drugs shove it.

Part 3. Meet the organ you are shrinking

What do fat cells actually do?

Body fat is not inert padding; it is a working endocrine organ. Fat cells manufacture and release hormones, called adipokines, that talk directly to your brain, liver, muscles, and immune system. The most famous is leptin, discovered when researchers cloned the “obese” gene in mice (Zhang, Proenca, Maffei, and colleagues; Nature, 1994). Leptin reports your fat stores to the brain, but as fat mass climbs, the brain grows resistant to the signal, so the “we have plenty stored” message stops landing. A second messenger, adiponectin, improves insulin sensitivity and calms inflammation, and it falls as fat mass rises, which is exactly backwards from what you would want. There is a darker channel too: enlarged fat tissue overproduces the inflammatory molecule TNF-alpha, which directly blunts insulin signaling (Hotamisligil, Shargill, Spiegelman; Science, 1993), and immune cells called macrophages move in as fat cells swell, approaching 40 percent of all the cells in the fat tissue of people with obesity (Weisberg and colleagues; Journal of Clinical Investigation, 2003). Low-grade inflammation from overloaded fat is a large part of how excess weight drives disease.

One more surprise, courtesy of Cold War physics: by carbon-dating fat cells using the radiocarbon left in the atmosphere by nuclear-bomb tests, researchers showed that your total number of fat cells is largely set by the end of adolescence and stays roughly constant through adult life, with about 10 percent of them replaced each year, even after marked weight loss (Spalding and colleagues; Nature, 2008). Losing weight mainly shrinks fat cells rather than deleting them. That is not discouraging; it is simply how the tissue works, and shrunken, calm fat cells behave far better than swollen, inflamed ones.

Why is belly fat more dangerous than hip fat?

Location matters more than the number on the scale. Fat under the skin of the hips and thighs is comparatively harmless and may even be protective. Fat packed deep inside the abdomen around the organs, called visceral fat, is the troublemaker: it is more inflamed, more metabolically active, and it drains straight into the liver. In the Framingham Heart Study’s imaging analyses, visceral fat was far more strongly tied to blood sugar, blood pressure, and unfavorable cholesterol than subcutaneous fat was (Fox and colleagues; Circulation, 2007).

This also solves a puzzle: some slim people are metabolically sick while some heavier people are not. Taylor and Holman proposed the “personal fat threshold”: each person can safely store only so much fat under the skin before the surplus spills into the liver, the muscles, and the belly and starts driving type 2 diabetes (Clinical Science, 2015). Cross your threshold and metabolism suffers, whatever your BMI; drop back under it and things often reverse. Keep that in mind when we reach the trials in the next section that drained fat out of liver and muscle cells: that is precisely the fat doing the damage.

Does losing weight release toxins into your body?

This deserves a straight answer, because the internet gets it wrong in both directions. Certain long-lived industrial pollutants, the persistent organic pollutants or POPs (PCBs, dioxins, and old pesticides like DDT), are fat-soluble and sit stored in adipose tissue for years. When you burn fat, some of that load re-enters the blood. This is well documented: in 39 obese adults who lost an average of 9.5 kg over 15 weeks, every one of the 19 organochlorine compounds detected rose in plasma (Chevrier and colleagues; International Journal of Obesity, 2000).

Does that make weight loss dangerous? No. There is no credible evidence that this rise causes clinical harm, and the health benefits of losing excess weight overwhelmingly dominate. The one nuance worth knowing: in a small study of 16 dieting men, the pollutant rise correlated with a dip in thyroid hormone and resting metabolic rate, and the authors themselves stressed that causation is unproven (Pelletier, Doucet, Imbeault, Tremblay; Toxicological Sciences, 2002). Treat it as a footnote, not a warning label. (Women who are pregnant or breastfeeding and considering rapid weight loss do have a genuine reason to loop in their clinician, since these compounds can transfer to an infant.)

Here is where this feeds the post’s thesis: the smart response is not to fear losing fat, it is to accumulate less of this cargo in the first place. These pollutants concentrate up the food chain, and the World Health Organization estimates that over 90 percent of human dioxin exposure arrives through food, chiefly meat, dairy, fish, and shellfish. Eat lower on the food chain and you simply carry less. The human data, while thin, point the same way: in a small pilot, nine vegans had significantly lower plasma levels of five PCB compounds than fifteen omnivores, with an overall trend toward less contamination (Arguin and colleagues; British Journal of Nutrition, 2010), and a classic early-1980s analysis found dramatically lower pollutant levels in vegetarian mothers’ breast milk (Hergenrather and colleagues; New England Journal of Medicine, 1981), though background exposures for everyone have fallen since that era. Fewer pollutants in means fewer to mobilize on the way out.

Is a vegan’s body fat different?

In several measurable ways, yes. First, plant-based eating preferentially attacks the dangerous depots. The 16-week vegan trial you will meet in the next section reduced visceral fat along with body weight, and cut fat inside liver cells by roughly a third and inside muscle cells by roughly a tenth (Kahleova and colleagues; JAMA Network Open, 2020); liver and muscle fat are prime drivers of insulin resistance. A cross-sectional study likewise found vegans carrying less muscle-cell fat and better insulin sensitivity than omnivores of the same body size (Goff and colleagues; European Journal of Clinical Nutrition, 2005). And in Stanford’s identical-twin trial, where genetics could not be the explanation, the twin randomized to a healthy vegan diet lost about 1.9 kg more in eight weeks than the twin eating a healthy omnivorous diet (Landry and colleagues; JAMA Network Open, 2023).

Second, the very chemistry of fat reflects the diet that built it. Adipose fatty acids turn over slowly, so your fat acts as a chemical diary of what you have eaten over roughly the past year or two (Hodson, Skeaff, Fielding; Progress in Lipid Research, 2008), and long-term vegans store more plant-derived linoleic acid and less saturated fat in their adipose tissue than omnivores do (Sanders, Ellis, Dickerson; American Journal of Clinical Nutrition, 1978). Whether that compositional difference changes how fat cells behave is still being worked out, so file it under promising rather than proven. The pattern across this whole section, though, is consistent: a whole-food plant-based diet does not just move the scale; it steers fat away from the organs, lightens the inflammatory and pollutant cargo the tissue carries, and leaves behind a calmer organ.

Part 4. Climbing the entire pyramid of evidence

What is the hierarchy of scientific evidence?

Scientists rank evidence like a pyramid. At the base sit mechanisms and small physiology studies (useful for why, weaker for proof). Above them sit prospective cohort studies, large groups followed for years, which are excellent for spotting real-world patterns but cannot fully prove cause and effect. Higher still are randomized controlled trials (RCTs), where people are randomly assigned to diets so researchers can isolate cause. At the very top are systematic reviews and meta-analyses, which pool many studies into one best estimate. A fad diet typically has a tiny, wobbly pyramid: a couple of short studies and a bestselling book. WFPB eating has a full one. Let us climb it.

Do meta-analyses show plant-based diets cause weight loss?

They do. A meta-analysis of controlled trials found that prescribing vegetarian diets produced a mean weight loss of 3.4 kg (about 7.5 lb) in intention-to-treat analysis and 4.6 kg (about 10 lb) among completers, notably even in studies that did not restrict calories (Barnard, Levin, Yokoyama; Journal of the Academy of Nutrition and Dietetics, 2015). A separate meta-analysis of RCTs confirmed greater weight loss on vegetarian diets than on comparison diets (Huang, Huang, Hu, Chavarro; Journal of General Internal Medicine, 2016). A more recent systematic review of vegan-diet RCTs in people with overweight or type 2 diabetes (11 trials, 796 participants) found meaningful improvements in body weight and cardiometabolic markers (Termannsen and colleagues; Obesity Reviews, 2022). On cholesterol, pooling 49 studies, plant-based diets lowered total cholesterol by about 12.5 mg/dL and LDL by about 12.2 mg/dL in clinical trials, with even larger gaps in observational data (Yokoyama, Levin, Barnard; Nutrition Reviews, 2017). And pooling long-term cohorts, vegetarians had a 25 percent lower risk of ischemic-heart-disease incidence or mortality (RR 0.75) and vegans a 15 percent lower total-cancer incidence (RR 0.85), though, in the interest of honesty, that same analysis found no statistically significant reduction in all-cause mortality (Dinu, Abbate, Gensini, Casini, Sofi; Critical Reviews in Food Science and Nutrition, 2017). The cardiovascular and cancer signals are on firmer footing than “vegans live longer.” The molecular half of this cholesterol story is told in Cholesterol, From the Ground Up.

What did the randomized controlled trials find?

This is the tier built to prove cause and effect. In the BROAD study, a New Zealand RCT, participants followed a non-calorie-restricted WFPB diet; at 6 months their mean BMI fell 4.4 points versus 0.4 in usual care (a 3.9-point difference), with roughly 12 kg of weight loss and a 4.2-point BMI reduction still in place at 12 months (Wright, Wilson, Smith, Duncan, McHugh; Nutrition and Diabetes, 2017). Worth savoring: the facilitated classes ran for only the first 12 weeks, so the results at 6 and 12 months were held largely on the participants’ own steam, which is precisely the sustainability test most diets fail. It was a small trial (65 people randomized), but the authors noted this was greater weight loss at 6 and 12 months than any prior trial that neither restricted calories nor mandated exercise.

In a 16-week RCT of 244 overweight adults, a low-fat vegan group lost about 5.9 kg while the control group barely moved, and the trial illuminated the mechanism: the thermic effect of food (post-meal calorie burn) rose about 14 percent, insulin sensitivity improved, and fat inside liver and muscle cells, the fat that drives insulin resistance, fell substantially (Kahleova, Petersen, Shulman, and colleagues; JAMA Network Open, 2020). In a head-to-head diabetes trial, a low-fat vegan diet beat the standard American Diabetes Association diet on blood sugar, weight, and cholesterol (Barnard and colleagues; Diabetes Care, 2006). And in a multicenter workplace RCT across corporate sites, employees on a low-fat vegan diet lost meaningful weight and improved cholesterol and HbA1c versus no-change controls (Mishra and colleagues, the GEICO study; European Journal of Clinical Nutrition, 2013).

Does a vegan diet beat other diets head-to-head?

When you line the diets up side by side, the fully plant-based end wins. In a six-month trial that randomized overweight adults to five eating patterns along a plant-to-animal gradient, the vegan group lost the most: 7.5 percent of body weight versus about 3 percent for the omnivore, semi-vegetarian, and pesco-vegetarian groups (Turner-McGrievy, Davidson, Wingard, Wilcox, Frongillo; Nutrition, 2015). And the edge holds up over time: an earlier two-year trial found vegan dieters kept off more weight than a conventional low-fat diet at both one year (about 4.9 versus 1.8 kg) and two years (about 3.1 versus 0.9 kg), without anyone counting calories (Turner-McGrievy, Barnard, Scialli; Obesity, 2007). Both trials were small, so treat the exact numbers as estimates, but the direction is consistent with everything above.

The NIH ward study: did people really eat less on a plant-based diet?

This is one of the most rigorous diet experiments ever run. Twenty adults lived on a metabolic ward where every calorie was measured, eating ad libitum either a minimally processed plant-based, low-fat diet or a minimally processed animal-based, ketogenic diet, then crossing over to the other (Hall and colleagues; Nature Medicine, 2021). On the plant-based diet, people spontaneously ate about 689 fewer calories per day, with no difference in reported hunger or meal enjoyment. Same fullness, same satisfaction, hundreds of fewer calories, driven by what was on the plate. One honest caveat, which I will state rather than let a critic catch: over these two weeks the ketogenic arm actually showed slightly more scale-weight loss (about 1.8 kg versus 1.1 kg), largely water and glycogen, so this trial is decisive proof that the plant-based pattern lowers spontaneous calorie intake, not proof that it wins the scale in a fortnight.

The same NIH lab had already shown the flip side of the coin. When people were fed ultra-processed versus unprocessed diets matched for presented calories, salt, sugar, fat, and fiber, they spontaneously ate about 500 more calories a day on the ultra-processed diet and gained about two pounds in two weeks, while on the unprocessed diet they lost about two pounds (Hall and colleagues; Cell Metabolism, 2019). Put the two experiments together and the prescription writes itself: plant, and whole. Each word is doing work.

Do vegans really have a lower BMI, over decades and across huge populations?

The Adventist Health Study-2 produced one of the cleanest figures in nutrition science: a stepwise BMI gradient by diet. Mean BMI was lowest in vegans (23.6), rising through lacto-ovo vegetarians (25.7), pesco-vegetarians (26.3), semi-vegetarians (27.3), and non-vegetarians (28.8), with type 2 diabetes prevalence climbing the same ladder from 2.9 to 7.6 percent (Tonstad, Butler, Yan, Fraser; Diabetes Care, 2009). On mortality in the same cohort, vegetarians overall had lower all-cause mortality than non-vegetarians (hazard ratio 0.88), with effects stronger in men; for vegans specifically the hazard ratio was 0.85 but crossed statistical significance, so it is suggestive rather than proven (Orlich and colleagues; JAMA Internal Medicine, 2013). And across more than 200,000 people in the Harvard cohorts, a diet built on whole plant foods predicted lower coronary disease, while a “plant-based” diet built on refined grains, sweets, and sugary drinks predicted higher risk (Satija and colleagues; Journal of the American College of Cardiology, 2017). Hold that last finding; it matters when we discuss vegan junk food.

Three more cohort threads are worth a line each. The pattern is not new: researchers were already reporting vegetarians’ lower body weights and far lower blood lipids in the New England Journal of Medicine back in 1975 (Sacks, Castelli, Donner, Kass; NEJM, 1975). It runs in the fattening direction too: in the EPIC-PANACEA analysis of hundreds of thousands of Europeans, higher meat intake, poultry included, predicted greater weight gain over the following years, even at the same calorie intake (Vergnaud and colleagues; American Journal of Clinical Nutrition, 2010). And on mortality, a U.S. national cohort found a healthful plant-based diet index associated with lower all-cause mortality (Kim, Caulfield, Rebholz; Journal of Nutrition, 2018), index-based evidence that complements, without replacing, the vegan-specific numbers above.

What long-lived populations eat, kept in proportion

Several of the world’s longest-lived populations eat diets built largely on beans, greens, whole grains, and other plants, with meat as a garnish. This is suggestive rather than conclusive, since these communities differ from the rest of us in many ways beyond diet, so treat it as hypothesis-generating support that rhymes with everything above, not as proof on its own.

Part 5. Yes, other approaches also cause weight loss. Here is why WFPB still wins.

Do Ozempic and other GLP-1 drugs work for weight loss?

They work, and powerfully. In the STEP 1 trial, semaglutide produced a mean weight change of about minus 14.9 percent at 68 weeks versus minus 2.4 percent for placebo (Wilding and colleagues; New England Journal of Medicine, 2021). But they are expensive, often cause nausea and other GI effects, can cost muscle mass, and are a lifelong prescription rather than a cure: in the STEP 1 extension, after stopping the drug, participants regained roughly two-thirds of their lost weight and their cardiometabolic gains drifted back toward baseline, leaving them about 5.6 percent below their starting weight (Wilding and colleagues; Diabetes, Obesity and Metabolism, 2022). There is also a body-composition wrinkle: a substantial share of the weight lost on GLP-1 drugs is lean tissue rather than fat, a recognized concern that has clinicians now pairing the drugs with protein and resistance training. A diet that raises GLP-1 on its own, the way whole plant foods appear to (see the Klementova trial above), sidesteps both the rebound and the muscle penalty.

Does keto, Atkins, or carnivore work, and what is the catch?

All of them reliably drop early weight, much of it water at first. The catch shows up in the numbers you cannot feel. Meta-analyses of ketogenic-diet trials find that LDL cholesterol tends to rise on average, by roughly 8 mg/dL (Bueno and colleagues; British Journal of Nutrition, 2013), and in a controlled feeding trial a ketogenic diet raised LDL in every participant tested (Buren, Ericsson, Damasceno, Sjodin; Nutrients, 2021). In fairness, and because I would rather name the counter-argument than have it thrown at me, the LDL response varies by body type: it climbs most dramatically in lean people and is more muted, occasionally even lowered, in some people with obesity. The scale itself can also mislead: in an NIH ward study, when overweight men were switched to a calorie-matched ketogenic diet, scale weight dropped quickly but the rate of body-fat loss actually slowed, because the early plunge was mostly water (Hall and colleagues; American Journal of Clinical Nutrition, 2016). But long-term adherence to these diets is poor, and when low-carb dieters are followed for decades, animal-based low-carb patterns track with higher mortality while plant-based low-carb patterns track with lower mortality. Same macros, opposite kingdoms, opposite outcomes.

The Twinkie diet: doesn’t junk food prove calories are all that matter?

In 2010, Mark Haub, a human-nutrition professor at Kansas State, ate mostly Twinkies, Oreos, and chips while capping calories at about 1,800 a day (down from roughly 2,600). Over 10 weeks he lost 27 pounds, his BMI dropped from 28.8 (overweight) to 24.9 (normal), his LDL fell 20 percent, his HDL rose 20 percent, his triglycerides dropped 39 percent, and his total cholesterol went from 214 to 184 (Park, CNN, 2010). It is the perfect illustration of the trap: weight loss improves biomarkers regardless of food quality, which is precisely why weight loss alone is a terrible measure of whether you are actually getting healthier. Haub, to his credit, said the same and did not recommend anyone repeat it.

If you feel better and look better, are you actually healthier?

Not necessarily, and this is the trap worth understanding. Losing weight almost always makes you feel more energetic, and it flatters the numbers you can see: the scale, your waistband, often your fasting glucose and blood pressure. But feeling better and looking better are not proof that everything under the hood improved. Some of the most dangerous changes are the ones you cannot feel at all. You do not feel your insulin sensitivity drift, you do not feel inflammation smoldering, and you certainly do not feel plaque building in a coronary artery. A diet can move the visible markers in the right direction while quietly moving a hidden one the wrong way. The clearest documented example is the LDL and ApoB rise that many people experience on very-low-carbohydrate diets even as the scale falls, exactly the keto pattern above: a worsening of the single most causal driver of heart disease that produces no symptoms whatsoever.

Coronary plaque is the starkest case. The plaque most likely to rupture and cause a heart attack is not the hard, calcified kind; it is soft, non-calcified plaque, and in the multicenter SCOT-HEART imaging study it was the strongest predictor of future heart attack, outperforming standard risk scores (Williams and colleagues; Circulation, 2020). It accumulates silently, over years, whatever the scale says. Imaging research has begun to raise concern that some carbohydrate-restricted dietary patterns may be associated with more of this high-risk plaque, though that specific evidence is early and genuinely contested, so treat it as a caution rather than a verdict. The safe conclusion is the humble one: judge a way of eating by what it does to the markers you cannot feel, measured over years, not by how you look in a month. For more on what LDL actually does inside an artery wall, and why you cannot feel any of it happening, see Cholesterol, From the Ground Up.

So why is a whole-food plant-based diet the best choice?

Here is the decisive difference. With most of these approaches, the weight-loss method and the health-building method are different things, and sometimes at odds (keto and rising LDL being the clearest example). With WFPB eating they are the same thing. The mechanism that takes the weight off, swapping calorie-dense, fiber-free animal and processed foods for fiber- and water-rich whole plants, is the identical mechanism that lowers cholesterol, improves insulin sensitivity, and is associated with less heart disease and cancer. You are not trading your long-term health for short-term thinness. And on the factor that trials show matters most for weight loss, adherence, vegan-diet trials repeatedly find participants rate the diets as satisfying and acceptable, with the BROAD and GEICO studies documenting good adherence and better quality of life.

Part 6. Why most diets fail and a well-planned vegan way of eating lasts

Why do most diets work at first and then stop working?

Most weight-loss plans ask you to eat less of the same food. Count calories, shrink portions, resist. That reliably works for a while. In a meta-analysis of 80 trials with at least a year of follow-up, dieters lost about 5 to 9 percent of their weight by six months, then plateaued and slowly regained (Franz and colleagues; Journal of the American Dietetic Association, 2007). A widely cited review put it bluntly: across long-term follow-ups, most dieters regain most or all of what they lost, and a meaningful share end up heavier than when they started (Mann and colleagues; American Psychologist, 2007). That review has real limitations (many of its follow-ups relied on self-reported weight and had high dropout), but its central point, that restriction is hard to sustain, is not seriously disputed.

The reason is not weak willpower. When you lose weight by restricting quantity, your body pushes back on two fronts. It burns fewer calories than your smaller size predicts: six years after “The Biggest Loser,” contestants still had resting metabolic rates roughly 500 calories a day below what their body composition predicted (Fothergill and colleagues; Obesity, 2016). And it turns up hunger: after diet-induced weight loss, the appetite hormones that drive you to eat stay shifted toward hunger, with leptin still about a third below baseline a full year out (Sumithran and colleagues; New England Journal of Medicine, 2011). A modeling analysis estimated that each kilogram lost pushes appetite up by roughly 100 calories a day, an effect several times larger than the metabolic slowdown (Polidori and colleagues; Obesity, 2016). Restriction, in other words, sets up a fight against your own biology that you have to win every single day, forever. Almost nobody does.

Why does a well-planned vegan diet last when other diets do not?

The escape is to change food quality rather than police food quantity. Whole plant foods are low in calorie density; they carry so much water and fiber that they fill you up for far fewer calories, so you can eat to fullness and still run an energy deficit without a calculator. This is measurable. In a year-long trial with no calorie or portion targets, women simply told to add water-rich fruits and vegetables ate a greater weight of food, reported less hunger, and lost about a third more weight (8.9 versus 6.7 kg) than women told only to cut fat (Ello-Martin, Roe, Ledikwe, Beach, Rolls; American Journal of Clinical Nutrition, 2007). You are working with your appetite instead of against it.

Plant-forward, eat-to-fullness patterns also hold up head to head. In the two-year vegan trial we met earlier, where both groups ate until full with no calorie limits, the vegan group kept off more weight at one and two years than a conventional low-fat diet (Turner-McGrievy, Barnard, Scialli; Obesity, 2007). And in the 24-month NEW Soul study, people assigned to a vegan pattern actually stuck with it better than those assigned an omnivorous diet, with higher diet quality at one year (Hu and colleagues; Nutrition Research, 2024). The honest caveat: these trials are modest in size, and direct multi-year comparisons of WFPB against calorie restriction remain few, so I am leaning on the mechanism plus what maintenance data we have, not overclaiming a settled result.

Here is the framing that matters, and it is an honest one. Every optimal way of eating, Mediterranean and DASH included, requires some planning and explanation to be its best version; that is not a mark against a well-planned vegan diet, it is simply true of them all. A well-planned whole-food plant-based way of eating is simple in practice: mostly vegetables, fruits, whole grains, beans, nuts, and seeds, eaten to fullness. (The detail on how to plan it well, protein, B12, and the rest, lives in the nutrition guide.) It lasts not because it eliminates nothing, but because it replaces restriction-of-quantity with improvement-of-quality, so staying the course does not depend on fighting hunger. No diet guarantees permanent loss, and plant-based eaters can regain too. But when the strategy stops asking you to white-knuckle against your own appetite, the odds of keeping the weight off, for years, tilt in your favor, and the payoff extends well past the scale.

Part 7. Why a vegan diet is the opposite of a fad diet

What makes a diet a fad diet?

Nutrition scientists point to a recognizable profile: severe or arbitrary restriction, promises of fast and effortless results, a thin and transient evidence base, poor long-term adherence, and claims that contradict the major health bodies. Hold WFPB eating up to that mirror and it fails every criterion. The evidence base spans decades and every tier of the pyramid, mainstream bodies endorse it rather than warn against it, and the adherence data are encouraging rather than dismal.

Isn’t going vegan just another elimination diet?

This is the counterintuitive part. A vegan diet does remove animal products, but in practice it expands your plate rather than shrinking it. Botanists count roughly 7,000 plant species that humans have used as food, against a modern Western plate where a handful of crops supply most of our calories. Across legumes, leafy greens, whole grains, fruits, vegetables, mushrooms, nuts, seeds, sea vegetables, herbs, and spices, the combinations are effectively limitless. Most people arrive eating a tight rotation of chicken, beef, cheese, and bread and discover lentils, chickpeas, farro, tempeh, jackfruit, a dozen leafy greens, and spice traditions from every cuisine on Earth. A fad narrows your world; a well-planned vegan diet widens it, and lets you tailor it to your culture, budget, and taste.

Part 8. What leading health organizations actually say

What does the Academy of Nutrition and Dietetics say about vegan diets?

The Academy, the largest body of nutrition professionals in the world, holds as its current position, approved in January 2025 and in effect through 2032, that appropriately planned vegetarian and vegan dietary patterns are nutritionally adequate for adults and can offer long-term cardiometabolic health benefits (Raj, Guest, Landry, Mangels, Pawlak, Rozga; Journal of the Academy of Nutrition and Dietetics, 2025). Its previous position went further, judging well-planned vegan diets appropriate for every stage of life, including pregnancy, infancy, childhood, and athletics (Melina, Craig, Levin; 2016); the 2025 update simply scoped its formal review to adults, the population its new systematic reviews covered.

Does Kaiser Permanente recommend plant-based diets?

One of the largest healthcare systems in the United States published guidance in its own physician journal urging doctors to consider recommending plant-based diets to patients as low-risk, cost-effective interventions that may lower BMI, blood pressure, HbA1c, and cholesterol (Tuso, Ismail, Ha, Bartolotto; The Permanente Journal, 2013).

Can a plant-based diet put type 2 diabetes into remission?

The American College of Lifestyle Medicine holds that sufficiently intensive lifestyle intervention, particularly a predominantly whole-food, plant-based dietary pattern, can achieve remission of type 2 diabetes with success comparable to bariatric surgery but with substantially fewer side effects (Kelly, Karlsen, Steinke; American Journal of Lifestyle Medicine, 2020).

What does the American Heart Association say?

The AHA’s dietary guidance emphasizes overall patterns over single nutrients and recommends choosing healthy protein sources mostly from plants, plus plenty and variety of fruits and vegetables and liquid plant oils rather than tropical or animal fats (Lichtenstein and colleagues; Circulation, 2021). The British Dietetic Association and Dietitians of Canada likewise affirm that well-planned vegan diets are suitable across the lifespan. None of these are fringe organizations.

Part 9. What if the transition is hard?

What about B12 and other nutrients on a vegan diet?

It would be dishonest to pretend everyone finds the switch easy overnight. There is a learning curve of new ingredients, new cooking rhythms, and a few nutrients to be intentional about. The big one is vitamin B12, and it is genuinely not a flaw in the diet, just a fact of modern sanitized life; a cheap, reliable supplement solves it completely, and, tellingly, farmed animals are themselves routinely supplemented with B12 or cobalt, so meat-eaters are often getting a supplement with extra steps. A little attention to vitamin D, omega-3s, iodine, and getting enough legumes rounds out the picture. This is bookkeeping, not rocket science, and it gets its own deep dive in The Great B12 Gamble.

How do you handle social pressure and stick with it?

Family dinners, restaurant menus, holidays, and the occasional contrarian influencer are real, and they are very solvable, especially with the right guidance. The best move if you want both results and respect for your values is to work with a physician, registered dietitian, or nutritionist who practices lifestyle medicine, since lifestyle medicine is evidence-based and specifically oriented around using food as primary therapy. You get a professional who will help you hit your weight and health goals and honor your reasons for eating this way, rather than talking you out of them.

What to actually do

None of this requires perfection on day one. A staged approach works better than a cold-turnkey overhaul, and it gives you benchmarks to tell whether it is working.

1. Weeks 1 to 2, crowd in rather than only cut out. Add a large serving of beans or lentils and a big pile of greens or vegetables to meals you already eat, to lower your plate’s calorie density and get used to feeling full on more food. Start a daily B12 supplement now. One trick with trial support: open meals with a low-calorie-density first course. In Penn State experiments, a small low-calorie salad reduced the total calories eaten at the meal instead of adding to them (Rolls, Roe, Meengs; Journal of the American Dietetic Association, 2004), soup preloads did the same (Flood, Rolls; Appetite, 2007), and a whole apple before lunch cut intake more than the same fruit as applesauce or juice, one more point for intact plants (Flood-Obbagy, Rolls; Appetite, 2009). Even two cups of water before each meal helped: middle-aged and older adults randomized to premeal water on a reduced-calorie diet lost about 2 kg more, a 44 percent greater decline in weight over 12 weeks (Dennis and colleagues; Obesity, 2010). Sign that you are on track: you feel satisfied at meals without hunger between them.
2. Weeks 3 to 6, rebuild the staples. Convert breakfast and lunch fully to whole-plant meals (oats and fruit; grain bowls with beans, vegetables, and a tahini or salsa sauce), and learn five to eight go-to recipes across different cuisines so that variety, not willpower, drives adherence. Sign that you are on track: the scale trends down about 0.5 to 1.5 lb per week without calorie counting.
3. Front-load your plate, and do not graze all day. When and how you distribute the same calories matters at the margins. In a 12-week trial of 93 women with metabolic syndrome, all eating about 1,400 calories a day, those who put the big meal at breakfast lost roughly 8.7 kg versus about 3.6 kg for the big-dinner group, with lower hunger hormones (Jakubowicz, Barnea, Wainstein, Froy; Obesity, 2013). A separate crossover trial in people with type 2 diabetes found two larger meals (breakfast and lunch) beat six small ones of the same total calories for losing weight and liver fat (Kahleova and colleagues; Diabetologia, 2014), which quietly retires the old “graze to stoke your metabolism” advice. Both trials relied partly on self-reported intake, so treat front-loading as a low-risk edge, not a guarantee.
4. Protect your sleep, or you will lose the wrong tissue. In a tightly controlled crossover trial, 10 dieters lost the same total weight on 8.5 versus 5.5 hours of sleep, but short sleep cut the share of weight lost as fat by 55 percent (1.4 versus 0.6 kg) and increased muscle loss by 60 percent (Nedeltcheva, Kilkus, Imperial, Schoeller, Penev; Annals of Internal Medicine, 2010). It is a small study, but the crossover design (each person as their own control) makes it hard to wave away: the scale can move while your body burns the wrong fuel.
5. Weeks 6 to 12, go fully WFPB and get baseline labs. Ask your clinician for weight and BMI, fasting glucose or HbA1c, and a lipid panel, then recheck at about 12 weeks. What good looks like: LDL cholesterol and, if elevated, HbA1c falling; waist shrinking; medications under review with your doctor.
6. Watch two thresholds. If weight is not moving after roughly 6 to 8 weeks of genuine WFPB eating, audit calorie density first, since refined vegan junk food, added oils, and liquid calories are the usual culprits (this is exactly the unhealthful plant-based pattern that Satija flagged). And if you take diabetes or blood-pressure medication, work with your prescriber before you start, because doses often need to come down quickly to avoid hypoglycemia or low blood pressure.
7. Get a lifestyle-medicine professional in your corner if you have an existing chronic condition, are pregnant, or want your cultural food traditions built into the plan rather than erased.

Where I’m keeping the claims honest

• Cohorts show association, not proof. The Adventist and Harvard cohorts cannot fully separate diet from the fact that these groups also tend to smoke less, exercise more, and drink less. Their value is that they line up with the RCTs and the mechanisms, which they do.
• The mortality claims deserve precision. Dinu 2017 found lower ischemic-heart-disease and cancer incidence but no statistically significant all-cause mortality reduction, and the Adventist vegan hazard ratio (0.85) crossed significance on its own. The weight, cholesterol, diabetes, and heart-disease findings are firmer than “vegans live longer.”
• Several of the strongest vegan RCTs are advocacy-affiliated. The Barnard, Kahleova, and GEICO trials come from researchers at the Physicians Committee for Responsible Medicine. Their results are corroborated by independent work (the NIH ward study, the Adventist cohorts, Rolls at Penn State, and the Termannsen and Dinu meta-analyses), which is why the overall case holds, but it is worth naming.
• The trials are mostly short and some are small. Most vegan RCTs run 12 to 74 weeks, BROAD randomized only 65 people, and the Kahleova trial was 16 weeks and mostly female. Very-long-term randomized data are scarce for essentially every diet, so the multi-decade picture leans on cohorts.
• “Plant-based” is not automatically healthy. French fries and soda are vegan. The evidence favors whole plant foods; ultra-processed vegan products do not earn the same benefits.
• The NIH ward study is about intake, not scale weight. Over its two weeks the keto arm lost slightly more scale weight, largely water. Its real finding is the spontaneous drop in calories eaten on the plant-based diet.
• Organizations endorse the approach; they do not order everyone to go vegan. The AHA and Academy back well-planned plant-forward and vegan patterns as legitimate and healthful. That is an endorsement of the approach, not a claim that every body has issued a “go vegan” mandate.
• The fat-biology section leans on famous, replicated science for its load-bearing claims, but its vegan-specific comparisons (plasma pollutants, muscle-cell fat, adipose fatty acids) rest on small or older cross-sectional studies (nine vegans in the Arguin pilot; fatty-acid data from the 1970s) and are stated directionally on purpose. The pollutant rise during weight loss is well documented; its clinical significance is not, and losing excess weight remains overwhelmingly net-beneficial.
• The diet-and-coronary-plaque link is an active, contested area. That non-calcified plaque is dangerous and silent is well established (SCOT-HEART). Whether specific carbohydrate-restricted diets accelerate it is still debated, with the most-discussed imaging study lacking a control group. I raise it as a caution about markers you cannot feel, not as a settled verdict.

Further reading

• Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries, on why the two food kingdoms package fat so differently and why the Twinkie result proves what it proves.
• Cholesterol, From the Ground Up, on what LDL actually does in an artery wall and why you cannot feel it happening.
• The Great B12 Gamble, the full, evidence-hierarchy case for the one supplement a well-planned vegan diet requires.
• Saturated Fat, Seed Oils, and Heart Disease, on the “seed oils are poison” narrative and what replacing saturated fat with plant oils actually does.
• The nutrition hub, for the broader plant-based nutrition picture.

References

1. Emmerich SD, Fryar CD, Stierman B, Ogden CL. Obesity and severe obesity prevalence in adults: United States, August 2021 to August 2023. NCHS Data Brief No. 508. National Center for Health Statistics; 2024.
2. Fryar CD, Afful J, Saif NT. Prevalence of overweight, obesity, and severe obesity among adults age 20 and older: United States, 1960 to 1962 through August 2021 to August 2023. NCHS Health E-Stat. National Center for Health Statistics; 2024.
3. Rolls BJ. Dietary energy density: applying behavioural science to weight management. Nutr Bull. 2017;42(3):246-253.
4. Rolls BJ, Roe LS, Meengs JS. Salad and satiety: energy density and portion size of a first course salad affect energy intake at lunch. J Am Diet Assoc. 2004;104(10):1570-1576.
5. Flood JE, Rolls BJ. Soup preloads in a variety of forms reduce meal energy intake. Appetite. 2007;49(3):626-634.
6. Flood-Obbagy JE, Rolls BJ. The effect of fruit in different forms on energy intake and satiety at a meal. Appetite. 2009;52(2):416-422.
7. Dennis EA, Dengo AL, Comber DL, et al. Water consumption increases weight loss during a hypocaloric diet intervention in middle-aged and older adults. Obesity (Silver Spring). 2010;18(2):300-307.
8. Holt SH, Miller JC, Petocz P, Farmakalidis E. A satiety index of common foods. Eur J Clin Nutr. 1995;49(9):675-690.
9. Howarth NC, Saltzman E, Roberts SB. Dietary fiber and weight regulation. Nutr Rev. 2001;59(5):129-139.
10. Pérez-Escamilla R, Obbagy JE, Altman JM, Essery EV, McGrane MM, Wong YP, Spahn JM, Williams CL. Dietary energy density and body weight in adults and children: a systematic review. J Acad Nutr Diet. 2012;112(5):671-684.
11. de Oliveira MC, Sichieri R, Venturim Mozzer R. A low-energy-dense diet adding fruit reduces weight and energy intake in women. Appetite. 2008;51(2):291-295.
12. Duncan KH, Bacon JA, Weinsier RL. The effects of high and low energy density diets on satiety, energy intake, and eating time of obese and nonobese subjects. Am J Clin Nutr. 1983;37(5):763-767.
13. Shintani TT, Hughes CK, Beckham S, O’Connor HK. Obesity and cardiovascular risk intervention through the ad libitum feeding of traditional Hawaiian diet. Am J Clin Nutr. 1991;53(6 Suppl):1647S-1651S.
14. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature. 1994;372(6505):425-432.
15. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science. 1993;259(5091):87-91.
16. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003;112(12):1796-1808.
17. Spalding KL, Arner E, Westermark PO, et al. Dynamics of fat cell turnover in humans. Nature. 2008;453(7196):783-787.
18. Fox CS, Massaro JM, Hoffmann U, et al. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116(1):39-48.
19. Taylor R, Holman RR. Normal weight individuals who develop type 2 diabetes: the personal fat threshold. Clin Sci (Lond). 2015;128(7):405-410.
20. Chevrier J, Dewailly E, Ayotte P, Mauriege P, Despres JP, Tremblay A. Body weight loss increases plasma and adipose tissue concentrations of potentially toxic pollutants in obese individuals. Int J Obes Relat Metab Disord. 2000;24(10):1272-1278.
21. Pelletier C, Doucet E, Imbeault P, Tremblay A. Associations between weight loss-induced changes in plasma organochlorine concentrations, serum T3 concentration, and resting metabolic rate. Toxicol Sci. 2002;67(1):46-51.
22. World Health Organization. Dioxins and their effects on human health. WHO fact sheet.
23. Arguin H, Sanchez M, Bray GA, et al. Impact of adopting a vegan diet or an olestra supplementation on plasma organochlorine concentrations: results from two pilot studies. Br J Nutr. 2010;103(10):1433-1441.
24. Hergenrather J, Hlady G, Wallace B, Savage E. Pollutants in breast milk of vegetarians. N Engl J Med. 1981;304(13):792.
25. Landry MJ, Ward CP, Cunanan KM, et al. Cardiometabolic effects of omnivorous vs vegan diets in identical twins: a randomized clinical trial. JAMA Netw Open. 2023;6(11):e2344457.
26. Goff LM, Bell JD, So PW, Dornhorst A, Frost GS. Veganism and its relationship with insulin resistance and intramyocellular lipid. Eur J Clin Nutr. 2005;59(2):291-298.
27. Sanders TA, Ellis FR, Dickerson JW. Studies of vegans: the fatty acid composition of plasma choline phosphoglycerides, erythrocytes, adipose tissue, and breast milk, and some indicators of susceptibility to ischaemic heart disease in vegans and omnivore controls. Am J Clin Nutr. 1978;31(5):805-813.
28. Hodson L, Skeaff CM, Fielding BA. Fatty acid composition of adipose tissue and blood in humans and its use as a biomarker of dietary intake. Prog Lipid Res. 2008;47(5):348-380.
29. Barnard ND, Levin SM, Yokoyama Y. A systematic review and meta-analysis of changes in body weight in clinical trials of vegetarian diets. J Acad Nutr Diet. 2015;115(6):954-969.
30. Huang RY, Huang CC, Hu FB, Chavarro JE. Vegetarian diets and weight reduction: a meta-analysis of randomized controlled trials. J Gen Intern Med. 2016;31(1):109-116.
31. Termannsen AD, Clemmensen KKB, Thomsen JM, Norgaard O, Diaz LJ, Torekov SS, Quist JS, Faerch K. Effects of vegan diets on cardiometabolic health: a systematic review and meta-analysis of randomized controlled trials. Obes Rev. 2022;23(9):e13462.
32. Yokoyama Y, Levin SM, Barnard ND. Association between plant-based diets and plasma lipids: a systematic review and meta-analysis. Nutr Rev. 2017;75(9):683-698.
33. Dinu M, Abbate R, Gensini GF, Casini A, Sofi F. Vegetarian, vegan diets and multiple health outcomes: a systematic review with meta-analysis of observational studies. Crit Rev Food Sci Nutr. 2017;57(17):3640-3649.
34. Wright N, Wilson L, Smith M, Duncan B, McHugh P. The BROAD study: a randomised controlled trial using a whole food plant-based diet in the community for obesity, ischaemic heart disease or diabetes. Nutr Diabetes. 2017;7(3):e256.
35. Kahleova H, Petersen KF, Shulman GI, Alwarith J, Rembert E, Tura A, Hill M, Holubkov R, Barnard ND. Effect of a low-fat vegan diet on body weight, insulin sensitivity, postprandial metabolism, and intramyocellular and hepatocellular lipid levels in overweight adults: a randomized clinical trial. JAMA Netw Open. 2020;3(11):e2025454.
36. Barnard ND, Cohen J, Jenkins DJA, Turner-McGrievy G, Gloede L, Jaster B, Seidl K, Green AA, Talpers S. A low-fat vegan diet improves glycemic control and cardiovascular risk factors in a randomized clinical trial in individuals with type 2 diabetes. Diabetes Care. 2006;29(8):1777-1783.
37. Mishra S, Xu J, Agarwal U, Gonzales J, Levin S, Barnard ND. A multicenter randomized controlled trial of a plant-based nutrition program to reduce body weight and cardiovascular risk in the corporate setting: the GEICO study. Eur J Clin Nutr. 2013;67(7):718-724.
38. Turner-McGrievy GM, Davidson CR, Wingard EE, Wilcox S, Frongillo EA. Comparative effectiveness of plant-based diets for weight loss: a randomized controlled trial of five different diets. Nutrition. 2015;31(2):350-358.
39. Turner-McGrievy GM, Barnard ND, Scialli AR. A two-year randomized weight loss trial comparing a vegan diet to a more moderate low-fat diet. Obesity (Silver Spring). 2007;15(9):2276-2281.
40. Franz MJ, VanWormer JJ, Crain AL, et al. Weight-loss outcomes: a systematic review and meta-analysis of weight-loss clinical trials with a minimum 1-year follow-up. J Am Diet Assoc. 2007;107(10):1755-1767.
41. Mann T, Tomiyama AJ, Westling E, Lew AM, Samuels B, Chatman J. Medicare’s search for effective obesity treatments: diets are not the answer. Am Psychol. 2007;62(3):220-233.
42. Fothergill E, Guo J, Howard L, et al. Persistent metabolic adaptation 6 years after “The Biggest Loser” competition. Obesity (Silver Spring). 2016;24(8):1612-1619.
43. Sumithran P, Prendergast LA, Delbridge E, et al. Long-term persistence of hormonal adaptations to weight loss. N Engl J Med. 2011;365(17):1597-1604.
44. Polidori D, Sanghvi A, Seeley RJ, Hall KD. How strongly does appetite counter weight loss? Quantification of the feedback control of human energy intake. Obesity (Silver Spring). 2016;24(11):2289-2295.
45. Ello-Martin JA, Roe LS, Ledikwe JH, Beach AM, Rolls BJ. Dietary energy density in the treatment of obesity: a year-long trial comparing 2 weight-loss diets. Am J Clin Nutr. 2007;85(6):1465-1477.
46. Hu EA, Turner-McGrievy GM, Wilson MJ, et al. Adherence to a culturally adapted soul food vegan diet among African American adults increases diet quality compared to an omnivorous diet in the NEW Soul Study. Nutr Res. 2024;128:1-13.
47. Klementova M, Thieme L, Haluzik M, Pavlovicova R, Hill M, Pelikanova T, Kahleova H. A plant-based meal increases gastrointestinal hormones and satiety more than an energy- and macronutrient-matched processed-meat meal in T2D, obese, and healthy men: a three-group randomized crossover study. Nutrients. 2019;11(1):157.
48. Hall KD, Guo J, Courville AB, et al. Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake. Nat Med. 2021;27(2):344-353.
49. Hall KD, Ayuketah A, Brychta R, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metab. 2019;30(1):67-77.
50. Tonstad S, Butler T, Yan R, Fraser GE. Type of vegetarian diet, body weight, and prevalence of type 2 diabetes. Diabetes Care. 2009;32(5):791-796.
51. Orlich MJ, Singh PN, Sabate J, Jaceldo-Siegl K, Fan J, Knutsen S, Beeson WL, Fraser GE. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Intern Med. 2013;173(13):1230-1238.
52. Satija A, Bhupathiraju SN, Spiegelman D, et al. Healthful and unhealthful plant-based diets and the risk of coronary heart disease in US adults. J Am Coll Cardiol. 2017;70(4):411-422.
53. Sacks FM, Castelli WP, Donner A, Kass EH. Plasma lipids and lipoproteins in vegetarians and controls. N Engl J Med. 1975;292(22):1148-1151.
54. Vergnaud AC, Norat T, Romaguera D, et al. Meat consumption and prospective weight change in participants of the EPIC-PANACEA study. Am J Clin Nutr. 2010;92(2):398-407.
55. Kim H, Caulfield LE, Rebholz CM. Healthy plant-based diets are associated with lower risk of all-cause mortality in US adults. J Nutr. 2018;148(4):624-631.
56. Wilding JPH, Batterham RL, Calanna S, et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). N Engl J Med. 2021;384(11):989-1002.
57. Wilding JPH, Batterham RL, Davies M, et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: the STEP 1 trial extension. Diabetes Obes Metab. 2022;24(8):1553-1564.
58. Jakubowicz D, Barnea M, Wainstein J, Froy O. High caloric intake at breakfast vs. dinner differentially influences weight loss of overweight and obese women. Obesity (Silver Spring). 2013;21(12):2504-2512.
59. Kahleova H, Belinova L, Malinska H, et al. Eating two larger meals a day (breakfast and lunch) is more effective than six smaller meals in a reduced-energy regimen for patients with type 2 diabetes: a randomised crossover study. Diabetologia. 2014;57(8):1552-1560.
60. Nedeltcheva AV, Kilkus JM, Imperial J, Schoeller DA, Penev PD. Insufficient sleep undermines dietary efforts to reduce adiposity. Ann Intern Med. 2010;153(7):435-441.
61. Bueno NB, de Melo ISV, de Oliveira SL, da Rocha Ataide T. Very-low-carbohydrate ketogenic diet v. low-fat diet for long-term weight loss: a meta-analysis of randomised controlled trials. Br J Nutr. 2013;110(7):1178-1187.
62. Buren J, Ericsson M, Damasceno NRT, Sjodin A. A ketogenic low-carbohydrate high-fat diet increases LDL cholesterol in healthy, young, normal-weight women: a randomized controlled feeding trial. Nutrients. 2021;13(3):814.
63. Hall KD, Chen KY, Guo J, et al. Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men. Am J Clin Nutr. 2016;104(2):324-333.
64. Williams MC, Kwiecinski J, Doris M, et al. Low-attenuation noncalcified plaque on coronary computed tomography angiography predicts myocardial infarction: results from the multicenter SCOT-HEART trial. Circulation. 2020;141(18):1452-1462.
65. Park M. Twinkie diet helps nutrition professor lose 27 pounds. CNN. November 8, 2010.
66. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian dietary patterns for adults: a position paper of the Academy of Nutrition and Dietetics. J Acad Nutr Diet. 2025.
67. Melina V, Craig W, Levin S. Position of the Academy of Nutrition and Dietetics: vegetarian diets. J Acad Nutr Diet. 2016;116(12):1970-1980.
68. Tuso PJ, Ismail MH, Ha BP, Bartolotto C. Nutritional update for physicians: plant-based diets. Perm J. 2013;17(2):61-66.
69. Kelly J, Karlsen M, Steinke G. Type 2 diabetes remission and lifestyle medicine: a position statement from the American College of Lifestyle Medicine. Am J Lifestyle Med. 2020;14(4):406-419.
70. Lichtenstein AH, Appel LJ, Vadiveloo M, et al. 2021 dietary guidance to improve cardiovascular health: a scientific statement from the American Heart Association. Circulation. 2021;144(23):e472-e487.

Read more...

The regenerative movement’s answer to every environmental critique is “it’s not the cow, it’s the how.” So let’s take that seriously, steelman the strongest version, then test it against the arithmetic, the peer-reviewed literature, and a 270,000-acre experiment in the Oregon desert.

The environmental defense of animal agriculture has evolved. Few people still argue that factory farms are good for the planet. The sophisticated position now is the regenerative one: it’s not the cow, it’s the how. Managed correctly, the story goes, moved in dense herds that mimic ancient bison, resting the land between rotations, cattle heal soil, sequester carbon, restore grasslands, and make beef part of the climate solution. It’s an appealing story, told by charismatic ranchers on beautiful land, and it deserves what every claim in this project gets: a fair hearing, followed by arithmetic.

So let’s do the arithmetic first, then hear from the land itself.

The arithmetic nobody disputes

How much land does animal agriculture use?

Start with the single most important set of numbers in food and environment, from the largest meta-analysis of global food systems ever conducted, Joseph Poore and Thomas Nemecek’s review of 38,700 farms across 119 countries, published in Science [1]. Agriculture occupies about half of all habitable land on Earth. Of that agricultural land, roughly 83%, pasture plus the cropland growing animal feed, serves meat and dairy production. And what does that 83% of the land deliver? About 18% of humanity’s calories and 37% of its protein [1, 2]. (That 83% is Poore and Nemecek’s own accounting, and the figure used throughout this series; Our World in Data expresses the same underlying data as “more than three-quarters.”) It takes on the order of 100 times more land to produce a calorie or a gram of protein from beef or lamb than from plants [2]. Run that comparison on a single plot and the gap is just as stark, two acres of good Carolina pasture against two acres of crops is its own post: Two Acres: What Beef Produces vs. What Plants Can.

Why is eating animals so much less efficient than eating plants?

That land gap is not an accident of how we farm; it is a consequence of physics, and it cannot be optimized away. Plants, working with soil microbes, are the base producers of the entire food supply: through photosynthesis they turn sunlight into the sugars, and with their microbial partners the amino acids, vitamins, and even the B12-producing bacteria, on which every animal ultimately depends. Animals do not create these nutrients. They concentrate and repackage what plants and microbes already made, and they charge a steep toll to do it. The second law of thermodynamics guarantees the toll exists: no energy transfer is ever free, and some fraction always disperses as unusable heat. Ecologists reduced the cost to a rule of thumb, roughly ten percent of the energy at one level survives into the next [47]. A cow eats the plant and spends most of those calories simply being a cow, respiring, walking, holding body temperature, growing hide and bone and organs nobody will eat. Measured directly, beef returns about 3% of the calories fed to cattle and about 3% of the protein, against a 7 to 8% average across all animal categories, making beef by far the least efficient [48].

Two points of precision, because this is the load-bearing claim of the whole land debate. That ten percent is an average, not a constant: Lindeman, who established it in 1942, measured efficiencies anywhere from a fraction of a percent up into the high thirties, and modern feed-conversion numbers vary the same way by species and system [47]. And the second law does not fix the exact size of the loss; most of it is respiration, movement, and inedible tissue rather than entropy in the abstract. What the second law does guarantee is the part that matters here: the transfer can never be free. Adaptive multi-paddock grazing does not repeal it. Holistic management does not repeal it. Better genetics do not repeal it. You can breed a faster-growing cow, but you cannot repeal the energy loss between trophic levels any more than you can build a perpetual-motion machine. The debate can only ever be about how large the tax is, never whether it is levied, and the 83% of farmland and the roughly hundredfold land penalty are simply that tax, scaled to continents.

How much land would we save if everyone went vegan?

Now run the scenario the regenerative advocates least want to discuss. If the world ate fully plant-based, total agricultural land use would fall from about 4.1 billion hectares to about 1 billion, a 75% reduction, an area roughly the size of North America and Brazil combined [2]. Every hectare of grazing land, nearly three-quarters of agriculture’s entire footprint, comes back. And cropland shrinks too, because the enormous acreage currently growing feed crops disappears faster than human-food cropland grows to replace the calories [2]. That is the fact behind the claim you may have heard: go vegan and you return essentially all of the world’s pasture to the wild, plus a meaningful slice of its cropland.

Isn’t most grazing land unsuitable for crops anyway?

Here the regenerative side raises its favorite objection: most grazing land is marginal, so you can’t grow crops there! Correct, about two-thirds of grazing land isn’t suitable for cropping [3]. Its companion claim, that 86% of livestock feed is inedible to humans, comes from an FAO analysis whose remaining figures rarely get quoted: the same study reports livestock eating about a third of global cereal production and consuming roughly 3 kg of human-edible feed per kg of meat produced [3]. The full breakdown of that statistic is its own post, The 86% Meme, Debunked. But notice what the marginal-land objection assumes: that the only alternative use of land is farming it. The freed land doesn’t need to grow crops. It needs to grow ecosystems.

What a billion freed hectares is actually worth

How much carbon could rewilded farmland absorb?

In 2021, researchers at NYU, Harvard, Colorado State, and Oregon State published in Nature Sustainability what they called the “carbon opportunity cost” of animal-sourced food: the carbon that the land under pasture and feed crops would pull from the sky if allowed to return to native vegetation [4]. The numbers are staggering. A global shift to plant-based diets by 2050 would allow the regrowing forests, grasslands, and wetlands to sequester an estimated 547 billion tonnes of CO2, roughly the past sixteen years of global fossil fuel emissions, with ecosystem soils potentially adding a couple hundred billion tonnes more [4]. Even the partial shift recommended by the EAT-Lancet Commission would sequester about 332 billion tonnes, nine years of fossil emissions [4]. Put another way: dietary change alone could recapture 99–163% of the entire remaining carbon budget for holding warming to 1.5°C, an amount of carbon removal that mainstream climate models otherwise assign to speculative machine-capture technology that barely exists [4]. And that’s before counting the biodiversity dividend: agricultural expansion is the leading driver of deforestation and extinction worldwide, with beef the single largest commodity driver of tropical forest loss, a debt this post settles in its own section below [2, 5]. Remember one of that carbon paper’s co-authors, William Ripple of Oregon State. He’ll reappear shortly, standing in a creek bed with a camera.

So the regenerative claim isn’t competing against factory farming. It’s competing against that, against returning a North-America-plus-Brazil-sized landmass to forests, wetlands, and wild grasslands. Keep that bar in mind as we examine the claim on its merits.

The rainforest the regenerative story would rather not mention

Is Amazon deforestation actually caused by beef?

Steelman it first: the regenerative advocate will say the Amazon is a governance failure, not a beef failure, that well-managed ranching need not fell a single standing tree, and that Brazil’s own satellites now show deforestation falling. Both halves of that are true. Under President Lula, Brazil’s INPE recorded Legal Amazon clearing of 5,796 square kilometers for the year to July 2025, down 11% and the third-lowest reading since monitoring began in 1988, while alert data for the year to July 2026 fell another 36% to roughly 2,874 square kilometers, the lowest since 2013 [37]. Clearing has fallen about half since 2022. Good. Now do the arithmetic on what is actually taking the forest down.

How many acres of rainforest are cleared for cattle each day?

Between 90% and 99% of tropical deforestation is driven, directly or indirectly, by agriculture, and the single largest commodity is beef [38]. Pendrill and colleagues attribute about 41% of tropical deforestation to pasture expansion for cattle, roughly 2.1 million hectares a year, an area about half the size of the Netherlands, or some 5,750 hectares, about 14,200 acres, every single day [39]. In Brazil, cattle ranching drives on the order of 72% of deforestation, and cattle pasture occupies roughly 70 to 80% of already-cleared Amazon land [40]. Soy is the usual scapegoat, but 77% of the world’s soy is fed to animals, not people, so most “soy deforestation” is, at one remove, meat deforestation too [41].

Is animal agriculture responsible for 91% of Amazon deforestation?

You will sometimes see the figure “91% of Amazon destruction” pinned on animal agriculture. Resist it, even though it favors this post’s argument. It traces through the film Cowspiracy to a misreading of a 2004 World Bank report by Sérgio Margulis, who attributed most, not 91%, of clearing to ranching [42]. The honest number, roughly 70 to 80% cattle pasture, is damning enough without inflation, and using it is precisely how this argument keeps its credibility. The same discipline applies to the daily-acreage figure: the dramatic “14,400 acres a day” sometimes quoted is not a separate, larger measurement but the same 2.1-million-hectare annual total run through a different unit conversion. One number, honestly stated, beats two that invite a fact-check.

What percentage of mammals on Earth are livestock?

And the biological cost is not abstract. By biomass, livestock now make up about 60% of all mammals on Earth, humans 36%, and wild mammals just 4%; all the world’s remaining wild land mammals together weigh about 20 million tonnes, roughly three kilograms per living person [43]. The frontier has moved beyond the Amazon, too, into the Cerrado savanna and the Gran Chaco, converted for soy and cattle at rates that rarely reach the headlines. Agriculture is the leading driver of terrestrial extinction, and the EU’s own Deforestation Regulation, covering cattle, soy, palm, cocoa, coffee, rubber, and timber, has now slipped to December 2026 [44]. The forest is not being lost to feed people. It is being lost to feed livestock.

Steelmanning the regenerative story, then weighing it

Does regenerative grazing actually sequester carbon?

Here is the claim at its strongest. Holistic and adaptive multi-paddock grazing, its advocates say, stimulates grass growth, drives roots deeper, builds soil carbon, and, per the most famous versions of the argument, can offset the cattle’s own emissions or even go carbon-negative, while reversing desertification. One celebrated farm’s sponsored life-cycle analysis reported net-negative beef. A famous TED talk claimed planned grazing could green the deserts. These are testable claims, and to their credit, serious scientists have tested them.

The most comprehensive evaluation is Grazed and Confused?, a two-year assessment by an international team led from Oxford’s Food Climate Research Network, dissecting every version of the grazing-as-climate-solution argument against the available science [6]. Its findings: grazing can boost soil carbon only under specific local conditions; the effect is small, time-limited (soils saturate within decades), and reversible (stop the practice, lose the carbon); and at the global maximum, roughly 0.3 to 0.8 billion tonnes of CO2 per year: it would offset only 20–60% of the emissions from the grazing animals themselves, never mind the rest of a livestock sector responsible for about 12% of all human-caused greenhouse emissions, around 6.2 billion tonnes of CO2-equivalent a year, of which cattle alone are 62% [6, 36]. The lead author’s conclusion was blunt: grass-fed livestock are not a climate solution; grazing animals are net contributors to the problem, and eating less meat, of all types, is what helps. A 2023 analysis in Nature Communications sharpened the point: for grassland carbon sequestration to genuinely offset current ruminant emissions, regional soil carbon stocks would have to grow by 25% to 2,000%, which is to say that it is not feasible [7].

Can grass-fed beef feed everyone?

Then there’s the scaling problem, which is really a land problem, which brings us back to the arithmetic. Grass-finished beef is slower and land-hungrier than feedlot beef. When researchers modeled converting all US beef to grass-fed, they found it would require a cattle population roughly 30% larger, and even then, the country’s grasslands couldn’t support current beef consumption, forcing either much less beef or the conversion of more wild and crop land to pasture [8]. Even the regenerative movement’s showcase: the sponsored life-cycle analysis of its most famous farm, conceded the operation required roughly 2.5 times the land of conventional production per unit of beef, with sequestration that plateaus as soils approach saturation [9]. And the experimental literature on rotational grazing itself, synthesized across decades of rangeland trials, finds no consistent superiority over continuous grazing on plant or animal production measures: the perception, the researchers concluded, has outrun the evidence [10].

If cows only ate grass and food scraps, how much meat could we have?

Now take the movement’s own premise seriously, seriously enough to model it. The regenerative case rests on upcycling: animals earn their place by converting grass and crop leftovers, things humans can’t eat, into food we can. Fine. What happens if livestock are restricted to only that? Researchers led by Hannah van Zanten at Wageningen ran exactly this scenario, and their answer is the most quietly devastating number in the debate. Livestock fed solely on leftover streams and grassland could supply roughly 9–23 grams of animal protein per person per day, against the 50–60 grams of total protein an adult needs; ruminants fed solely on existing grassland come to about 7 grams [30]. Set that beside current global supply of about 27 grams of terrestrial animal protein per person per day [30], and beside the far larger figures in wealthy countries, where Europe averages 51 grams. The upcycling world contains a fraction of today’s meat, and rich-country consumption has the furthest to fall. This is the deepest irony in the regenerative argument: it borrows the moral glow of a model whose own arithmetic forbids the scale of production it’s used to defend. Advocates invoke the leftover-eating cow to justify the steak; the leftover-eating cow’s math says the steak mostly has to go.

Is grass-fed beef better for the climate than feedlot beef?

The scale problem shows up in the actual production numbers too. Grass-fed beef supplies roughly one gram of protein per person per day globally, ruminants as a whole contribute about 13, which is to say that a rounding error in the world’s protein supply is being deployed to defend a global industry [6]. And the carbon accounting, when done carefully, doesn’t rescue it either: a 2025 analysis in PNAS found US grass-fed beef is about as carbon-intensive as industrial beef and roughly ten times more emissions-intensive per unit of protein than common protein-dense alternatives, even after crediting the soil sequestration its defenders cite [31]. The same paper ran the counterfactual this post keeps returning to: cropland reallocated from beef production to food grown directly for people delivers three to seven times more protein per kilogram of CO2 emitted, per hectare occupied, and per kilogram of reactive nitrogen used [31].

Notice the shape of this dispute, because it mirrors the health debate exactly. The regenerative claims live in TED talks, brand-sponsored analyses, and gorgeous documentary footage. The rebuttals live in multi-year systematic assessments, Nature-family journals, and rangeland-science syntheses. We built a ladder for exactly this situation. And even granting regenerative grazing its best case, modestly less-bad beef on modestly healthier soil, less bad is not the alternative on offer. The alternative is the land coming back entirely. Which is not a hypothetical. It has a name, an acreage, and thirty years of data.

The 270,608-acre control group

What happened when Hart Mountain removed all its cattle?

In the sagebrush steppe of southeastern Oregon sits Hart Mountain National Antelope Refuge: the subject of the documentary Rewilding a Mountain, and the closest thing the American West has to a controlled experiment on the regenerative question [11]. The refuge was created in 1936 in a desperate bid to save the pronghorn, an animal built by a million years of evolution to outrun extinct American cheetahs but not the rifles and fences of settlement: from an estimated thirty to thirty-five million animals before European settlement, the continental population had collapsed to roughly thirteen thousand by the 1920s [11, 33]. The land itself, meanwhile, had been continuously grazed by cattle since the late 1800s, straight through the refuge’s creation and for a half-century after it.

By the late 1980s, the contradiction was impossible to ignore. Riparian zones: the green ribbons along streams that occupy under 5% of the arid West’s landscape but support the majority of its wildlife species, a standard finding across Great Basin riparian ecology, were functionally dead: bare, caving banks, compacted soil that one scientist compared to a worn-out sponge that won’t hold water, and almost no willow or young aspen anywhere [11]. When refuge manager Barry Reiswig, himself from a livestock family, asked the world’s leading pronghorn authority which grazing system would best serve the antelope, the answer was three words: there isn’t one [11]. A multi-year environmental impact study led by riparian ecologist Boone Kauffman and avian ecologist David Dobkin reached the conclusion the data forced: grazing was incompatible with the refuge’s legal mandate. So in 1990–91, Hart Mountain became one of the first, and among the largest, western refuges to remove livestock entirely, buying out ranchers’ permits and beginning two decades of fence removal across 270,608 acres [11, 12]. No reseeding campaigns, no engineered stream repairs at scale. Mostly, the management plan was subtraction: take the cattle off and watch.

How fast does land recover after livestock are removed?

What happened next is now peer-reviewed. Oregon State’s Jonathan Batchelor and William Ripple, yes, the same Ripple who later co-authored the global carbon-opportunity-cost paper, compared 64 matched photographs of refuge streams taken before cattle removal in 1990 against retakes 23 years later [12]. Bare soil declined by 90%. Exposed, eroding channel declined by 63%. Willow cover increased 388% and rushes 389%; channel widths narrowed at 64% of sites and eroding banks healed at 73%; grasses, sedges, and forbs all gained [12]. The authors’ conclusion was that removing cattle produced dramatic riparian recovery even in a semi-arid landscape without active restoration, passive restoration, nature doing the work for free [12]. The tree rings tell the same story: after grazing ended in 1990, aspen recruitment rose by more than an order of magnitude, over 1,000%, filling the “missing middle” age class left by a century of cattle eating every young shoot [34]. The birds followed the plants: repeated transect surveys found overall bird abundance in the riparian aspen up about a third, with ground- and understory-nesting species specifically more than doubling, a 133% increase, even as many of those same species were declining regionally [13, 14]. A 24-year follow-up found the gains held, with three times as many species colonizing as dropping out, though cavity-nesting birds declined locally, for want of the old, rot-softened trees a century of grazing never let grow, an honest reminder that recovery can’t instantly undo everything [35]. Springs that longtime staff remembered as bone-dry now audibly run [11]. And the pronghorn, after an initial dip that critics gleefully mistook for failure, climbed from summer counts in the low hundreds in the 1970s to recent estimates of 2,000–3,500 as the last fences came down [11, 12].

Do cattle really mimic bison on native grassland?

Two lessons from Hart Mountain cut straight through the regenerative story. The first is ecological specificity. The “cattle mimic bison” argument borrows its plausibility from the Great Plains, where sod-forming grasses co-evolved with massive migrating herds. Great Basin bunchgrasses did not: they evolved with no such grazing pressure, so cattle eat them before they seed, native cover collapses, and invasive cheatgrass and desertification move in, which is precisely what the surrounding grazed lands show while the ungrazed refuge regains species found nowhere off it [11]. Regenerative grazing’s premise is, at very best, regional; marketed as universal, it is simply false. The second lesson is the documentary’s own closing verdict: millions of acres across the West are managed today the way Hart Mountain was before removal, and reaching these ecological milestones through any active management that kept cattle on the land would likely not have been possible [11]. Passive restoration worked. The regenerative movement promises that the right kind of grazing can heal land; Hart Mountain demonstrates what the land does when asked a simpler question, what if we just stopped? It’s worth adding that no rancher was ruined to find out: permits were purchased, not confiscated, a working template for a just transition, one buyout and one fence at a time.

The wildlife that doesn’t get to come back

Hart Mountain shows what returns when livestock leave. There’s a grimmer corollary: on the land where livestock stay, wildlife doesn’t merely get crowded out, a federal program kills it.

How many wild animals does the US government kill for livestock?

Wildlife Services, an arm of the USDA, exists largely to remove animals that agricultural producers consider threats. Its own reported figures are substantial: 375,045 native animals killed in 2023, including 68,562 coyotes, 24,603 beavers, 2,122 foxes, 469 bobcats, 430 black bears, 305 gray wolves, and 235 mountain lions [26]. Prior years run similar, over 400,000 native animals in 2021 [27]. Add invasive-species removals and the annual program total climbs into the millions. The methods include aerial gunning from low-flying aircraft, neck snares, leghold traps, den gassing, and M-44 sodium cyanide devices, and by the agency’s own accounting they kill thousands of unintended animals a year, golden and bald eagles, grizzlies, bighorn sheep, family pets, and, with real irony, livestock-protection dogs [26, 27].

Do predators actually kill much livestock?

Two facts make this hard to defend on its own terms. First, predators account for a small share of livestock losses overall, disease, weather, and birthing complications kill far more, so the ecological price is being paid for a modest slice of a modest problem [28]. Second, an NPR investigation using records obtained under FOIA found roughly 11,000 animals killed at locations with no recorded livestock damage at all; at one Montana site the program killed 318 coyotes with no documented coyote predation there, and across three years it killed 71 wolves at five locations where wolves had harmed 61 animals out of a state herd of roughly 2.5 million, about 6% of Montana’s wolves for predation on 0.002% of its livestock [29]. This is not the exception the program’s defenders describe; it’s the routine.

Set that beside the arithmetic from the top of this post. Livestock occupy about 83% of agricultural land, and on a meaningful share of that land, native carnivores and ecosystem engineers like beavers are killed at public expense to keep the arrangement intact. Beavers deserve their own footnote here, because Hart Mountain explains why: they build the wetlands, raise water tables, and restore exactly the riparian function the refuge regained, meaning the program is killing, by the tens of thousands, the animals that would do the restoration work for free. The wildlife-versus-livestock conflict isn’t a hypothetical tradeoff at the margins. It’s a line item in the federal budget, and dietary demand is what funds it. In the nine-column body count of the food system, field deaths, slaughter, wild capture, predator control, disease culls, on-farm mortality, pesticide kills, dead zones, and habitat loss: this is the column almost nobody tallies; The Crop Deaths Argument, Counted Honestly counts them all.

The fertility objection: “you can’t farm crops without cows”

Can you farm crops without cows for manure?

One more argument deserves the steelman treatment, because it sounds like ecology and travels fast: without livestock, farming would depend entirely on synthetic agrochemicals, manure is the natural fertility loop, grazing enriches the land, and plowing for crops poisons and kills everything anyway. Every clause of that inverts the actual accounting.

Where does nitrogen in fertilizer actually come from?

Start with what a cow contributes to soil: nothing she didn’t eat. Cattle manufacture no nitrogen, phosphorus, or potassium, manure is plant nutrients passed through an animal, minus substantial losses to air and water along the way. So the fertility question is really: where did the plants’ nitrogen come from? In today’s system, overwhelmingly from the Haber-Bosch process, synthetic fertilizer that already feeds roughly half of humanity [21] and that saturates the very feed crops whose nutrients end up in the manure pile. The livestock system doesn’t shield agriculture from agrochemicals; it is the agrochemical system, with the cow inserted as a leaky middleman. Readers of the health post will recognize the shape of this argument: it’s B12 again: the industry feeds its animals the supplement, then points to the animal as the source. Nitrogen’s actual natural source is bacteria, rhizobia on legume roots, fixing it from the air, which is precisely what the clover in a “self-sufficient” pasture is doing beneath the cattle’s feet.

Does animal agriculture cause more pesticide use?

The pesticide charge inverts the same way. Roughly two-thirds of America’s crop calories are fed to livestock rather than people [22], which means the pesticide- and herbicide-intensive corn–soy monoculture exists chiefly in service of animal agriculture. Federal regulation documents this in its own numbers: glyphosate tolerances on animal feed commodities run 1,500 to 4,000 times those permitted on the fruits and vegetables sold to people, a direct readout of where chemical intensity concentrates [32]. Shrink the livestock and you shrink the monoculture, its chemistry, and the harvested acreage, along with the field-animal deaths that acreage entails, since routing crops through animals multiplies the acres a food system must harvest, not the other way around: the full body count runs the same direction. The full version of that argument, including where the residues end up inside the animal, is its own post, The Pesticide Argument the Carnivore Movement Can’t Answer. Nor is manure the benign input of the story: its runoff helps feed the dead zone we met earlier, its nitrates seep into rural groundwater, and produce-contamination outbreaks have repeatedly pointed investigators toward neighboring cattle operations.

What is veganic farming and does it work?

Can farms actually run without any animal inputs, no manure, no blood meal, no bone meal, no grazing? They already do, with decades of receipts. The Stockfree Organic standard, developed with the Vegan Organic Network, certifies exactly this, and its flagship is Tolhurst Organic in England: the first certified stockfree farm in the world, growing on the order of 120 tonnes of vegetables a year on about 20 acres of low-grade land, its fertility built entirely from green-manure rotations, undersown cover crops, and woodchip compost from the farm’s own hedges and coppice, sustained now for more than three decades [23, 24]. The quiet irony is that mainstream organic standards already concede the principle: they specify that primary fertility must come from fertility-building crops, with manure permitted only as an adjunct [23]. Veganic growing isn’t a hypothetical awaiting invention; it’s the fertility-building-crop rule taken to its logical conclusion, with the bonus, as the stockfree growers note, of removing manure-borne pathogen pathways from vegetable production entirely [23].

Does grazing improve or degrade soil?

As for grazing “enriching” the land: the IPCC lists overgrazing among the leading human causes of soil degradation and desertification worldwide [25], and Hart Mountain’s soils told the same story in miniature, hardpan you could barely dent after a century of cattle, sponge-soft and water-holding within two decades of their removal [11, 12]. The land did not need the animals. It needed the absence of them.

The rest of the footprint, briefly

What else does animal agriculture damage?

Land and carbon are the headline, but the same subtraction pays down every other column. Routing crops through animals is why feed-crop fertilizer and manure runoff feed the Gulf of Mexico’s recurring dead zone [15]. In 2023, more than six million kilograms of medically important antibiotics were sold for use in US food animals, 56% of all antibiotics sold for livestock, with swine and cattle taking the largest shares, a primary engine of antibiotic resistance [16]. About three-quarters of emerging infectious diseases are zoonotic: the 2009 pandemic flu emerged from swine, and H5N1 bird flu, first confirmed in US dairy cattle in March 2024, had spread to more than a thousand herds across some seventeen states by early 2026, with a second independent spillover into Nevada cattle in 2025, and dense confinement operations are purpose-built incubators [17]. The response to those outbreaks is itself one of the largest body counts in the food system, and it goes almost entirely uncounted: more than 185 million birds have been killed in US flocks since February 2022, on top of the 50.5 million lost in the 2014–15 outbreak [45]; and African swine fever is estimated to have killed or forced the culling of some 143 million pigs in China across 2018–19, a figure that blends disease deaths, official culling, and panicked liquidation by farmers dumping herds ahead of the virus [46]. The waste has nowhere to go: industrial hog operations store feces and urine in open-air lagoons that emit ammonia and hydrogen sulfide over surrounding communities and have flooded into waterways during hurricanes; since 2013, the largest US pork producer, Smithfield, has been owned by China’s WH Group, meaning much of the pork is exported while American communities keep the lagoons [18]. And all of it exists to process more than 80 billion land animals through slaughter every year [19], which is the point where the environmental column meets the ethical one. As the health post noted, veganism was never a diet; it is the commitment to avoid animal exploitation as far as possible and practicable [20]. The remarkable thing is how completely the ledgers converge: the same act that spares the animals frees the land, and the same plate that compresses human morbidity un-compresses the wild.

The bottom line

What is the single biggest thing you can do for the planet?

Every serious climate solution humanity is pursuing requires building something, turbines, panels, grids, carbon-capture machinery. Rewilding through diet is the rare one that runs on subtraction. Stop routing 83% of agricultural land through animals to produce 18% of calories, and a landmass roughly the size of the US, China, the EU, and Australia combined begins pulling sixteen years’ worth of fossil carbon out of the sky while the willows, the warblers, and the pronghorn move back in, and the coyotes, wolves, and beavers stop being shot to keep them out [1, 2, 4, 11, 12, 26, 29]. The regenerative story asks us to admire a slightly better way of occupying that land. Hart Mountain, soil soft as a marshmallow where hardpan used to be, streams audible where there was silence, shows what happens when we give it back. On the refuge, they tell a story about the day after a fence came down: a dozen pronghorn approached the old line, hesitated where the wire had always been, then burst through open country as the crew cheered from the rimrock [11]. That, scaled to a planet, is what’s actually on the table.

Related reading in this series

• What the Longest-Living People Actually Eat: the health ledger: the same steelman-then-arithmetic approach applied to diet, longevity, and the hierarchy of evidence.
• Two Acres: What Beef Produces vs. What Plants Can: the acre-by-acre math: roughly 500,000 calories of beef versus 20–32 million calories of plants from the same land.
• The 86% Meme, Debunked: the “livestock eat food we can’t” statistic, examined with the very study it cites.
• The Pesticide Argument the Carnivore Movement Can’t Answer, where the chemical load actually concentrates: feed crops, and the organs and fat of the animals that eat them.
• The Crop Deaths Argument, Counted Honestly: the field-mouse gotcha traced to its source, the correction that flipped it five-to-one, and the full nine-column ledger.

References

1. Poore J, Nemecek T. Reducing food’s environmental impacts through producers and consumers. Science. 2018;360(6392):987–992. Link
2. Ritchie H. If the world adopted a plant-based diet, we would reduce global agricultural land use from 4 to 1 billion hectares. Our World in Data; 2021. Link
3. Mottet A, de Haan C, Falcucci A, Tempio G, Opio C, Gerber P. Livestock: on our plates or eating at our table? A new analysis of the feed/food debate. Global Food Security. 2017;14:1–8.
4. Hayek MN, Harwatt H, Ripple WJ, Mueller ND. The carbon opportunity cost of animal-sourced food production on land. Nature Sustainability. 2021;4:21–24. Link
5. Ritchie H, Roser M. Drivers of deforestation. Our World in Data. Link
6. Garnett T, Godde C, Muller A, et al. Grazed and Confused? Ruminating on cattle, grazing systems, methane, nitrous oxide, the soil carbon sequestration question, and what it all means for greenhouse gas emissions. Food Climate Research Network, University of Oxford; 2017. Link
7. Wang Y, de Boer IJM, Persson UM, et al. Risk to rely on soil carbon sequestration to offset global ruminant emissions. Nature Communications. 2023;14:7625. Link
8. Hayek MN, Garrett RD. Nationwide shift to grass-fed beef requires larger cattle population. Environmental Research Letters. 2018;13:084005.
9. Rowntree JE, Stanley PL, Maciel ICF, et al. Ecosystem impacts and productive capacity of a multi-species pastured livestock system. Frontiers in Sustainable Food Systems. 2020;4:544984.
10. Briske DD, Derner JD, Brown JR, et al. Rotational grazing on rangelands: reconciliation of perception and experimental evidence. Rangeland Ecology & Management. 2008;61(1):3–17.
11. Rewilding a Mountain (documentary film on Hart Mountain National Antelope Refuge).
12. Batchelor JL, Ripple WJ, Wilson TM, Painter LE. Restoration of riparian areas following the removal of cattle in the northwestern Great Basin. Environmental Management. 2015;55(4):930–942. Link
13. Dobkin DS, Rich AC, Pyle WH. Habitat and avifaunal recovery from livestock grazing in a riparian meadow system of the northwestern Great Basin. Conservation Biology. 1998;12(1):209–221.
14. Earnst SL, Dobkin DS, Ballard JA. Changes in avian and plant communities of aspen woodlands over 12 years after livestock removal in the northwestern Great Basin. Conservation Biology. 2012;26(5):862–872. Link
15. National Oceanic and Atmospheric Administration. Gulf of Mexico hypoxic zone (annual monitoring); nutrient runoff from agricultural sources in the Mississippi basin.
16. US Food and Drug Administration. Summary report on antimicrobials sold or distributed for use in food-producing animals (annual).
17. Taylor LH, Latham SM, Woolhouse MEJ. Risk factors for human disease emergence. Philosophical Transactions of the Royal Society B. 2001;356(1411):983–989.
18. Reuters. Shuanghui (WH Group) completes acquisition of Smithfield Foods. September 2013.
19. FAO. FAOSTAT: livestock primary, annual global slaughter statistics.
20. The Vegan Society. Definition of veganism (established 1944).
21. Erisman JW, Sutton MA, Galloway J, Klimont Z, Winiwarter W. How a century of ammonia synthesis changed the world. Nature Geoscience. 2008;1:636–639.
22. Cassidy ES, West PC, Gerber JS, Foley JA. Redefining agricultural yields: from tonnes to people nourished per hectare. Environmental Research Letters. 2013;8:034015.
23. Tolhurst Organic; Stockfree Farming. What is Stockfree Organic? Farm profile and standards (tolhurstorganic.co.uk; stockfreefarming.org).
24. Soil Association. Stockfree organic farming: abundance without animals? March 2023.
25. IPCC. Climate Change and Land: An IPCC Special Report. 2019. Link
26. USDA APHIS Wildlife Services. Program Data Reports, FY2023 (as compiled and reported by the Center for Biological Diversity, April 2024).
27. USDA APHIS Wildlife Services. Program Data Reports, FY2021; WildEarth Guardians analysis of FY2021–2022 data.
28. USDA National Agricultural Statistics Service. Cattle and Sheep Death Loss reports (predation as a share of total livestock losses).
29. NPR Investigations. How a federal program kills wildlife for private interests. October 10, 2024. Link
30. van Zanten HHE, Herrero M, van Hal O, et al. Defining a land boundary for sustainable livestock consumption. Global Change Biology. 2018;24(9):4185–4194. Link
31. Eshel G, Flamholz AI, Shepon A, Milo R. US grass-fed beef is as carbon intensive as industrial beef and ≈10-fold more intensive than common protein-dense alternatives. PNAS. 2025;122(12):e2404329122. Link
32. US Environmental Protection Agency. Glyphosate tolerances, 40 CFR § 180.364 (commodity-by-commodity residue limits for human food vs. animal feed).
33. O’Gara BW, Yoakum JD. Pronghorn: Ecology and Management. Wildlife Management Institute / University Press of Colorado; 2004 (historic continental population estimates).
34. Beschta RL, Painter LE, Levi T, Ripple WJ. Long-term livestock grazing alters aspen age structure in the northwestern Great Basin. Forest Ecology and Management. 2015;329:30–36. Link
35. Poessel SA, Hagar JC, Haggerty PK, Katzner TE. Removal of cattle grazing correlates with increases in vegetation productivity and in abundance of imperiled breeding birds. Biological Conservation. 2020;241:108378. Link
36. FAO. Pathways towards lower emissions: A global assessment of the greenhouse gas emissions and mitigation options from livestock agrifood systems. Rome; 2023. (Livestock ≈6.2 Gt CO2-eq, ~12% of anthropogenic emissions; cattle 62% of terrestrial animal-source-food emissions. Earlier estimate: Gerber PJ et al., Tackling Climate Change through Livestock, FAO 2013, 14.5%.)
37. Instituto Nacional de Pesquisas Espaciais (INPE). PRODES and DETER Legal Amazon deforestation data, 2024–2026 (via gov.br/Secom; Mongabay reporting). Link
38. Pendrill F, Gardner TA, Meyfroidt P, et al. Disentangling the numbers behind agriculture-driven tropical deforestation. Science. 2022;377(6611):eabm9267. Link
39. Pendrill F, Persson UM, Godar J, et al. Agricultural and forestry trade drives large share of tropical deforestation emissions. Global Environmental Change. 2019;56:1–10; and Ritchie H, Drivers of Deforestation, Our World in Data (beef ≈41% of tropical deforestation, ≈2.1 Mha/yr). Link
40. Ritchie H, Roser M. Drivers of Deforestation and Soy. Our World in Data; Greenpeace/Mongabay (cattle ranching ≈72% of Brazilian deforestation; ~70–80% of cleared Amazon land in pasture). Link
41. Ritchie H. Soy. Our World in Data (77% of global soy used as animal feed; 7% direct human food). Link
42. Margulis S. Causes of Deforestation of the Brazilian Amazon. World Bank Working Paper No. 22; 2004 (cited to document, and preempt, the inflated “91%” advocacy figure popularized by the film Cowspiracy).
43. Bar-On YM, Phillips R, Milo R. The biomass distribution on Earth. PNAS. 2018;115(25):6506–6511; and Greenspoon L, Krieger E, Sender R, et al. The global biomass of wild mammals. PNAS. 2023;120(10):e2204892120. Link
44. Regulation (EU) 2025/2650 amending the EU Deforestation Regulation (application dates 30 December 2026 / 30 June 2027).
45. USDA Animal and Plant Health Inspection Service. Highly Pathogenic Avian Influenza: commercial and backyard flock detections and depopulation totals, February 2022–present; and HPAI 2014–2015 outbreak summary. Link
46. FAO and USDA Economic Research Service. African swine fever in China: herd losses and culling, 2018–2019.
47. Lindeman RL. The trophic-dynamic aspect of ecology. Ecology. 1942;23(4):399-417. Link
48. Shepon A, Eshel G, Noor E, Milo R. Energy and protein feed-to-food conversion efficiencies in the US and potential food security gains from dietary changes. Environmental Research Letters. 2016;11:105002. Link

Read more...

There’s a book making the rounds in carnivore circles called Why Vegans Have Smaller Brains and How Cows Can Reverse Climate Change, by Anita Tagore, Alison Morgan and David Ellis. Anthony Chaffee gave it a two-part interview. Shawn Baker had them on. Judy Cho hosted all three. It’s been passed around social media with confident captions about brain atrophy and B12 and DHA. So let’s go to the source. The title comes from a single 2008 study out of Oxford. Here is Anita Tagore describing it, twice, to two different hosts:

“What this study shows is that they’ve followed a group of middle-aged and elderly people over five years, they’ve noted what they eat, they’ve done blood tests to look at various markers, and they’ve done brain scans at the beginning and end. And what they found is that the less animal food that you have in your diet, the more your brain shrinks with age. So the vegans in this study had the most brain atrophy, the vegetarians next, and the meat eaters the least.”

That study is Vogiatzoglou et al., Vitamin B12 status and rate of brain volume loss in community-dwelling elderly, published in Neurology in 2008 (PMID 18779510). It followed 107 volunteers aged 61–87 in Oxford for about five years with annual MRI. At baseline it measured plasma total B12, holotranscobalamin, methylmalonic acid, homocysteine and serum folate. It found that people in the lowest tertile of B12 status lost brain volume fastest. It did not assess diet. There were no vegans in it. There were no vegetarians in it. There were no meat-eaters in it — not as groups, not as a variable, not at all. Participants were sorted by a blood marker, not by what was on their plate. The words “vegan” and “vegetarian” do not appear. The famous “six times” figure that gets repeated by Valerie Smith and attributed to Ben Bikman — “those who ate a meat-free diet were found to be suffering from 6X the rate of brain shrinkage” — is the odds ratio comparing the lowest to highest B12 tertile. It has nothing to do with diet. There is no Oxford study showing meat-free eaters had six times the brain shrinkage, because nobody in that study was asked what they ate. That’s the whole trick. A B12 biochemistry paper gets a vegan costume put on it, and the costume becomes a book title.

Before we go further: what the evidence actually says about vegans

If you eat a well-planned vegan diet, nothing in this book applies to you. And the professional position isn’t grudging. The Academy of Nutrition and Dietetics’ current position paper — approved January 2025, in effect until 2032, built on systematic reviews — holds that appropriately planned vegetarian and vegan dietary patterns “can be nutritionally adequate and can offer long-term health benefits such as improving several health outcomes associated with cardiometabolic diseases.” Its 2016 predecessor was blunter about which ones: vegetarians and vegans are “at reduced risk of certain health conditions, including ischemic heart disease, type 2 diabetes, hypertension, certain types of cancer, and obesity.” That’s the baseline. Not tolerable if you’re careful. Not adequate with effort. Reduced risk of the diseases that actually kill people in wealthy countries. The large cohorts behind that — EPIC-Oxford, Adventist Health Study-2 — are observational, with the usual confounding caveats, and I’ll apply that same skepticism to the book’s evidence later. But it’s a substantial, consistent body of it, and it points the opposite direction from the cover of this book. We’ve covered it in more depth on the Adventist studies and what the longest-living populations actually eat. So: a person eating a varied plant-based diet, taking B12, and eating enough food has no reason to be alarmed by any of this. That’s the conclusion, stated at the top rather than buried at the bottom.

What “well-planned” means here, and what it doesn’t

Throughout this post, “a vegan diet” means reliable B12, genuine variety across whole plant foods, and enough calories. That isn’t a special exemption carved out for veganism. It’s the baseline every way of eating gets. Nobody evaluates omnivorous eating by pointing at a man living on hot dogs and beer, and nobody should. Every diet has nuance and every diet has failure modes. The Institute of Medicine already tells omnivores over fifty to take supplemental B12. The authors of this book cannot say what the nutrient requirements of their own recommended diet are — more on that below, in their own words. And I’m not going to soften the other half, because softening it would be both dishonest and dangerous: an unsupplemented vegan diet is genuinely risky, and in pregnancy and infancy it can cause real and sometimes permanent harm. That is true, it is in this post, and anyone who tells you otherwise is not doing you a favour. But notice that this is where the book’s argument quietly changes shape. Its evidence is about deficient people — B12-deficient elderly in the 2008 study, B12-deficient mothers in the case reports. Its title is about vegans as a category. Those are not the same set, and the argument only functions if you don’t catch the substitution. A vegan who supplements is not in the data the book cites. Neither, for that matter, is a vegan of any kind. Read every concession below against that. “Unsupplemented vegans are at risk” is not a concession about veganism. It’s a statement about deficiency, which is the only thing anyone here has actually measured.

The tell: when pressed, they retreat to babies

Here’s what makes this hard to write off as an honest mistake. In her long solo interview with the dentist who hosts “Dr Abs,” Tagore describes the same study accurately — B12 tertiles, brain scans, no diet groups. She knows what it says. The false version appears in the soundbites, where it does the rhetorical work; the accurate version appears when she’s speaking carefully at length. And when Judy Cho asked her directly — do vegans actually have smaller brains, and why did you title it that? — watch where she goes:

“There is plenty of evidence that diet does affect the size of your brain. The best evidence is in babies who’ve been breastfed exclusively by vegan or vegetarian mothers.”

That’s the concession. The “best evidence” isn’t the adult study the title is built on. It’s a set of infant case reports. And she adds, unprompted: “we’ve also chosen the title because it’s controversial and gets attention.” On Chaffee’s show she was even more explicit about the origin — a friend on the phone suggested it, she thought “vegetarians” would offend too many people, so she swapped in “vegans,” “because we decided we wanted the attention.” So let’s take the babies seriously, because that part is real.

The infant B12 cases: real, serious, and not about veganism

There is a documented, published literature on severe brain atrophy in exclusively breastfed infants of B12-deficient mothers. Case reports come from Turkey, India, Switzerland, Japan and elsewhere. The atrophy can be dramatic on MRI. And — this is the important part — it is often completely reversible with B12 treatment if caught early. Tagore describes this accurately. This is the strongest nutritional point in the entire book, and it deserves to be stated plainly rather than waved away: an unsupplemented vegan mother can cause serious, sometimes permanent neurological harm to her breastfed infant. That is not carnivore propaganda. That is real. But notice what the cases actually show. They are about maternal B12 deficiency, not about veganism. The same picture appears in infants of mothers with pernicious anaemia, with malabsorption, after bariatric surgery, and in undernourished non-vegan populations in low-income settings. B12 deficiency is the cause. Veganism is one of several routes to it — and the only route that comes with a free, cheap, universally recommended fix. Every major dietetic body — the Academy of Nutrition and Dietetics, the British Dietetic Association, the Canadian and Australian equivalents — treats reliable B12 as mandatory for vegans, and especially in pregnancy and lactation. A vegan mother who follows standard guidance is not the mother in these case reports. Which means the honest version of the argument is: unsupplemented vegan mothers can harm their infants’ brains. That’s true. It’s also a sentence with an obvious solution in it, which is presumably why it didn’t make the cover. For the full picture on B12 — sources, testing, dosing, and why “it’s not in plants” is a weaker argument than it sounds — see our B12 article.

The B12 argument, done properly

Two things collapse the book’s B12 case as an argument against veganism specifically. First: the intervention that actually worked in trials was a supplement, not meat. The Oxford group didn’t stop at observation. The VITACOG trials (Smith et al. 2010, PLoS One; de Jager et al. 2012; Douaud et al. 2013, PNAS) randomised people with mild cognitive impairment to B-vitamin supplements and found slowed brain atrophy in those with elevated homocysteine. Tagore even cites the follow-up finding correctly — Jernerén et al. 2015 in the American Journal of Clinical Nutrition showed the B-vitamin benefit appeared mainly in people with higher omega-3 status. She’s right about the study. But look at what it means. The thing that slowed brain atrophy in a randomised trial was a pill. Not a ribeye. Not organ meat. Not the Sapiens Diet. If your evidence that animal foods protect the brain is a trial where the intervention was a B-vitamin capsule, you have accidentally made the case for supplementation. Second: modern supplemented vegans do not have low B12. The book’s model of a vegan is someone in 1985 who has never heard of a supplement. Unsupplemented vegans genuinely are at high risk — the Pawlak review they cite is real. But in the German RBVD cohort, 97% of vegans supplemented (92% specifically with B12), and their B12 status was fine. A 2023 study from that same group is titled almost on the nose around adequate B12 status in supplemented vegans. Supplemented vegans routinely match or exceed omnivores. And there’s a slightly awkward corollary the book never addresses: if low B12 shrinks brains and supplemented vegans often have higher B12 than meat-eaters, then by the book’s own logic, the arrow points the other way.

The double standard hiding inside “well-planned”

Every version of this argument runs through a phrase that sounds like a fair caveat and functions as a loaded one: a vegan diet is fine if well-planned. The qualifier is accurate. The asymmetry in how it gets applied is the problem. Every way of eating requires planning, and B12 in particular is not a vegan problem. It’s an absorption-and-intake problem that happens to have one dietary route into it. Start with the recommendation almost nobody in this debate quotes. The Institute of Medicine’s Dietary Reference Intakes advise that adults aged 51 and over should obtain most of their B12 from fortified foods or supplements, because 10–30% of older adults have food-bound cobalamin malabsorption. The NIH Office of Dietary Supplements puts it to consumers without hedging: “People over 50 should get most of their vitamin B12 from fortified foods or dietary supplements.” Read that again. The US national nutrition authority tells every adult over fifty, whatever they eat, to get B12 by precisely the mechanism vegans use. For a large slice of the population meat is the unreliable source and crystalline B12 is the reliable one. Stomach acid declines with age, B12 bound to animal protein never gets released, and the supplemental form doesn’t need releasing. The prevalence figures follow from that. Deficiency runs around 25% in over-60s. In the Trinity-Ulster cohort of 3,299 adults aged 60 and over who were not taking B12 supplements, atrophic gastritis was present in 15% — and 38% of that group were B12 deficient, against 21% of proton-pump-inhibitor users and 15% of controls. Thirty-seven percent of the whole cohort had used PPIs for six months or longer. Then add long-term metformin, pernicious anaemia, bariatric surgery, H. pylori, and heavy alcohol use. And here is the finding that ought to end the argument: in participants with atrophic gastritis, regular consumption of fortified foods was inadequate to restore normal B12 status. Eating more steak was never going to. None of this means vegan diets carry no B12 risk. Unsupplemented, the risk is high, and I said so above. The point is narrower and worse for the book: B12 deficiency is a population-wide problem with a single universal solution, and the book has repackaged it as a vegan behaviour problem. Dean Sherzai, who treats dementia for a living, disposes of it in a word — deficiency is “actually ubiquitous.” Now apply the standard symmetrically. In four long interviews, nobody once asks whether the Sapiens Diet is well-planned. Seventy percent of calories from animal fat and near-zero plants — what’s the vitamin C position? Folate? Potassium? Magnesium? Fibre? Tagore has an answer, and it’s the most revealing thing she says in any of these conversations. On low-carbohydrate diets, she argues, requirements for vitamin C and iodine are lower than the guidelines state, because the guidelines “are all in the context of a standard western diet, which is high carb.” And then:

“So we don’t know what the actual amounts that we need on a low carb diet.”

That is the author of Why Vegans Have Smaller Brains conceding that the nutrient requirements of her own recommended diet are unknown — in the same conversation where she tells vegans theirs shrinks brains. One diet gets “we don’t know.” The other gets a book title.

DHA: where the book overclaims, and where honestly nobody knows

DHA deserves its own article and I’ll write one, because the evidence here is genuinely messier than either side admits. But three of the specific claims made in these interviews are checkable right now. Claim 1: “It’s not just fish — DHA is in cattle, in beef, and in the organ meats.” This is quantitatively wrong, and not by a little. Analyses of US beef find total long-chain omega-3 (EPA + DPA + DHA combined) running roughly 2–19 mg per 100 g in grain-finished beef and 5–33 mg per 100 g in grass-fed — and the research notes these are “primarily as EPA + DPA,” with only trace amounts of DHA specifically. The reason is rumen biohydrogenation: cattle destroy most unsaturated fatty acids before they reach the tissue. Compare oily fish: salmon delivers roughly 0.9–1.4 g of DHA per 140 g portion. That’s on the order of a hundredfold difference, and beef’s small omega-3 contribution isn’t even mostly DHA. Beef is not a DHA source. Saying it is doesn’t become true because you farm cattle. As for her “the best source is the brain of ruminant animals” — evolutionarily interesting, practically irrelevant, and worth noting that bovine brain and spinal cord are classed as specified risk material and banned from the food chain across the UK and EU following BSE. Tagore is British and runs a cattle farm. This is not a food she can recommend. Claim 2: “The conversion from ALA can grind to a halt completely with high omega-6.” The conversion is genuinely poor — low single-digit percentages at best, and omega-6 intake competes for the same desaturase enzymes. That much is fair. “Grinds to a halt completely” is an overstatement of a real limitation. But it’s also beside the point, because DHA isn’t animal-exclusive. Algal oil is where fish get their DHA in the first place. Algal DHA supplements raise blood DHA comparably to fish oil in randomised trials. You can skip the fish and go to the source. Claim 3 — and here’s the honest part: does any of this matter for cognition? Vegans do have lower plasma and red-cell DHA. That’s not in dispute. What is in dispute — and what neither the book nor its loudest critics will tell you — is whether that translates into anything. Because when you actually test it, the trials come back empty. The Cochrane review of omega-3 supplementation for cognitive decline and dementia prevention (Sydenham, Dangour & Lim, 2012) pooled three high-quality RCTs in 3,536 people aged 60+ and found no benefit to cognitive function over 6 to 40 months. Cochrane’s own summary is blunt: “three high quality randomised trials show no benefit for cognitive function.” Later reviews in MCI and Alzheimer’s are mixed at best — some signal for hippocampal volume in early-stage disease, negligible effects in others. So the state of play is: lower DHA in vegans, uncertain significance, and an RCT literature that has repeatedly failed to show that raising DHA improves cognition. That cuts against the book’s confident causal story at least as hard as it cuts against anyone else’s. I’m not going to tell you DHA doesn’t matter. I don’t know that, and neither does anyone else. What I’ll tell you is that anyone claiming certainty here — in either direction — is ahead of the evidence, and that the cheap, sensible move for a plant-based eater is an algal DHA supplement rather than a dietary philosophy.

What actual brain doctors say

The book’s medical authority is a retired GP with a one-year MSc in nutrition — who, on the Dr Abs podcast, said this about why she did it:

“I have to admit, part of the reason why I did the masters was just to give myself credibility to write a book on nutrition. It wasn’t to actually learn what was on the course.”

Fine. Let’s compare that to people who actually treat brain disease for a living.

Drs Dean and Ayesha Sherzai

Dean Sherzai, MD, PhD is a behavioural neurologist and neuroscientist, trained in neurology at Georgetown with fellowships in neurodegenerative disease and dementia at the NIH and UC San Diego, plus a PhD in healthcare leadership. Ayesha Sherzai, MD is a vascular neurologist with dual residencies in preventive medicine and neurology at Loma Linda, a fellowship in vascular neurology and epidemiology at Columbia, and a Master of Advanced Sciences in clinical research from UCSD; she received the American Heart Association’s Trudy Bush Fellowship in 2015 for research on cardiovascular disease in women. They co-directed the Brain Health and Alzheimer’s Prevention Program at Loma Linda University, then the equivalent programme at Cedars-Sinai, and are now at Charles R. Drew University of Medicine and Science, where Dean is Executive Director of Clinical Research and Professor of Neurology and Neuroscience. They wrote The Alzheimer’s Solution and The 30-Day Alzheimer’s Solution, built the NEURO framework — Nutrition, Exercise, Unwind, Restore, Optimize — and appear as expert commentators in Netflix’s You Are What You Eat: A Twin Experiment. They are, in other words, exactly the specialists this book is trying to talk over. They eat a whole-food plant-based diet themselves — and notably, per Viva!’s profile of them, not for ethical or environmental reasons but because they read the evidence as pointing there for brain health. They are not advocates who became neurologists. They are neurologists whose reading of the literature made them plant-based. Here’s Dean Sherzai on the two nutrients the book builds its entire case on:

“B12 deficiency, yes, there’s a lot of it in plant-based but it’s actually ubiquitous… You just have to be aware of it, make sure that you eat a complex diet… and be aware of your levels.” “Omega-3, be aware of it, eat foods that are rich in omega-3s like chia and flax seeds… But even then, if you’re worried, especially in certain times of life like developing brain, children, like pregnancy, or aging brain, take a supplement, algae-based.”

And on the specific question this book aims at pregnant women, Ayesha Sherzai has spoken about her own history:

“There was a time when I was pregnant and when I was having my kids, I supplemented with omega-3 fatty acids, but right now we’re taking vitamin B12.”

They are a long-term whole-food plant-based family — by their own account one that moved from lacto-ovo vegetarian toward whole-food plant-based over time, so I won’t claim more about the pregnancies themselves than she says. What she says is enough: she took the omega-3 supplement then, the family takes B12 now. That is exactly the protocol she and her husband give patients, and exactly the protocol Why Vegans Have Smaller Brains exists to tell prospective mothers is inadequate. That’s the position two dementia neurologists hold. Not “plants shrink your brain.” Not “you must eat ruminant fat.” Just: know your levels, supplement B12, take algal omega-3 at the life stages where it matters most. The book’s central claim is that this is impossible. Two people who run a dementia prevention programme live it and recommend it — and, as covered above, they classify B12 deficiency as a general clinical problem rather than a vegan one.

Dean Ornish

Ornish ran the trial that carnivore advocates most need to explain away: the first randomised controlled trial of intensive lifestyle change in mild cognitive impairment and early Alzheimer’s disease, published in Alzheimer’s Research & Therapy in June 2024 (PMID 38849944, co-authored with Rudolph Tanzi, Steven Arnold, Miia Kivipelto, Rob Knight and Dorene Rentz — this is not a fringe author list). Fifty-one participants aged 45–90, randomised 1:1, 20 weeks. The intervention: a whole-food, minimally processed plant-based diet low in refined carbohydrates, with selected supplements including B12, plus moderate exercise, stress management and support groups. Results after 20 weeks: statistically significant differences favouring the intervention on CGIC (p=0.001), CDR-SB (p=0.032) and CDR Global (p=0.037), with ADAS-Cog borderline (p=0.053). Plasma Aβ42/40 ratio improved in the intervention group and decreased in controls. Extended 40-week results have since been published. Now the limitations, because they’re real and I’m not going to hide them: 51 people is small. Twenty weeks is short. The intervention was multimodal, so you cannot isolate diet from exercise, stress management or social support. It was unblinded. Ornish and colleagues have financial interests in Ornish Lifestyle Medicine, which is disclosed in the paper. This is a phase 2 signal, not a settled result. But notice what it is: a randomised controlled trial where a plant-based diet plus B12 supplementation was associated with improved cognition and function in early Alzheimer’s, while the usual-care controls declined. Whatever you think of its size, it is more than the carnivore side has for brain health, which is zero randomised trials and a pile of testimonials.

The affirmative case, which the book never has to answer

Everything so far has been defensive — this study didn’t say that, that citation drops its caveat. But the positive case is the more interesting half, and it runs through mechanisms the book’s own argument depends on.

Inflammation

Chronic low-grade inflammation is increasingly treated as a driver of neurodegeneration, and plant-based eaters have less of it. A systematic review and meta-analysis in Scientific Reports (Menzel et al. 2020, PMC7730154) pooled 21 studies and found vegan diets associated with lower C-reactive protein than omnivorous diets — a mean difference of −0.54 mg/L (95% CI −0.79 to −0.28, p < 0.0001). Those were cross-sectional, so a 2026 meta-analysis in Nutrition, Metabolism and Cardiovascular Diseases went looking for randomised evidence and found seven RCTs covering 541 participants; plant-based dietary patterns significantly lowered CRP versus omnivorous diets. The authors rate the certainty as low given heterogeneity and bias risk, and I’ll flag that rather than bury it — but the observational and randomised evidence point the same way. The plausible drivers are the ones the carnivore literature treats as liabilities: fibre, polyphenols, flavonoids and carotenoids, which is a topic we’ve covered separately. Anthony Chaffee’s “plants are trying to kill you” and the actual inflammatory biomarker data are not compatible.

The vascular route, which is the biggest one

This is the mechanism most people underrate. What damages arteries damages the brain — the vasculature feeding it is the same vasculature. Midlife hypertension and elevated LDL are among the most firmly established modifiable dementia risk factors, and vascular contributions to cognitive impairment are a major share of late-life decline, including in people who also have amyloid pathology. So the cholesterol chapter isn’t a separate argument from the brain chapter. It’s the same argument. A dietary pattern that raises ApoB and blood pressure is not neutral for the brain, whatever it does for your triglyceride-to-HDL ratio — and this is the thread running through our cholesterol article. It is also, not coincidentally, why Ornish’s intervention and the Sherzais’ NEURO framework both spend as much attention on blood pressure and lipids as on any single nutrient.

Methionine and homocysteine — the book’s own mechanism, running backwards

Here’s where it gets awkward for them. The entire brain-atrophy argument in this book runs through homocysteine: low B12 → elevated homocysteine → accelerated brain volume loss. Fine. Now ask where homocysteine comes from. It’s a metabolite of methionine, and methionine comes disproportionately from animal protein — plant proteins are lower in it, which is why studies comparing the groups have found vegans with measurably lower methionine intake. In animal models the downstream effects are not subtle. Feeding mice a 2% high-methionine diet for nine weeks produced learning and memory impairment, hippocampal and cortical neuronal damage, elevated serum homocysteine, and increased brain Aβ1-40 and Aβ1-42, along with raised APP and BACE1 and reduced amyloid-clearing enzymes (Pi et al. 2021, Behavioural Neurology, PMC8046555). Diet-induced high homocysteine in 3xTg mice worsened all three hallmarks at once — memory deficits, amyloid, and tau. And a 2026 paper in Alzheimer’s & Dementia found late-life methionine restriction reduced amyloid deposition and neuroinflammation in aged Alzheimer’s mice. Two honesty notes. First, the mechanism isn’t that methionine “becomes” amyloid — it’s methionine → homocysteine → amyloid and tau pathology, an indirect route. Second, it isn’t uniform: in one tauopathy model (rTg4510), methionine restriction increased tau and inflammatory markers. And all of this is rodent and mechanistic work, not human outcome data. I’m not going to do to this literature what the book did to Vogiatzoglou. But the shape of it stands. The book selected homocysteine as its causal pathway and then declined to follow it upstream to the amino acid that generates it.

And homocysteine itself doesn’t behave the way the book needs it to

Unsupplemented vegans do have higher homocysteine. That’s consistent across meta-analyses, including Niklewicz et al. (2024, Nutrition Bulletin), and I’m not going to pretend otherwise. But homocysteine is not a fixed property of eating plants. It’s a modifiable number, and B12 is the modifier. Niklewicz’s own subgroup analysis found vegan supplement users had higher B12 and holotranscobalamin and lower MMA and homocysteine than non-supplementing vegans. Haddad et al. (1999) found supplementing vegans at 8.0 µmol/L against omnivores at 7.9 — statistically indistinguishable. Crane et al. (1998) gave vegans 500 µg/day of B12 for two months and measured homocysteine below 5 µmol/L, which is low by any standard. In Adventist Health Study-2, serum B12, holoTC, MMA and homocysteine did not differ across diet groups on average. I won’t overclaim this: “supplemented vegans have lower homocysteine than meat-eaters” is not an established general finding. What is established is narrower and sufficient. The single variable this book identifies as the driver of brain atrophy is one that vegans control directly, cheaply, and — when they supplement — well. The book presents that variable as an indictment of a diet. It’s an argument for a pill, and the pill costs a few cents a day.

“Meat made our brains big” — and the Lamarckian error hiding inside it

The book leans hard on evolution. Dave Ellis’s 24-hour-clock metaphor: 23 of 24 hours as carnivores, agriculture in the last hour, dietary guidelines in the last ten seconds, and brain size shrinking ever since. Three problems. First, the meat hypothesis is not the consensus explanation. It’s one candidate among several: cooking (Wrangham), tubers and dietary carbohydrate (Hardy et al. 2015 in the Quarterly Review of Biology), fat specifically rather than muscle meat, social and ecological drivers. It is a live debate, not a settled finding. Second, the hypothesis’s own author says the book is overreading it. The Expensive Tissue Hypothesis is Leslie Aiello’s. In 2008 she delivered the George Peabody Founder’s Lecture at Harvard, revisiting it after fifteen years of scrutiny. The Harvard Gazette’s report is worth quoting exactly:

“Our human ancestors were not wholly carnivores — ‘that would be silly,’ said Aiello, who does not argue that meat-eating caused bigger brains — just that it made bigger brains possible.”

The same article notes that ETH now “has theoretical competitors,” that some researchers point to bipedalism or reduced muscle mass instead, and that “ETH doesn’t hold true for all animals, including birds and bats.” Navarrete, van Schaik and Isler found no brain–gut size correlation across mammals in Nature in 2011. The originator of the hypothesis says our ancestors weren’t carnivores, calls the idea silly, and explicitly declines to claim causation. The book claims causation and builds a hypercarnivore diet on it. Third — and this is the load-bearing error nobody names — the argument is Lamarckian. Even granting everything: meat enabled brain expansion over roughly two million years. That is Darwinian selection acting on populations. It means individuals with certain gut-brain configurations out-reproduced those without, given an energy-dense diet. It does not mean that you, eating a steak in 2026, grow a bigger brain. It does not mean your children will have bigger brains. You will pass on the genes you have. The only way to shift brain-size distribution in a population is differential reproduction — some lineages having fewer descendants than others. “Meat made our brains big over two million years, therefore you personally need meat for your brain today” is a category error dressed as ancestral wisdom. It’s the same error as “our ancestors evolved lactase persistence, therefore drink milk.” Evolutionary history explains why a trait exists. It does not prescribe your dinner. And on the brain-shrinkage-since-agriculture point: it’s real but contested. DeSilva et al. (2021) dated the decline to ~3,000 years ago; Villmoare and Grabowski rebutted the timing and dataset in 2022. Leading explanations include reduced body size, self-domestication and externalised cognition — and in several analyses the decline predates agriculture, which is inconvenient for a story blaming grain. Within the normal human range, brain volume correlates with cognition only weakly (r ≈ 0.24–0.4). Bigger is not straightforwardly better. There’s also positive evidence for plant eating in hominins that the book skips: starch granules in Neanderthal dental calculus from Shanidar and El Sidrón, the AMY1 amylase copy-number expansion, and recent isotope work on the Taforalt Iberomaurusians in Morocco indicating a heavily plant-based pre-agricultural diet.

The cholesterol chapter

On Judy Cho’s show, Tagore and Ellis run through the standard set. Worth taking one at a time. “Cholesterol is the fire engine at the fire, not the cause of the fire.” This analogy has been circulating for years and it fails at the same place every time: fire engines don’t show up before the fire. LDL does. The causal evidence for LDL and ApoB in atherosclerosis runs through Mendelian randomisation (people genetically randomised to lifelong low LDL get less heart disease), through familial hypercholesterolaemia (lifelong high LDL, early heart disease, no other explanation), through a dose-response gradient across dozens of LDL-lowering RCTs with different drug mechanisms, and through the European Atherosclerosis Society consensus statement (Ference et al. 2017). It’s about as settled as nutrition-adjacent science gets. Full treatment in our cholesterol article. “Triglyceride-to-HDL ratio matters more than LDL.” Trig:HDL is a decent proxy for insulin resistance and worth knowing. It does not erase ApoB. These are not competing claims — you can improve insulin sensitivity and raise your atherogenic particle count, which is precisely what happens to a lot of people on this diet. “In older people, higher cholesterol is associated with lower all-cause mortality.” A real observation with a well-understood explanation: reverse causation. Frailty, cancer, and undiagnosed illness lower cholesterol. Sick people have low cholesterol because they’re sick. “It’s glucose that damages cholesterol — carbs are the new smoking.” The comparison to smoking is theatre. And it flattens the distinction between refined and whole carbohydrate that does most of the actual work. The cleanest counterexample is the Tsimane of Bolivia: Kaplan et al. in The Lancet (2017) CT-scanned 705 adults and found the lowest coronary artery disease of any population ever measured, roughly five times less than the US — on a diet that’s about 72% carbohydrate. On animal fat specifically, see our article on fat and animal fat versus coconut oil. The “just stop vegetable oil” chapter gets its own treatment in our seed oils piece.

Fiber, the colon, and the breastfed baby

Tagore makes specific, checkable claims here, and this is a case where she cites something real and then strips off the caveat. On colon cancer, she describes “a Cochrane review of controlled trials where they increased fiber for 2 to 8 years, and that didn’t lower the risk of colon cancer, and paradoxically after four years there was a slight increase.” She’s referring to the Cochrane review on dietary fibre and recurrent colorectal adenomas (Asano & McLeod 2002, updated by Yao et al. 2017, CD003430). And the numbers are as she says: no significant difference for at least one adenoma (5 RCTs, n=3,641, RR 1.04), and for colorectal cancer diagnoses the results “favoured the control group over the dietary fibre group” (2 RCTs, n=2,794, RR 2.70, 95% CI 1.07–6.85). What she leaves out is the sentence Cochrane wrote immediately afterward:

“However, these results may be unreliable and should be interpreted cautiously, not only because of the high rate of loss to follow-up, but also because adenomatous polyp is a surrogate [outcome].”

Low-quality evidence. Two trials. A confidence interval running from 1.07 to 6.85 — which is another way of saying “we have almost no idea.” And a surrogate endpoint measured over three to four years for a cancer that develops over decades, in trials that mostly tested fibre supplements rather than dietary patterns. Set against that: Reynolds et al. in The Lancet (2019), covering ~135 million person-years across 185 prospective studies and 58 clinical trials, found the highest fibre consumers had a 15–30% lower all-cause and cardiovascular mortality and 16–24% lower incidence of coronary heart disease, stroke, type 2 diabetes and colorectal cancer. WCRF/AICR grade fibre-rich whole grains as protective against colorectal cancer. Meanwhile IARC classifies processed meat as a Group 1 carcinogen and red meat as Group 2A — a fact that goes unmentioned in a book warning about colon cancer. More on that in our piece on the meat and bowel cancer link. On constipation, both Tagore and Judy Cho lean on a Singapore study. That’s Ho et al. 2012 in the World Journal of Gastroenterology (PMID 22969234): 63 patients with idiopathic constipation, all placed on a no-fibre diet for two weeks, then allowed to self-select their intake. No proper control group. Non-randomised. A highly selected population of people already constipated. Some constipated people improve on less fibre. That’s clinically useful and I’ve no quarrel with it. It does not make fibre “a natural junk food” for everyone else. And then there’s the baby argument, which both of them treat as a knockdown. Judy Cho: “he never had fiber in that one year and he massively grew… there’s no fiber in that.” Tagore: “breastfed babies have zero fiber in their diet, they have no problems with constipation, they don’t need fiber at all.” Human milk oligosaccharides are the third most abundant solid component of human breast milk, after lactose and lipids. More than 200 distinct HMO structures have been identified. They run roughly 5–20 g/L in mature milk and 20–25 g/L in colostrum. A breastfed infant consumes up to around 10 g per day of them. They are indigestible by the infant’s own enzymes. They pass intact through the upper GI tract and reach the colon, where they act as selective substrates for beneficial bacteria — principally Bifidobacterium. That is fibre. It is functionally, definitionally fibre: indigestible carbohydrate that reaches the colon and feeds the microbiome. Breast milk is loaded with it, at metabolic expense to the mother, which in evolutionary terms is about as strong a signal of importance as you can get. And the follow-on argument makes it worse for them. Tagore points out that breastfed and formula-fed infants have different short-chain fatty acid profiles. Correct — and SCFAs are produced by bacterial fermentation of exactly these indigestible carbohydrates. She has cited the mechanism by which fibre produces its benefits as evidence that fibre is unnecessary. One more: human milk is roughly 40–45% of calories from carbohydrate, as lactose. For a book arguing we evolved to be ketogenic, the food evolution designed specifically for the fastest-growing human brain is a high-carbohydrate, high-fibre liquid. On lectins, oxalates and the broader “plants are trying to kill you” framework — including Chaffee’s “Brussels sprouts contain 130 carcinogens” — see our article on plant toxins and antinutrients. On the beta-carotene conversion argument they raise, see here.

The clinical claims

Tagore says she’d recommend this diet “to absolutely everybody.” Chaffee claims MS reversal and tumour regression. Baker’s guests describe reversed autoimmune disease. There is no randomised controlled trial of a carnivore or near-carnivore diet with hard clinical endpoints. Not one. The largest dataset is Lennerz et al. (2021), a self-reported online survey of 2,029 people recruited from carnivore-diet social media. Self-selected. Self-reported outcomes. No control group. And structurally incapable of seeing the people who tried it, felt terrible, and left — which is the entire population you’d need to hear from. The Hungarian Paleomedicina clinic on which the Sapiens Diet is explicitly based publishes mainly uncontrolled case reports and small case series, often in low-tier venues, claiming reversal of cancer and inflammatory bowel disease from single cases. That is the weakest possible evidence for the strongest possible claims. Low-carbohydrate diets can drive type 2 diabetes remission — that’s real. But so can other routes: DiRECT achieved 46% remission at one year using a low-calorie liquid formula diet. Weight loss is doing much of the work, and there are many ways to lose weight. More in our weight loss article. On “leaky gut”: intestinal permeability is real and measurable. “Leaky gut syndrome” as a unified cause of autoimmune disease is not established. Tagore’s own evidence is a PEG-400 test she ran on herself before and after four weeks — an n=1, unblinded, uncontrolled self-experiment using an assay that is no longer commercially available. Commercial zonulin ELISAs, meanwhile, have been shown not to actually measure zonulin.

Two small things worth checking

Tagore states that brain atrophy runs about 1% per year from the mid-20s, accelerating to 2% per year in the 50s. The figures in the neuroimaging literature are considerably lower for midlife — typically well under 1% annually before older age. I’d treat her numbers as inflated, though I’d encourage you to check the primary sources rather than take my word for it. And “the brain is 60% fat,” repeated by everyone in these interviews as though it settles something, refers to dry weight. The brain is roughly three-quarters water. More to the point, the brain synthesises its own cholesterol de novo behind the blood-brain barrier — dietary cholesterol doesn’t cross it. “Your brain is made of fat, therefore eat fat” has the same logical structure as “your bones are made of calcium, therefore eat chalk.”

The environmental half, briefly

The subtitle claims cows reverse climate change. Three facts, then I’ll point you elsewhere. The ice-core record shows atmospheric methane stable at roughly 700 ppb for millennia and now around 1,900 ppb — unprecedented in 800,000 years. That alone refutes Alison Morgan’s claim that we have ruminants today “in similar numbers to the numbers we have today, wild and domesticated.” Bar-On, Phillips & Milo (2018, PNAS) found livestock biomass vastly exceeds wild mammal biomass; humans and livestock together are ~96% of mammalian biomass on Earth. And Poore & Nemecek (2018, Science), using consistent life-cycle boundaries across ~38,700 farms — which answers their complaint about inconsistent accounting — still finds beef roughly 25 times more emissions-intensive per gram of protein than tofu. One new claim from the Judy Cho interview deserves flagging because it’s the most dubious thing either of them said: Morgan asserts that 45% of UK soy imports go to vegan and vegetarian meat alternatives, and that this rather than beef drives deforestation. This is not consistent with UK soy footprint data — the overwhelming majority of soy imported into the UK goes into animal feed, primarily for poultry and pigs. I’d treat that figure as unsupported pending a source. We’ve handled the rest at length elsewhere: regenerative farming, rewilding versus regenerative, crop deaths, the pesticide argument, and the “livestock turn food we can’t eat into protein” claim.

The true statements in the book, and why not one of them reaches the title

A book can contain fifty accurate claims and one false title, and the fifty do not repair the one. That’s card stacking, and it’s the standard architecture of this genre. You assemble real citations, real mechanisms, real concerns — none of them individually objectionable — and let the accumulated weight of “well, that part’s true” carry a conclusion that none of it supports. The reader comes away thinking the case was cumulative when it was only long. So I’m not interested in grading the book. What follows isn’t credit to the authors; it’s the handful of things in the underlying science that a reader deserves to have straight no matter who’s saying them, each with the reason it doesn’t get you within a mile of the cover. Unsupplemented vegans are genuinely at risk of B12 deficiency, and the infant cases are real. Plant-based advocates should say this more loudly, not less. But the title isn’t “unsupplemented vegans have smaller brains.” The condition is prevented by a supplement every dietetic body already mandates, and the same deficiency arises from pernicious anaemia, malabsorption, bariatric surgery, metformin and PPIs. The pregnancy claim is the book’s most aggressive, and it’s the one where professional guidance is least ambiguous. A well-planned vegan pregnancy is endorsed as safe by the 2016 Academy of Nutrition and Dietetics position paper, by British Dietetic Association and NHS-aligned midwifery guidance — which puts it plainly that “well-planned vegan and plant-based diets are safe during pregnancy and lactation” — and by the expert panel of the Scientific Society for Vegetarian Nutrition, whose review concluded a completely plant-based diet “is suitable during pregnancy, lactation, infancy, and childhood, provided that it is well-planned.” A 2024 analysis in the Journal of the Academy of Nutrition and Dietetics went further and modelled it directly, finding that vegan, lacto-vegetarian, ovo-vegetarian and pescatarian adaptations of the USDA Healthy Vegetarian Dietary Pattern can all be nutritionally adequate for pregnancy with supplementation. And the claim has been tested where it counts. An analysis from the Avon Longitudinal Study of Parents and Children compared 78 vegetarian and 2,144 omnivorous mothers in early pregnancy, and 91 versus 2,552 in late pregnancy, then followed the children. The vegetarian mothers had lower blood concentrations of docosahexaenoic acid, arachidonic acid and cobalamin in both early and late pregnancy — precisely the exposure this book says damages developing brains. The paper’s title is its finding: Vegetarian Diet during Pregnancy Is Not Associated with Poorer Cognitive Performance in Children at Age 6–7 Years. After controlling for confounders, there was no association with poorer neurocognitive development. That’s vegetarian rather than vegan, and it’s one study. But it is the closest thing anyone has to a direct test of the book’s central claim about mothers and children’s brains, it ran the exposure the book nominates, and it came back null. Well-planned here means a slightly longer checklist than B12 alone: reliable B12, iodine, vitamin D, iron, choline, zinc, algal DHA, and enough calories. Before anyone reads that list as an indictment, look at what every omnivorous pregnant woman is handed. Folic acid, from before conception through the first trimester — universal, non-negotiable, and a supplement. Vitamin D daily, per NHS guidance, for every pregnant woman in Britain regardless of diet. Iron at 27 mg a day — the pregnancy RDA — for everyone. Iodine, where deficiency has been documented in pregnant women in the south-east of England eating ordinary mixed diets. Choline, which most pregnant women fall short of on any dietary pattern. DHA, where omnivorous women get an advisory telling them which fish to avoid because of mercury. There is no such thing as an unplanned pregnancy diet that anyone recommends. Every pregnant woman on earth is told to take supplements, watch specific nutrients, and see her doctor. Nobody calls that regime “well-planned omnivorous eating,” because the qualifier only ever gets attached to the diet somebody wants to make sound precarious. The vegan list differs by exactly one item that moves from optional to mandatory: B12. Everything else on it is the same list, handed to the same women, by the same clinicians. So the advice is identical in shape whatever you eat. Cover your bases, take the prenatal, get your bloods checked, and talk to your doctor or midwife. A pregnant vegan taking B12 is not the woman in this book’s case reports, and never was. Now the birth weight data, which is real but far weaker than the use they put it to. A 2024 meta-analysis found strict vegetarian diets associated with small-for-gestational-age births at a pooled odds ratio of 2.71 — with a 95% confidence interval running from 1.24 to 5.95. A 2020 Journal of Perinatology analysis reported RR 5.9, CI 1.2 to 21.8. Intervals that wide are the statistical equivalent of a shrug. The 2024 Nutrients systematic review concluded the available studies had insufficient power and serious limitations. More to the point, the specialist literature already separates the two populations the meta-analyses blur together. Baroni and colleagues, writing on vegan nutrition for mothers and children, state it directly: “The average birthweight of infants born to vegan mothers does not differ significantly from that of infants of omnivorous mothers. Macrobiotic vegan women, whose diets can be highly restricted in calories and nutrients, in contrast to well-planned vegan diets, give birth to infants whose weights are significantly lower than expected.” Much of the older data comes from exactly those restricted communities. Layer on the fact that vegans are leaner with lower gestational weight gain — both strong independent determinants of birth weight — and a smaller baby born to a smaller mother gets coded SGA without anything having gone wrong. The same authors note that plant-rich eating in pregnancy may be protective against pre-eclampsia. On energy specifically: plant foods are less energy-dense, and in populations where most adults are now overweight or obese — around three in four US adults and roughly two in three in the UK — that is overwhelmingly an asset — it’s one of the mechanisms by which plant-based eating helps with weight. Pregnancy is the narrow window where it cuts the other way and intake needs deliberate attention. A planning point for one life stage, not a defect in plants. And even granting the signal in full: birth weight is not brain size, and brain size is not cognition. And as above, the one study that actually followed the children of mothers with lower DHA and B12 through to cognitive testing found no deficit. Tagore’s closing message to prospective mothers — that without animal foods their children “won’t fulfill their cognitive potential” — is not a finding. It is the book’s thesis restated as a warning to pregnant women, aimed at the group least able to shrug it off. The Cochrane adenoma finding is real — RR 2.70 for colorectal cancer diagnoses favouring control. But Cochrane’s own next sentence calls it unreliable, on two low-quality trials with heavy loss to follow-up and a surrogate endpoint. Citing the number while dropping the caveat is not reporting a finding; it’s using one. The Jernerén DHA–B vitamin interaction is real and cited correctly. But the intervention was a B-vitamin capsule. It is evidence for supplementation, which is the thing the book exists to argue against. Nutritional epidemiology has real weaknesses. Food frequency questionnaires are bad, confounding is rampant, and the under-reporting research they cite is legitimate. But this one boomerangs hardest of all, because the entire book rests on an observational study its lead author described inaccurately on two podcasts. The elderly cholesterol association is a real observation. And reverse causation explains it. Sick people have low cholesterol because they’re sick. Run the list again and notice what’s absent. Not one item is evidence that vegans have smaller brains. Every one of them is either a solved problem, a surrogate endpoint, an argument for supplements, or a standard of evidence the authors decline to apply to themselves.

Where our side overreaches

Precision is the whole point of this post, so: Mic the Vegan’s response video — which is right about the central claim — describes the Ornish trial as running “5 years.” It ran 20 weeks. His methane “airborne fraction” argument misunderstands how GWP is calculated (the CO₂ term already incorporates the atmospheric impulse response, including ocean and land uptake). And he sourced ruminant population figures via ChatGPT, on air, and said so. If we’re going to hold Tagore to the standard of describing studies accurately, we hold our own to it too. More broadly: vegan diets require planning. They are not automatically optimal. EPIC-Oxford found higher haemorrhagic stroke risk in vegetarians alongside lower ischaemic heart disease. Anyone telling you a plant-based diet is bulletproof is selling something, same as the people telling you it shrinks your brain. One study you’ll see cited at you is worth describing accurately, because almost nobody has. Norgren et al. (JAMA Network Open, March 2026) analysed 2,157 adults from the SNAC-K cohort in central Stockholm, mean age 71.2, followed up to 15 years, and reported that higher unprocessed meat intake tracked with slower cognitive decline and lower dementia risk in carriers of APOE ε3/ε4 and ε4/ε4. Carnivore media ran with it, some under RFK Jr. headlines. What the headlines left out. The association was not observed in the study population as a whole, and the Science Media Centre’s expert reaction describes it as “only marginally significant” within the genotype subgroups. Across every genotype, a higher proportion of processed meat was associated with increased dementia risk. And the cohort — Swedes born roughly 1920–1941 — contains no meaningful vegan population. This compares lower-meat omnivores to higher-meat omnivores. Nobody in it was following a planned plant-based diet or supplementing B12. Credit where due: the model adjusted for education, occupation type, living arrangements, physical activity, smoking, alcohol, total energy intake and the Alternative Healthy Eating Index, so socioeconomic and diet-quality confounding was partly addressed. Reverse causation is harder to dismiss — people in early undiagnosed decline eat differently and recall diet worse, over fifteen years in a cohort aged 60+. The mechanism the authors reached for is the interesting part. Post-hoc analyses using a proxy for B12 absorption suggested APOE ε4 carriers took up more vitamin B12 per unit of meat than other genotypes — what Norgren calls a “food matrix” effect, and explicitly labels hypothesis-generating: “we can only speculate.” Note what that implies. If the operative variable is B12 delivery from the food matrix, a crystalline B12 supplement bypasses the food matrix entirely — which is exactly why the Institute of Medicine recommends the supplemental form to everyone over fifty. Hussein Yassine’s lab has published a careful critique of the paper worth reading if you want the full methodological picture. So: real study, serious group, genuinely open question for APOE4 carriers eating little meat and supplementing nothing. It says nothing whatsoever about a supplementing vegan, because it did not contain one. Which is, once again, the entire problem with this genre. Set against it, the largest recent signal runs the other way: a 2025 Neurology analysis from Harvard T.H. Chan and Mass General Brigham found a quarter-serving of processed red meat daily associated with 13% higher dementia risk and roughly 1.6 years of additional cognitive ageing per daily serving, with substitution modelling suggesting 19% lower risk swapping in nuts and legumes and 28% swapping in fish. Critics have fairly noted that unprocessed red meat showed no significant association with objective cognitive measures there. Both can be true. Neither gets you to the cover of this book. On the population-level evidence, see our work on the longest-living populations, Adventist Health Studies, and both sides of the Blue Zones argument.

What to actually do

The honest version of the concern this book is exploiting is short, and it looks nothing like the Sapiens Diet.

• Take B12. Reliably, forever, and get your levels checked. Non-negotiable, and doubly so in pregnancy and lactation.
• Consider algal DHA/EPA — the Sherzais’ recommendation, especially during pregnancy, in childhood, and in older age. The evidence that it improves cognition is weak. The cost of covering the base is trivial.
• Cover iodine, vitamin D, iron, zinc and choline. See our nutrition overview.
• Do the things with the strongest dementia evidence, which are mostly not on your plate: exercise, sleep, blood pressure control, hearing correction, social connection, cognitive engagement. Both Ornish’s trial and the Sherzais’ NEURO framework are multimodal for a reason.
• Eat plants. Fibre, polyphenols, and vegetable intake track with better outcomes across essentially every large dataset we have.

That’s it. That’s the whole brain-health protocol, and it’s compatible with a fully plant-based diet in a way that two practising dementia neurologists have been recommending publicly for years. One note on the professional guidance, since it’s being misrepresented. The Academy of Nutrition and Dietetics’ 2016 position paper stated that appropriately planned vegetarian and vegan diets are “appropriate for all stages of the life cycle, including pregnancy, lactation, infancy, childhood, adolescence, older adulthood, and for athletes.” That paper expired at the end of 2021 and was replaced in January 2025 by Vegetarian Dietary Patterns for Adults — which reaffirms nutritional adequacy and long-term benefit in adults, and does not address pregnancy or children because AND judged those to need separate guidance. That is a scope decision, not a reversal, and anyone telling you AND now warns against vegan diets for children or pregnancy is misreading a table of contents. For pregnancy and lactation specifically, British Dietetic Association and NHS-aligned midwifery guidance, Dietitians of Canada, and the Italian Scientific Society for Vegetarian Nutrition all hold that well-planned vegan diets are appropriate, and a 2024 Journal of the Academy of Nutrition and Dietetics modelling study found vegan adaptations of the USDA Healthy Vegetarian Dietary Pattern nutritionally adequate for pregnancy with supplementation. And the American Academy of Pediatrics’ Pediatric Nutrition Handbook, in its 5th edition (2004), stated that “children exhibit good growth and thrive on most lacto-ovo vegetarian and vegan diets when they are well planned and supplemented appropriately,” with the 6th edition (2008) taking the same line on adequacy with fortified foods or supplements as needed. One honest complication, since this post is about describing evidence accurately. The dietetic and obstetric bodies are supportive; some paediatric bodies are more cautious about infancy and childhood specifically. In August 2025 the ESPGHAN Nutrition Committee published a position paper (Verduci et al., Journal of Pediatric Gastroenterology and Nutrition) — the first based on a systematic search rather than narrative review — which found only 10 qualifying studies covering roughly 1,500 vegan children, concluded that supplementation “including vitamin B12, is essential during paediatric age when following a strict vegan diet,” and called for high-quality trials to close the evidence gaps. The Canadian Paediatric Society is likewise more comfortable when dairy and eggs are included. That’s a finding of insufficient evidence and a requirement for proper supervision, not a finding of harm, and its emphasis on B12 is the same one made throughout this post. But it’s real, it’s recent, and it deserves stating rather than skipping — particularly since it makes the same criticism of expert-opinion position papers that this post makes of the book. It also concerns weaning and childhood rather than pregnancy, and it says nothing about brain atrophy in adults, which is what the book on the table actually claims.

On raising children this way

The book’s closing message is aimed at prospective mothers, so it’s worth ending where it does. Start with what the leading bodies actually say. Vegetarians and vegans are at reduced risk of ischaemic heart disease, type 2 diabetes, hypertension, several cancers and obesity — that’s the Academy of Nutrition and Dietetics, not an advocacy group, and its 2025 position paper reaffirms long-term benefit for cardiometabolic outcomes in adults. Those are, between them, most of what kills people early in wealthy countries and most of what steals the last decade of a life. Now add the timing. Atherosclerosis doesn’t wait for adulthood. The Bogalusa Heart Study and PDAY found fatty streaks and early lesions in children and adolescents, tracking with LDL. The disease process the adult data is measuring begins in childhood. And dietary patterns established early tend to persist. If a way of eating lowers the risk of the diseases in question, the window in which it operates opens young. Now the objection, and why it’s weaker than it sounds. No cohort has followed vegan-raised children from conception into late adulthood and measured chronic disease outcomes. True. It has also never been done for any other diet. Nobody has run that study on omnivorous children, and nobody demands it before handing out the standard advice — advice attached to a dietary pattern that has never cleared the bar being set here and that produces the population outcomes this post has spent several thousand words discussing. That standard, applied consistently, would forbid all paediatric nutrition guidance. The reason children are told to eat vegetables and limit sugar is not a lifetime randomised trial. It is adult outcome data, plus mechanistic plausibility, plus short-term paediatric evidence — the identical structure of inference at work here. Demanding more of veganism specifically is the same asymmetry as “well-planned,” wearing a lab coat. So the honest formulation isn’t “we don’t know.” It’s this: given that the best available evidence says this way of eating lowers the risk of the diseases most likely to kill your child in sixty years, and given that the disease process starts in childhood, raising them that way follows. Not proven in the sense that no dietary recommendation for children is proven. But reasoned, and reasoned from a stronger evidence base than the default it’s being measured against. Which is what the parents I’ve interviewed are actually doing when they say — and they say it often — that raising their children vegan is the best gift they could give them. That isn’t a testimonial about how someone felt after a diet change. It’s a conclusion drawn from the same evidence the professional bodies publish, by people who then took the B12 seriously, had the bloods done, and got on with it. They’ve also factored in something the risk calculations don’t capture: the animals, and the planet those children inherit. A values position resting on a defensible evidence base. Which is a different and more honest object than the book on the table — a values position resting on a study that never studied a vegan. The book’s title is a marketing decision its own authors admitted to on two podcasts. The study underneath it never looked at a vegan. And when you press on the “best evidence,” what you find is a solved problem with a two-cent solution, wearing a much more dramatic hat.   If you spot an error above, tell me and I’ll correct it. That’s supposed to be how this works. Read more...

Mic the Vegan’s video covers some of this really well. So I wanted to include it here. He does such fantastic work. I have never been one to inject politics into my work. But the U.S. government has now crossed a line so clearly that silence would be complicity.

The finalized 2025–2030 Dietary Guidelines for Americans are not a faithful reflection of the scientific evidence they claim to represent. They are a political document — not a scientific one — and they directly contradict the findings of their own expert advisory committee.

That matters, because these guidelines influence school lunches, hospital food, military rations, public health messaging, and medical norms for hundreds of millions of people.

The Science Was Clear. The Government Ignored It.

The 2025 Dietary Guidelines Advisory Committee (DGAC) released its Scientific Report in late 2024. That report explicitly recommended a “bold shift” toward plant-based protein, based on strong, consistent evidence that higher intake of plant protein is associated with lower risk of chronic disease.

Specifically, the committee recommended reordering protein sources to lead with:

• Beans, peas, and lentils
• Followed by nuts, seeds, and soy
• With animal proteins listed afterward

This was not ideological. It was evidence-based.

The final government-issued guidelines rejected that recommendation.

Instead, on page 3 under “Prioritize Protein Foods at Every Meal,” the published document leads with animal products:

“Consume a variety of protein foods from animal sources, including eggs, poultry, seafood, and red meat…”

Plant proteins are mentioned only afterward — a direct reversal of the scientific recommendation.

This is not a neutral formatting choice. Order signals priority. And the government chose tradition and industry comfort over science.

Internal Contradictions That Expose the Agenda

The final document goes further, encouraging full-fat dairy and claiming healthy fats are “plentiful in meats, poultry, and eggs” — while simultaneously advising Americans to reduce saturated fat, cholesterol, and trans fats.

These positions cannot coexist logically.

Animal products are the primary dietary source of saturated fat in the U.S. population. The science is unequivocal on this point. To promote them as foundational while warning against their defining nutritional liabilities is not guidance — it’s gaslighting.

Vegan Diets Were Singled Out — and Misrepresented

Most disturbing is how the guidelines treat vegan diets.

Every dietary pattern has potential shortcomings when poorly planned. Yet only vegan diets were singled out and framed as nutritionally suspect — despite decades of data showing the opposite.

Vegans do not suffer unique deficiencies when diets are adequately planned. In fact:

• Vitamin B12: Vegans who supplement routinely achieve higher and more stable serum B12 levels than omnivores, whose intake often depends on animals that were supplemented themselves.
• Other nutrients cited (B2, B6, choline, vitamin A): These are abundant in legumes, greens, whole grains, mushrooms, nuts, seeds, and vegetables — staples of well-planned vegan diets. Deficiency occurs only in extreme caloric restriction, not veganism.

The Academy of Nutrition and Dietetics (January 2025) reaffirmed that appropriately planned vegan diets are:

“Healthful, nutritionally adequate, and beneficial for the prevention and treatment of certain diseases.”

The Health Outcomes the Guidelines Conveniently Omitted

The scientific report — the one the government sidelined — acknowledges outcomes the final guidelines barely mention:

• Lower coronary heart disease risk (up to 26%)
• Lower type 2 diabetes risk
• Lower LDL cholesterol and insulin levels
• Slower biological aging in recent twin studies
• Greater longevity, with some models estimating up to 10 additional years of life expectancy

These are not fringe benefits. These are the leading chronic diseases driving U.S. healthcare collapse — and the final guidelines largely omit them while portraying vegan diets as risky.

The Real Nutritional Crisis Isn’t Veganism

The real failures belong to the standard omnivorous diet:

• Fiber: Less than 10% of Americans meet recommended intake
• Legumes: Over 80% fail to meet targets for beans, peas, and lentils
• Saturated fat: ~75% already exceed recommended intake from meat, poultry, and eggs

So why did the administration move animal foods ahead of plants?

Because the Final Document Is Political — Not Scientific

The truth is simple and admitted outright:

The DGAC report is advisory. The final Dietary Guidelines are approved by USDA and HHS leadership — not scientists.

In other words, the administration is legally allowed to ignore empirical evidence and publish public health policy that contradicts it.

That is exactly what happened.

One Final Truth We Refuse to Say Out Loud

The website is called “Real Food.” Animals are not food any more than humans are. We are animals too. What makes us human is humane behavior — and needless exploitation, violence, and consumption of sentient beings is neither humane nor ethical.

A society that knows better — and chooses otherwise — is not acting rationally. It is acting politically.

And that is the real scandal behind the 2025–2030 Dietary Guidelines for Americans.

  Read more...

The cow, done fairly. On good pasture, the kind we have in the Carolinas, one cow/calf pair needs about 2 acres. (On dry Western rangeland it is 20 to 50-plus acres, so this is beef’s best case.) That pair raises roughly one calf a year. Finished out to about 1,350 lbs, that calf becomes roughly 450 to 500 lbs of boneless beef, around 500,000 calories. At 2,000 calories a day, that feeds one person for about 250 days. Not even a year.

What two acres of beef actually produces

How many calories does one cow produce per acre?

That is being generous, because the 2 acres only get the calf to weaning. Finishing it to slaughter weight takes months of additional feed, usually grain grown on cropland somewhere else. Along the way, the cow and calf consume roughly 20 to 24 million calories of forage and feed to hand us back half a million, a 2 to 3% conversion rate, which matches peer-reviewed estimates for beef.

But don’t cattle eat grass we can’t eat anyway?

Most of those calories are grass humans cannot eat, of course; the FAO’s own feed analysis puts it at 86%, and livestock advocates quote that number constantly. Two things they leave out. First, it is a tally by dry weight, so a tonne of straw counts the same as a tonne of grain; measured in the currency that actually feeds people, feed crops take 36% of global crop calories and 53% of global plant protein. Second, the same paper found livestock eat one third of the world’s cereal production and require 2.8 kg of human-edible feed per kg of boneless ruminant meat (3.2 for pigs and poultry), roughly three pounds of human food in for one pound of meat out, by the most livestock-friendly accounting in print. The full autopsy of that statistic is here. And the real land cost shows up in the feed: by some estimates roughly two-thirds of the calories grown on US cropland are fed to animals rather than people.

Now put the same two acres in plants

How many people can two acres of potatoes feed?

USDA data puts average commercial potato yields around 455 cwt (about 45,000 lbs) per acre, roughly 16 million calories per acre. Two acres: about 32 million calories. That is one person’s calories for about 43 years, or 43 people fed for a year, roughly sixty times what the same land returns when you run it through a steer.

Two honest objections deserve answers: you cannot plant potatoes on the same ground year after year (disease pressure demands a 3-to-4-year rotation), and calories alone are not a diet. Fair enough; rotate with other food crops. It makes the soil healthier and the diet complete.

A 2-acre mix that actually feeds people

Rotating plots each season:

• 0.8 acre field corn, about 13 million calories
• 0.5 acre sweet potatoes, about 4 million calories plus vitamin A (North Carolina grows more sweet potatoes than any state in the country)
• 0.5 acre soybeans, about 3 million calories and roughly 500 lbs of protein (dry beans work too, at lower yields)
• 0.2 acre greens, squash, and vegetables, for vitamins C, K, and folate

Can two acres of crops provide enough protein?

That is about 20 million calories, a year of food for about 27 people, plus over 1,000 lbs of protein between the beans and corn. A person needs about 45 lbs of protein a year; the steer’s beef contains roughly 95 lbs of it, protein for about two people. Even this diversified, rotation-friendly, nutritionally rounded plot outproduces the steer roughly 40x on calories and 10x-plus on protein.

What about B12?

The one nutrient it cannot grow is B12, which is made by bacteria, not cows. Pigs and chickens cannot produce it at all in confinement, so industrially fermented B12 is a standard, regulator-approved additive in their feed; cattle get supplemental cobalt so their gut microbes can make it. Animal agriculture does not avoid the supplement; it just routes it through an animal first. We can take it directly and skip the middleman.

The health evidence, every rung of the ladder

Nutrition science answers big questions with a hierarchy of evidence, and plant-based eating holds up at every level of it.

The mechanism

Saturated fat raises LDL cholesterol, and LDL’s causal role in heart disease is among the most firmly established facts in medicine, proven through drug trials and confirmed by human genetics. Processed meat carries nitrites and heme iron that form carcinogenic compounds in the gut, the biological basis for its cancer classification.

What do randomized controlled trials show?

In Stanford’s identical-twin study (JAMA Network Open, 2023), twins randomized to a vegan diet showed lower LDL cholesterol, fasting insulin, and body weight than their omnivore siblings in just eight weeks, with genetics perfectly controlled. A randomized trial of a low-fat vegan diet in type 2 diabetes (Diabetes Care, 2006) outperformed the standard diabetes diet on glycemic control. Portfolio-diet trials show plant-based patterns cutting LDL by amounts approaching statin therapy. A meta-analysis of randomized trials (Circulation, 2019) found that replacing red meat with legumes, soy, and nuts lowers total and LDL cholesterol. And the Lifestyle Heart Trial (Lancet, 1990) documented measurable regression of coronary artery narrowing on a program centered on a low-fat plant-based diet, alongside exercise and stress management.

What do the large population studies find?

After reviewing more than 800 studies, the WHO’s cancer agency placed processed meat in its highest evidence category for causing cancer and classified red meat as probably carcinogenic. A 2023 meta-analysis of 24 prospective cohorts found that replacing processed meat with nuts, legumes, or whole grains was associated with 23 to 36% lower cardiovascular disease risk. Reviews of protein-substitution studies consistently find lower all-cause and cardiovascular mortality when plant protein replaces animal protein.

Is a vegan diet officially considered safe?

The Academy of Nutrition and Dietetics’ 2025 position paper, built on systematic reviews and in effect through 2032, states that appropriately planned vegetarian and vegan dietary patterns in adults can be nutritionally adequate and can offer long-term health benefits, including improved cardiometabolic outcomes. Plan the diet around whole plant foods (fries and soda are technically plants; they do not count), supplement B12, and it is not just safe; it is protective.

No single study settles a question this big. Convergence across every tier of evidence does, and every tier points the same way.

What happens at scale

Two acres is a thought experiment. Animal agriculture is not; it operates on billions of animals and billions of acres, and the damage radiates outward in every direction the science can measure.

Forests, soil, and land

Beef is the largest single driver of tropical deforestation, linked to roughly 41% of it, more than double the next commodity, and over three-quarters of the world’s soy is grown for animal feed, not tofu. Livestock use 83% of the world’s farmland while supplying 18% of humanity’s calories and 37% of its protein (Science, 2018). The IPCC’s land report lists overgrazing among the leading human causes of soil degradation and desertification, and classic global soil surveys attribute roughly a third of degraded land to it. The result of a century of this arithmetic: farmed animals now make up about 60% of Earth’s mammal biomass. Humans are most of the rest. Wild mammals, every elephant, deer, whale, and mouse combined, are down to roughly 4% (PNAS, 2018).

Water and dead zones

Every summer, fertilizer and manure washing out of Midwest feed-crop country feed an oxygen-starved dead zone in the Gulf of Mexico measured by NOAA in the thousands of square miles, where little can live. Here in North Carolina it is more personal: thousands of open-air hog waste lagoons sit in the eastern part of the state, and hurricanes like Floyd and Florence have flooded them into rivers. Peer-reviewed studies of the communities around industrial hog operations, disproportionately Black, Latino, and Native communities, document higher rates of asthma symptoms, blood pressure, anemia, kidney disease, and infant mortality (NC Medical Journal, 2018). Nearly a third of agriculture’s global freshwater footprint runs through livestock.

The people who do the work

Government audits consistently rank slaughter and processing among the most dangerous jobs in America, with injury rates well above manufacturing overall and a steady log of amputations in OSHA’s severe-injury reports. When COVID hit, a congressional investigation found at least 59,000 infections and 269 deaths among workers at just the five largest meatpacking companies in the pandemic’s first year. The psychological toll is documented too: a 2023 systematic review (Trauma, Violence, & Abuse) found consistently elevated depression, anxiety, and trauma symptoms among slaughterhouse workers, and a peer-reviewed analysis found that counties hosting slaughterhouses see higher violent-crime rates than comparable manufacturing towns.

Pandemics and antibiotic resistance

The CDC estimates three out of four new or emerging infectious diseases in people come from animals. The 2009 flu pandemic emerged from swine. In 2024, H5N1 bird flu spread through US dairy herds and infected dozens of farm and poultry workers; in January 2025, a separate H5N1 case became the first US death. The UN Environment Programme had already named rising demand for animal protein and intensified animal agriculture among the leading drivers of pandemic risk. Meanwhile, roughly three-quarters of the world’s antibiotics are sold for use in farmed animals (Science, 2017), a practice the WHO has formally asked producers to stop in healthy animals, because antibiotic resistance already kills over a million people a year directly (Lancet, 2022) and ranks among the WHO’s top ten threats to global health.

The animals, counted honestly

More than 80 billion land animals are slaughtered every year, about ten per human alive, plus farmed and wild fish numbering from the hundreds of billions into the trillions. The layer industry grinds or gasses 6 to 7 billion male chicks on their first day of life as standard practice. Castration, dehorning, and tail docking are routinely performed without pain relief in the US; most breeding sows spend part of each pregnancy in crates too narrow to turn around in; and the EU’s own food-safety authority classifies the standard CO2 method used to stun pigs as a serious welfare concern, because the gas is painful before it renders them unconscious. Most US states exempt “customary” agricultural practices from their animal-cruelty statutes, so the same act can be a felony in a living room and lawful in a barn.

Don’t crops kill animals too?

For anyone reaching for that card, the argument has been run through peer review and it fails. When an agricultural scientist proposed in 2003 that a pasture-and-ruminant diet might kill fewer animals than a vegan one, a rebuttal in the Journal of Agricultural and Environmental Ethics showed the calculation had compared deaths per hectare instead of per consumer; corrected with the original’s own field-mortality numbers, the plant-based diet came out to roughly 0.3 animal deaths per person per year against 1.5 for the proposed omnivorous one, five times fewer. The full body count, field, slaughterhouse, predator control, and nets, gets its own post. Tally the two categories the way that correction demands, animals killed in the field plus animals killed at the slaughterhouse, per calorie delivered, and every animal product costs more lives than every plant food. Per calorie, chicken and eggs are the most lethal foods on the table by a wide margin, since it takes so many small birds to fill the same plate one steer would. The reason is the arithmetic already on this page: a system that pushes two-thirds of US crop calories through livestock multiplies the harvested acres, and the field deaths with them, then adds more than 80 billion deliberate deaths on top. The plate with meat on it carries more crop deaths, not fewer.

Chemicals and monocrops

The corn-soy monoculture that blankets roughly 180 million US acres exists chiefly to feed animals and fuel tanks, and USGS pesticide data show those two crops are where the herbicide tonnage goes, with insecticide-treated seed on the great majority of corn acres. It shows up in federal law, too: the legal glyphosate ceiling on animal feed runs 1,500 to 4,000 times the ceiling on the fruits and vegetables sold to people (the pesticide argument, followed upstream). Shrink the feed demand and you shrink the monoculture that requires it.

The land argument, without the strawman

But isn’t most grazing land unsuitable for crops?

Yes, most grazing land cannot grow crops, roughly two-thirds of it, by the FAO’s own analysis. But when plants produce 40 to 60 times the calories per acre, we do not need the hillside to grow food. We need it to grow ecosystems. The Science analysis of 38,000 farms found a global shift to plant-based diets would cut agricultural land use by roughly three-quarters, cropland included, not just pasture, while still feeding everyone. And researchers in Nature Sustainability put a number on what that freed land is worth: allowed to return to native forest, grassland, and wetland, it would draw down an estimated 547 billion tonnes of CO2, about sixteen years of global fossil-fuel emissions.

Doesn’t regenerative grazing fix this?

That is the alternative “regenerative” grazing actually competes against, and it loses. Oxford’s two-year Grazed and Confused? assessment concluded that grazing-driven soil carbon gains are small, saturate within decades, reverse when the practice stops, and at their global best offset only 20 to 60% of the grazing animals’ own emissions. For the full autopsy of the regenerative story, including the 270,000-acre Oregon refuge that removed its cattle and logged three decades of peer-reviewed recovery, see Rewilding the Plate. And if some animal foods remain, the numbers have said for over a century that beef is the worst place to put land: even in 1917, USDA data showed dairy and pork each producing about five times the food per acre that beef does.

The receipts go back 109 years

USDA Farmers’ Bulletin 877 (1917) ran this exact comparison: beef, 130,000 calories per acre; sweet potatoes, 2,851,200, a 22x gap. And their beef was fed on cropland corn, silage, and clover, not scrubland. A century of crop breeding has only widened the gap: potato and corn yields are 5 to 8x higher today, while a steer is still a steer.

Add it up

Every nutrient beef provides is available from plants plus the same B12 supplement the industry already feeds its animals, so there is no nutritional need to eat them. Meanwhile the reasons not to keep stacking: a system that discards roughly 97% of the calories fed to beef cattle; that ties up 40 to 60 times the land; that drives deforestation, soil loss, and dead zones; that carries the disease risks documented above; that breeds resistant bacteria and incubates pandemic flu; that injures and traumatizes its own workforce; that sickens the communities downwind and downstream; that kills more field animals, not fewer, and then 80 billion more on purpose; and that, per the largest analysis of global food systems ever conducted (Science, 2018), emits more greenhouse gas per gram of protein than any staple plant food, often 25-fold more. A potato asks for none of that, and nothing has to die for it.

There is no reason to eat animals, and every reason not to.

Related reading in this series

• The Crop Deaths Argument, Counted Honestly: the field-mouse gotcha traced to its academic source, the correction that flipped it five-to-one, and the nine columns of the death ledger the argument never counts.
• The 86% Meme, Debunked: the “livestock eat food we can’t” statistic, examined with the very study it cites, one third of the world’s grain, 3 kg of human-edible feed per kg of meat.
• Rewilding the Plate: what the freed land becomes, the regenerative-grazing story audited, and the peer-reviewed recovery that follows when the cattle come off.
• The Pesticide Argument the Carnivore Movement Can’t Answer: where the chemicals actually land, feed-crop tolerances thousands of times higher than produce, and why liver and fat are the wrong things to recommend.

Sources

• USDA NASS Quick Stats: potato yield statistics, and Prospective Plantings for corn and soybean acreage.
• USDA Farmers’ Bulletin 877 (1917), Human Food from an Acre of Staple Farm Products.
• IARC Monographs Volume 114 (2015), and Bouvard, V., et al. (2015). Carcinogenicity of consumption of red and processed meat. Lancet Oncology, 16(16), 1599–1600.
• Landry, M. J., Ward, C. P., Gardner, C. D., et al. (2023). Cardiometabolic effects of omnivorous vs vegan diets in identical twins. JAMA Network Open, 6(11), e2344457.
• Barnard, N. D., et al. (2006). A low-fat vegan diet improves glycemic control and cardiovascular risk factors in type 2 diabetes. Diabetes Care, 29(8), 1777–1783.
• Jenkins, D. J. A., et al. (2003). Effects of a dietary portfolio of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive protein. JAMA, 290(4), 502–510.
• Guasch-Ferré, M., et al. (2019). Meta-analysis of randomized controlled trials of red meat consumption. Circulation, 139(15), 1828–1845.
• Ornish, D., et al. (1990). Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. The Lancet, 336(8708), 129–133.
• Neuenschwander, M., et al. (2023). Substitution of animal-based with plant-based foods on cardiometabolic health and all-cause mortality. BMC Medicine, 21, 404.
• Raj, S., et al. (2025). Vegetarian dietary patterns for adults: a position paper of the Academy of Nutrition and Dietetics. Journal of the Academy of Nutrition and Dietetics, 125(6), 831–846.
• Poore, J., & Nemecek, T. (2018). Reducing food’s environmental impacts through producers and consumers. Science, 360(6392), 987–992.
• Shepon, A., et al. (2016). Energy and protein feed-to-food conversion efficiencies in the US. Environmental Research Letters, 11(10), 105002.
• Cassidy, E. S., et al. (2013). Redefining agricultural yields: from tonnes to people nourished per hectare. Environmental Research Letters, 8(3), 034015.
• Mottet, A., et al. (2017). Livestock: On our plates or eating at our table?. Global Food Security, 14, 1–8.
• Pendrill, F., et al. (2022). Disentangling the numbers behind agriculture-driven tropical deforestation. Science, 377(6611); and Our World in Data, Drivers of Deforestation.
• World Resources Institute, Global Forest Review.
• IPCC Special Report on Climate Change and Land (2019).
• Bar-On, Y. M., Phillips, R., & Milo, R. (2018). The biomass distribution on Earth. PNAS, 115(25), 6506–6511.
• NOAA Gulf of Mexico hypoxia monitoring.
• Kravchenko, J., et al. (2018). Mortality and health outcomes in North Carolina communities near hog CAFOs. North Carolina Medical Journal, 79(5), 278–288.
• Wing, S., & Johnston, J. (2014). Industrial hog operations in North Carolina disproportionately impact African-Americans, Hispanics and American Indians. UNC Chapel Hill.
• US House Select Subcommittee on the Coronavirus Crisis (2021), meatpacking worker infections and deaths.
• Slade, J., & Alleyne, E. (2023). The psychological impact of slaughterhouse employment: a systematic literature review. Trauma, Violence, & Abuse, 24(2), 429–440.
• Fitzgerald, A. J., Kalof, L., & Dietz, T. (2009). Slaughterhouses and increased crime rates. Organization & Environment, 22(2), 158–184.
• CDC, About zoonotic diseases.
• UNEP (2020), Preventing the Next Pandemic.
• Van Boeckel, T. P., et al. (2017). Reducing antimicrobial use in food animals. Science, 357(6358), 1350–1352.
• WHO guidelines on use of medically important antimicrobials in food-producing animals (2017).
• Murray, C. J. L., et al. (2022). Global burden of bacterial antimicrobial resistance in 2019. The Lancet, 399(10325), 629–655.
• EFSA (2020), Welfare of pigs at slaughter, EFSA Journal, 18(6), e06148; and EFSA (2015), vitamin B12 as a feed additive, EFSA Journal, 13(7), 4112.
• Mekonnen, M. M., & Hoekstra, A. Y. (2012). A global assessment of the water footprint of farm animal products. Ecosystems, 15(3), 401–415.
• USGS Pesticide National Synthesis Project, estimated annual agricultural pesticide use; and US EPA glyphosate tolerances, 40 CFR § 180.364.
• FAO FAOSTAT, livestock primary statistics.
• Hayek, M. N., Harwatt, H., Ripple, W. J., & Mueller, N. D. (2021). The carbon opportunity cost of animal-sourced food production on land. Nature Sustainability, 4, 21–24.
• Garnett, T., Godde, C., et al. (2017). Grazed and Confused? Food Climate Research Network, University of Oxford.
• Davis, S. L. (2003), JAEE 16(4), 387–394; and Matheny, G. (2003), JAEE 16(5), 505–511.

Read more...

A carnivore doctor made the rounds this week with a line that sounds like a mic drop: there are essential amino acids, there are essential fatty acids, and there are no essential sugars. He is right about the chemistry. There really is no such thing as an essential carbohydrate. What he does not mention is that the very institution that decides which nutrients are “essential,” the National Academies, uses that same rulebook to recommend a diet built on plants. So let us take his premise seriously and follow it all the way to the end, because it does not land where he wants it to.

Here is the short version. “Essential” is a narrow word with a specific meaning, and it was never a shopping list for a healthy diet. Once you understand what it actually means, the carnivore talking point deflates, and the case for eating plants gets stronger, not weaker.

What “essential” actually means in nutrition

In nutrition science, a nutrient is “essential” for one reason only: your body cannot build it from scratch, so you have to eat it. That is the whole definition. It is a statement about your internal chemistry, not a verdict on what belongs on your plate.

Why there are no essential carbohydrates

Carbohydrate fails the “essential” test for a boring reason. Your liver can manufacture all the glucose your body strictly needs out of protein and fat, through a process called gluconeogenesis. That is exactly why the National Academies, in their reference report on carbohydrate, fat and protein, wrote that the lower limit of dietary carbohydrate compatible with life is apparently zero, as long as you eat enough protein and fat (National Academies DRI report). Carnivore advocates love that sentence. But read the next thought in the same report: the amount of carbohydrate that provides for optimal health is unknown, and the entire rest of the document goes on to recommend a high-carbohydrate, high-fiber, plant-heavy pattern. “Compatible with life” is the standard you use for a lifeboat, not for a life.

Non-essential does not mean unnecessary

This is the pivot the meat-only crowd never makes. Fiber is also not classified as “essential.” Neither is any single antioxidant or phytochemical. Yet fiber is one of the most protective things you can put in your body, and it exists only in plants. So the label “not essential” tells you nothing about whether something keeps you healthy for fifty years. It only tells you whether its absence causes an immediate deficiency disease like scurvy.

Here is the cleanest way to see how weak the standard is. By the same definition, exercise is not essential either. Your body synthesizes no exercise, you can survive without it, and no deficiency disease follows from skipping the gym. Nobody concludes that exercise is optional for health. The things that actually kill people in wealthy countries, heart disease, cancer, type 2 diabetes, are not deficiency diseases. They are diseases of dietary pattern. “Essential” was the wrong yardstick from the start.

What about creatine, carnitine and taurine

Carnivore advocates often pivot here, arguing that meat supplies compounds plants do not. It does, and it does not matter, for exactly the reason above running in reverse. Creatine, carnitine and taurine are not essential nutrients precisely because your body makes them. If “your body can make it” disqualifies carbohydrate from mattering, it disqualifies these too. You cannot use the essentiality rule in one direction and abandon it in the other. Athletes who want creatine can take it as a cheap vegan powder, and vegans tend to show a larger performance response because they start with lower muscle stores.

The same academy that defines “essential” tells you to eat plants

When the doctor invokes essential amino acids and essential fatty acids, he is borrowing the authority of the group that named them. Fair enough. So let us see what that same group recommends when it moves from “what keeps you alive” to “what keeps you well.”

The carbohydrate recommendation is higher than the fat and protein ceilings

The National Academies set an Acceptable Macronutrient Distribution Range: carbohydrate 45 to 65 percent of calories, fat 20 to 35 percent, protein 10 to 35 percent. Look at the floor of the carbohydrate range. Forty-five percent is higher than the top of both the fat range and the protein range, which each cap out at 35 percent. In plain English, the body that carnivore influencers cite as the referee for “essential” puts carbohydrate at the center of the plate and treats a meat-dominated diet as off the map entirely. They also set a carbohydrate reference intake of 130 grams a day, based on the average amount of glucose the brain burns.

They say keep saturated fat and cholesterol as low as possible

The same report concludes that saturated fat, trans fat and dietary cholesterol have no known beneficial role in preventing chronic disease and are not required at any level in the diet. They set no safe upper number, not because these are harmless at any dose, but because every incremental gram nudges LDL cholesterol up, so the only defensible target is “as low as possible while still eating a nutritionally complete diet.” Notice the symmetry. By the doctor’s own logic, there is no essential saturated fat either. That argument cuts against him at least as hard as it cuts against sugar. We dig into this more in our piece on dietary cholesterol and your heart and our overview of dietary fat.

Can you get all your protein and amino acids from plants

Yes, and the “you have to combine proteins at every meal” idea has been dead in the professional literature for decades.

Do plants have all nine essential amino acids

They do, collectively. Every one of the nine indispensable amino acids is present across plant foods, and the Academy of Nutrition and Dietetics states plainly that a varied plant diet eaten over the course of a day supplies enough protein and amino acids, with no need to consciously pair rice with beans at a single sitting (Academy position paper, 2025). The old “complementary protein” rule from the 1970s was retracted by its own author.

What the real-world data on vegan protein shows

A 2025 analysis of long-term vegans found that a majority comfortably met protein requirements, with the caveat that lysine, concentrated in legumes, was the amino acid most likely to run low in a poorly planned diet (PLOS ONE, 2025). The fix is not a steak. The fix is beans, lentils, tofu, tempeh and peas. This is what the word “well-planned” is doing in every serious recommendation, and it is not a high bar. For the practical side, see our guide to vegan nutrition basics.

Can you build muscle on a vegan diet

When total protein is matched, plant protein builds muscle just as well as animal protein. A controlled resistance-training trial found no difference in muscle or strength gains between habitual vegans and omnivores over twelve weeks at about 1.6 grams of protein per kilogram of bodyweight (Sports Medicine, 2021).

Is plant protein better than animal protein for your health

Adequacy is the low bar. The more interesting question is what happens when you swap one for the other, and the answer favors plants. Harvard’s Nurses’ Health Study and Health Professionals Follow-up Study, following 131,342 people, found that substituting plant protein for animal protein was associated with lower all-cause and cardiovascular mortality (JAMA Internal Medicine, 2016). An NIH-AARP analysis of more than 400,000 participants found that replacing just 3 percent of energy from animal protein with plant protein was associated with roughly 10 percent lower all-cause mortality (JAMA Internal Medicine, 2020). So the protein question does not end at “can you get enough.” It ends at “which source does better,” and it is not the steak.

How to get essential fats without eating animals

Where vegans get omega-3

The two truly essential fatty acids, linoleic acid and alpha-linolenic acid, both come from plants: seeds, nuts and their oils. The longer-chain omega-3s, EPA and DHA, are the ones people are told to worry about. Here is the part the fish industry does not advertise: fish do not make EPA and DHA either. They accumulate it by eating algae. So even the case for eating fish is really a case for algae, the original source. If you have been told seed oils are the villain, our seed oils explainer and our coconut oil versus animal fat comparison are worth a read.

Do vegans really need to worry about DHA and EPA

Less than the supplement marketing implies, and this is honestly unsettled rather than closed. If you are a healthy adult eating a well-planned whole-food vegan diet, there is no good evidence that you need to supplement DHA and EPA, and no good evidence that a small dose harms you. It is a genuinely low-stakes decision, and anyone selling you certainty in either direction is going past the data.

The usual worry is that the body converts the plant omega-3, ALA, into the long-chain forms inefficiently. As a raw tracer percentage that is true, but the percentage is close to the wrong question. Most ALA gets burned for energy, which is not a failure, it is what fat is for, and a tracer figure says nothing about whether total synthesis meets tissue demand. Run the arithmetic instead: PET imaging in living humans puts brain DHA use at under 4 milligrams a day with a half-life in the brain of roughly two and a half years, against an adult fat store measured in tens of grams (Umhau et al., Journal of Lipid Research, 2009). There is also evidence that people who eat no fish convert ALA more actively than fish-eaters do, though the vegan subgroup in that particular cohort was small enough that it should be held loosely (Welch et al., American Journal of Clinical Nutrition, 2010).

Meanwhile the supplement trials keep coming up empty. Large randomized trials have repeatedly failed to show cognitive benefit, including a 2026 trial that proved the DHA actually reached the brain and still found no difference in cognition or brain structure. ALA on its own, by contrast, has its own favorable outcome data for mortality and heart disease (Naghshi et al., BMJ, 2021). So the plant-omega-3 position is not a consolation prize.

What to do in practice is simple. Eat ALA daily: a tablespoon of ground flaxseed or chia covers the baseline, and walnuts and hemp work too. Grind the flax, because whole seeds pass through largely intact. Do not cook everything in high-omega-6 oils, though the “ratio panic” is overblown and omega-6 is not a villain as a class. The strongest case for taking an algal supplement anyway belongs to pregnancy and breastfeeding, infants and young children, older adults with cognitive concerns, and anyone with a confirmed very low omega-3 index. For those cases, or for anyone who simply wants insurance against an open question, roughly 250 mg a day of algal EPA and DHA is the rough consensus among clinicians who recommend it at all, and it sits far below any dose where risks appear. We go through the trials, the biochemistry, the risks and the expert disagreement in full in our complete guide to whether vegans need DHA and EPA.

Plants are where these essential nutrients come from in the first place

Step back and notice what just happened in the two sections above. The essential amino acids and the essential fatty acids both trace back to the same place: plants. This is not a slogan, it is basic biochemistry, and it quietly dismantles the idea that meat is the “primary” source of anything.

Do plants make the essential amino acids

They do. Plants and microbes can build all twenty amino acids from scratch, including the nine that humans and other animals cannot make and therefore have to eat. When a cow eats grass, it is not creating those amino acids, it is harvesting them from the plant and burning most of them for its own life in the process. The protein in a steak is, in a real sense, repackaged plant protein that has already been routed through another animal’s metabolism. Eating the plant means going to the factory. Eating the animal means buying second-hand.

Where do essential fatty acids originate

Same story. Animals, humans included, lack the enzymes needed to build linoleic acid and alpha-linolenic acid, the two essential fatty acids. Plants and algae have those enzymes. So every bit of essential fat in an animal’s body was first synthesized by a plant or by algae and then eaten. We already saw this with the long-chain omega-3s in fish, which come from algae. The pattern is total: for the fats your body cannot make, the origin is always photosynthesis, never the animal.

Why eating plants means going straight to the source

Put those two facts together and the “purity” argument flips completely. Plants are the first-hand, unprocessed source of the essential building blocks of life. The animal is the middleman that took those building blocks, used up most of them, and passed along what was left, bundled with its own saturated fat, cholesterol, and, in the case of processed meat, a Group 1 carcinogen. If you want these nutrients at the source, you eat the source.

Does grass-fed beef fix this

It does not, and the reason is a detail the “grass-fed is different” defense never survives. Fat from cattle, sheep and goats does not reach your plate the way it left the pasture. It passes first through the rumen, a fermentation chamber whose microbes hydrogenate unsaturated fats. Those microbes saturate omega-3s just as readily as omega-6s, which is exactly why even grass-fed beef stays a mediocre omega-3 source. The process is also incomplete, so beef and dairy fat carry a few percent naturally occurring trans fats, and beef fat’s most abundant saturate is palmitic acid, the classic LDL-raiser, much of which the animal synthesizes itself regardless of what it ate. So the middleman does not merely waste energy. It chemically degrades the very fats you were told to eat it for. The chemistry is taken apart in our dietary fat deep dive and in coconut oil versus animal fat.

What do plants have that animal foods simply do not

Fiber is the famous one, but the bigger gap gets almost no airtime: antioxidants. When researchers assayed the total antioxidant content of more than 3,100 foods, beverages and spices, plant foods averaged on the order of 64 times more antioxidants than animal foods, with most animal foods sitting near zero (Carlsen and colleagues, Nutrition Journal, 2010). That headline number is a mean lifted by antioxidant-dense herbs, spices and berries, but the gap holds for ordinary foods too. These are the same compounds tied to lower rates of exactly the diseases that dominate modern mortality. A diet with no plants in it is not just missing fiber. It is missing almost the entire chemical class.

The thermodynamics: why eating animals wastes most of the energy

There is a physics reason the middleman is so costly, and it is the second law of thermodynamics. Every time energy is converted from one form to another, some of it escapes as heat. Nothing in the universe transfers energy at 100 percent efficiency, and living things are no exception. Ecologists have a rule of thumb for this: at each step up a food chain, only about 10 percent of the energy carries over, while roughly 90 percent is burned off and lost.

Follow the chain. A plant captures energy from the sun. A cow eats the plant and loses about 90 percent of that energy just staying alive, standing up and walking around. Then a person eats the cow and loses about 90 percent again. By the time it reaches your plate as beef, you are recovering something on the order of one percent of the energy the plant first captured. Eat the plant directly and you skip an entire tier of that loss. This is not an opinion, it is thermodynamics, and it is the reason raising animals for food swallows so much more land, water and crop than feeding people plants directly. We work through what that means on the ground in the crop deaths math and the “livestock upcycle food we can’t eat” myth.

Fiber is the one nutrient that is only in plants

What fiber actually does for your body

Fiber feeds the bacteria in your gut, which ferment it into short-chain fatty acids like butyrate. Butyrate is the primary fuel for the cells lining your colon. This is a direct mechanical link between eating plants and a healthy gut, and it is a link a meat-only diet severs completely, because animal foods contain zero fiber.

How much fiber lowers your risk of dying

The largest synthesis on this question, commissioned by the World Health Organization, pooled 185 studies and 58 trials. People eating the most fiber had 15 to 30 percent lower all-cause and cardiovascular death, and 16 to 24 percent lower rates of heart disease, stroke, type 2 diabetes and colon cancer, with the sweet spot around 25 to 29 grams a day and benefits still climbing beyond that (The Lancet, 2019). For context, most people in Western countries eat around half of that, and only about seven percent of adults hit the target (American Society for Nutrition). A carnivore diet does not just fall short here. It scores a structural zero on the single dietary factor with the most consistent link to living longer.

Saturated fat, LDL cholesterol and your heart

Is LDL cholesterol really the cause of heart disease

This is as settled as nutrition gets. A consensus panel of the European Atherosclerosis Society reviewed more than two million people across cohort studies, genetic studies and randomized trials and concluded that LDL does not merely correlate with heart disease, it causes it, in a dose-and-time relationship (European Heart Journal, 2017). The genetic evidence is the clincher, because people born with naturally lower LDL have less heart disease across their whole lives, which rules out the “it is just a marker” dodge.

But I heard studies cleared saturated fat

You heard about a couple of meta-analyses that found no overall link, and carnivore advocates cite them constantly. The catch is what saturated fat was being compared against. In those analyses it was often swapped for refined carbohydrates, which are also harmful, so the two washed each other out. When saturated fat is replaced specifically with unsaturated fat from plants, heart events fall. When it is replaced with white bread and sugar, they do not. What you replace it with is the entire question, and the answer points at whole plant foods, not butter and not white flour.

What about the keto study that said high LDL was fine

You may see a 2025 study on “lean mass hyper-responders” passed around as proof that sky-high LDL on a low-carb diet does not build plaque. Be careful with that one. The headline paper, whose selling point was that ApoB does not matter, was retracted by the journal in 2026 over methodology problems the editors judged too large to fix. And even the raw imaging data showed rapid growth of soft plaque over a single year in that very-high-LDL group. It is not the vindication it is billed as.

Do plant-based diets lower cholesterol

Reliably, and by a lot. The Portfolio diet, a plant-based pattern built around soy, nuts, viscous fiber and plant sterols, lowered LDL by around 30 percent under controlled conditions, in the range of a starting statin dose (JAMA, 2003). A clever Stanford study put genetically identical twins on either a healthy vegan or a healthy omnivore diet for eight weeks; the vegan twin ended up with meaningfully lower LDL and fasting insulin than their own sibling, controlling for genes and upbringing (JAMA Network Open, 2023).

Is white meat better than red meat for cholesterol

Not really, and this one surprises people. The APPROACH feeding trial compared red meat, white meat and non-meat plant protein head to head and found red and white meat raised LDL about equally, while plant protein beat both (American Journal of Clinical Nutrition, 2019). Swapping beef for chicken is not the fix people think it is. And in a six-week randomized trial, partly replacing red and processed meat with legumes lowered total and LDL cholesterol in the legume group while both rose in the meat group, with the legume eaters also losing more weight and keeping adequate B12, iodine and iron status (European Journal of Nutrition, 2025).

What the evidence says about red and processed meat

Is processed meat really a carcinogen

The World Health Organization’s cancer agency classifies processed meat as a Group 1 carcinogen, the same confidence category as tobacco smoke for the strength of the evidence, though not the same size of risk, based on its link to colorectal cancer. Red meat sits one rung down as “probably carcinogenic” (WHO Q&A). If you have seen the claim that a single 2019 reanalysis overturned all of this, it did not, and we walk through exactly why in our fact-check of the meat and bowel cancer link.

Does eating meat raise diabetes risk

A federated analysis of nearly two million adults across twenty countries found that a modest daily serving of processed meat was linked to 15 percent higher type 2 diabetes risk, and unprocessed red meat to 10 percent higher (The Lancet Diabetes and Endocrinology, 2024). If weight and blood sugar are your concern, our plant-based weight loss guide covers the practical steps.

What is TMAO and why does it matter

When gut bacteria digest carnitine, which is abundant in red meat, some of them produce a compound called TMAO that accelerates artery disease in animal studies. People who eat meat regularly cultivate the bacteria that make more of it; long-term vegans and vegetarians make far less after the same dose, because their gut ecosystem has shifted (Nature Medicine, 2013).

What happens to people who actually eat this way

Do vegans and vegetarians live longer

In the large Adventist Health Study-2, all vegetarians combined had lower overall death rates than meat-eaters, with an adjusted hazard ratio of 0.88 across 73,308 participants (JAMA Internal Medicine, 2013). Being precise matters here: the vegan-specific mortality estimate of 0.85 trended the same way but did not reach statistical significance on its own, so anyone claiming this study proves vegans outlive everyone is overreading it. Where the vegan signal is robust in that cohort is cardiometabolic. Type 2 diabetes prevalence ran 2.9 percent in vegans against 7.6 percent in non-vegetarians, an adjusted odds ratio of 0.51 even after correcting for body weight, and vegans had markedly less hypertension. We break this cohort down in our Adventist Health article, and look at long-lived populations generally in what the longest-living people eat.

Wait, did not one study show vegetarians had more strokes

Yes, and honesty matters here. The British EPIC-Oxford cohort found vegetarians had about 22 percent less heart disease but around 20 percent more stroke over eighteen years (BMJ, 2019). The likely culprit was poor vitamin B12 status, which is fixable with a cheap supplement, not with a burger. And the finding did not replicate: later cohorts in Taiwan and the United States found vegetarians had the same or lower stroke risk. Even in EPIC-Oxford itself, the much larger drop in heart disease meant vegetarians came out ahead on total cardiovascular disease. This is a footnote in favor of supplementing B12, not an argument for eating animals.

Can a plant-based diet reverse type 2 diabetes

Randomized trials of low-fat vegan diets show larger drops in blood sugar, more weight loss and bigger LDL reductions than conventional diabetes diets (Diabetes Care, 2006). Diet is doing real clinical work here, not just trimming a risk score. Ornish’s Lifestyle Heart Trial went further still, showing measurable regression of coronary artery narrowing on a plant-based program without cholesterol-lowering drugs, while the control group progressed.

Why can you eat until full on plants and still lose weight

Because fiber and water dilute the calories. On a metabolic ward where every calorie was measured, people eating a minimally processed plant-based diet spontaneously ate roughly 689 fewer calories a day than the same people on an animal-based ketogenic diet, with no difference in reported hunger or meal enjoyment. That is the whole argument of our plant-based weight loss guide: you can eat until comfortably full and still run a deficit.

What about the Blue Zones controversy

You may have seen a researcher named Saul Newman argue that the Blue Zone longevity claims rest on shaky record-keeping. He raises fair points about data quality, and we take them seriously rather than hand-waving them away in our Newman fact-check and our broader look at whether the Blue Zones are debunked. The case for plants does not depend on the Blue Zones. It stands on the controlled trials and the mechanistic evidence above.

The honest gaps: B12 and a few nutrients to plan for

Where does B12 actually come from

Here is a fact that reframes the whole debate. B12 is not made by plants, and it is not made by animals either. It is made by bacteria. Animals only carry B12 because microbes in their gut or their feed produce it, and modern farmed livestock are routinely given cobalt and B12 supplements to keep their levels up (review of cobalt and B12 in cattle; Watanabe and Bito, 2018). So most meat-eaters are already taking a B12 supplement, just laundered through an animal first. Vegans cut out the animal and take it directly. Being upfront about this makes the position stronger, not weaker, and we cover the details in our full B12 guide.

What else should a vegan plan for

A well-planned vegan diet needs deliberate sources of B12 (a supplement, non-negotiable), vitamin D, iodine, and attention to iron, zinc and calcium, with daily ALA from ground flax, chia or walnuts and an optional algal omega-3 if you want the insurance. Note the order there: B12, vitamin D and iodine matter far more than the omega-3 question that gets all the airtime. None of this is exotic; it is a short checklist, and the major dietetic associations affirm that a well-planned vegan diet is adequate for adults, with the British Dietetic Association and others extending that to every life stage including childhood. If you are feeding a family this way, start with raising children vegan.

What the carnivore diet can actually show for itself

Is there any real research on the carnivore diet

Almost none. The flagship “study” carnivore advocates cite is an online survey of about 2,000 self-selected enthusiasts who reported feeling great (Lennerz et al., 2021). There was no control group, no lab verification, and an obvious survivorship problem: anyone who felt awful and quit was never counted. There is not a single long-term trial or cohort measuring heart attacks, cancer or death on an all-meat diet. Compare that to the mountain of controlled and cohort data behind plant-based eating and the asymmetry is stark.

What nutrients does an all-meat diet miss

Fiber, entirely. Vitamin C, down to trace amounts in muscle meat. Folate, vitamin E, magnesium, potassium and the entire universe of plant phytochemicals, all low or absent. Advocates point to the Inuit and to Vilhjalmur Stefansson’s year-long all-meat experiment, but traditional Inuit ate fresh and raw organs that supply vitamin C, which is not the same as a plate of cooked steak. We get into that in our piece on raw animal foods. And if the pitch is that plants are full of toxins and antinutrients, we handle that head-on in plant toxins and antinutrients.

Are humans built to be apex predators

Our own genome argues otherwise, and vitamin C is the giveaway. Most mammals manufacture their own vitamin C. Our lineage carries a broken GULO gene and cannot make any, which only makes evolutionary sense if the dietary supply from plants was so dependable that maintaining the enzyme stopped being worth the cost. Add expanded salivary amylase gene copies for digesting starch, a colon adapted to fermenting plant fiber, and 25 functional bitter taste receptors. Obligate carnivores have none of that profile. The species that needs to eat plants to avoid scurvy is not the one built for an all-meat diet.

Why does no medical body recommend a meat-only diet

Step back and notice the asymmetry. Pediatric, cardiology, oncology, diabetes and dietetic organizations across four continents affirm well-planned plant-based eating. Plant-based nutrition has entire evidence-based medical organizations built around it, and the Ornish program for reversing heart disease has been Medicare-covered since 2010. There is no College of Carnivore Medicine, no professional body for keto or paleo, and no Medicare-covered program built on meat. Where omnivorous patterns do earn official endorsement, like the Mediterranean and DASH diets, they earn it precisely for being plant-forward and for telling people to cut red and processed meat. The reason is not a conspiracy. Position papers require evidence, and the evidence has to exist first.

The planet and the animals

How much land and emissions does meat really cost

The largest food-system analysis ever assembled, covering roughly 38,000 farms, found that animal products deliver only about 18 percent of the world’s calories while using around 83 percent of its farmland, and generating well over half of food’s greenhouse emissions. Moving to plant-based eating could cut global farmland use by more than 75 percent and still feed everyone (Poore and Nemecek, Science, 2018). The 2025 update to the EAT-Lancet planetary health work links a plant-rich diet to roughly 27 percent lower risk of early death while fitting inside the planet’s limits (the 2025 EAT-Lancet Commission, The Lancet).

What about the crop deaths argument

The claim that plant farming kills more animals collapses on arithmetic: livestock eat far more crops than they return as food, so feeding animals grows more fields, not fewer. We lay out the numbers in the crop deaths math and the “livestock upcycle food we can’t eat” myth. For the regenerative and rewilding counterpoints, see regenerative farming, rewilding, and the pesticide argument the carnivore movement can’t answer.

When policy and science disagree

Did not the new 2025 guidelines push meat and butter

They leaned that way, and it is worth being honest about it. The 2025 to 2030 Dietary Guidelines for Americans, released in January 2026, kept the cap on saturated fat at under 10 percent of calories but simultaneously raised protein targets and presented butter and beef tallow as acceptable fats. Cardiologists and nutrition scientists pointed out the contradiction immediately. The American College of Cardiology warned the new graphic could be read as an upside-down pyramid with steak and cheese on top and said the evidence does not support promoting butter or tallow (ACC review, 2026). Harvard’s nutrition department called the mixed messaging likely to raise saturated fat intake and cardiovascular risk (Harvard Nutrition Source). The key point: guidelines are policy documents subject to lobbying and politics. The underlying biochemistry of LDL and saturated fat did not change. When the politics and the physiology diverge, follow the physiology.

The bottom line

There are no essential sugars. That is true, and you can concede it cheerfully, because it changes nothing. There are also no essential saturated fats, no essential cholesterol, and no essential red meat. “Essential” only ever meant “your body can’t make it,” never “you should build your diet around it.” The organization that carnivore advocates lean on to define the word turns around and recommends a plate that is mostly plants, rich in fiber, and low in the animal fats that raise LDL. The controlled trials, the genetic evidence and the population data all pull in the same direction. A well-planned vegan diet is not the fragile, deficient thing it gets painted as. It is the pattern the science keeps landing on, and the one thing it asks of you in return is a B12 supplement that most meat-eaters are quietly taking anyway.

Related reading on VeganLinked

• Plants over animals: the full nutritional case
• Do vegans need DHA and EPA?
• The great B12 gamble
• Cholesterol, from the ground up
• Animal fat, plant fat, and the journey to your arteries
• Saturated fat, seed oils and heart disease
• Coconut oil versus animal fat
• Plant toxins and antinutrients
• Raw and cooked
• The best diet to lose weight
• Raising children vegan
• The Adventist Health Studies
• What the longest-living people eat
• Blue Zones, debunked or not
• Saul Newman versus the Blue Zones
• Meat and bowel cancer, fact-checked
• The crop deaths argument
• Do livestock turn food we cannot eat into protein?
• The pesticide argument the carnivore movement cannot answer
• Two acres: regenerative farming
• Rewilding the plate

Sources

• National Academies, Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids
• Reynolds et al., carbohydrate quality and human health, The Lancet, 2019
• American Society for Nutrition, fiber intake in US adults
• Ference et al., LDL cholesterol as a cause of atherosclerotic cardiovascular disease, European Heart Journal, 2017
• Jenkins et al., the Portfolio diet and LDL, JAMA, 2003
• Landry et al., identical-twin vegan versus omnivore trial, JAMA Network Open, 2023
• Barnard et al., low-fat vegan diet and type 2 diabetes, Diabetes Care, 2006
• Retraction notice, KETO-CTA lean mass hyper-responder study, JACC: Advances
• World Health Organization, carcinogenicity of red and processed meat
• Li et al., meat and type 2 diabetes across 1.97 million adults, The Lancet Diabetes and Endocrinology, 2024
• Koeth et al., carnitine, TMAO and atherosclerosis, Nature Medicine, 2013
• Orlich et al., vegetarian patterns and mortality, Adventist Health Study-2, JAMA Internal Medicine, 2013
• Tong et al., heart disease and stroke in vegetarians, EPIC-Oxford, BMJ, 2019
• Hevia-Larrain et al., vegan versus omnivore resistance training, Sports Medicine, 2021
• Song et al., animal and plant protein intake and mortality, JAMA Internal Medicine, 2016
• Huang et al., plant versus animal protein and mortality, NIH-AARP, JAMA Internal Medicine, 2020
• Bergeron et al., APPROACH trial, red meat, white meat and plant protein on LDL, American Journal of Clinical Nutrition, 2019
• Back et al., replacing red and processed meat with legumes, European Journal of Nutrition, 2025
• Welch et al., ALA to long-chain omega-3 conversion in non-fish-eaters, EPIC-Norfolk, American Journal of Clinical Nutrition, 2010
• Umhau et al., brain DHA incorporation and turnover measured by PET, Journal of Lipid Research, 2009
• Naghshi et al., dietary ALA and mortality, dose-response meta-analysis, BMJ, 2021
• Soh et al., protein and amino acid adequacy in vegans, PLOS ONE, 2025
• Academy of Nutrition and Dietetics, position paper on vegetarian dietary patterns, 2025
• Watanabe and Bito, vitamin B12 sources and bioavailability, 2018
• Cobalt and vitamin B12 in dairy cattle nutrition, review
• Lennerz et al., self-reported health among carnivore dieters, 2021
• Poore and Nemecek, environmental impacts of food, Science, 2018
• The EAT-Lancet Commission on healthy, sustainable and just food systems, The Lancet, 2025
• American College of Cardiology, review of the 2025 to 2030 Dietary Guidelines
• Harvard Nutrition Source, on the 2025 to 2030 Dietary Guidelines

Read more...

For decades, the dominant cultural narrative has forced vegans into the defensive hot seat. We are constantly badgered with the same repetitive, exhausting questions: “Where do you get your protein?” “Isn’t a plant-based diet deficient?” “What about our evolutionary history?”

But the scientific landscape has fundamentally transformed.

When you strip away the social conditioning, the defensive cognitive dissonance, and the nitpicking of nutritional epidemiology, a massive biological reality becomes clear.

The burden of proof is no longer on vegans to defend their health. The burden of proof has entirely shifted to those who eat animals.

🏛️ The Science Is Settled: Meat Is a Choice, Not a Necessity

The foundational argument for eating animals has always been biological necessity. If humans must consume animal flesh and secretions to survive and thrive, then the associated violence, environmental destruction, and public health risks could be argued as unfortunate but necessary evils.

However, the world’s largest, most prestigious medical and scientific bodies have officially shattered that premise. The consensus across mainstream medicine is absolute: a well-planned vegan diet is healthy, nutritionally adequate, and actively reduces chronic disease risk for every single stage of life.

Because animal consumption is completely unnecessary for human survival, eating meat shifts from a metabolic requirement to an elective preference.

But it’s not a choice when there are victims involved.

When you eat animals, you are supporting the most violent jobs, the most atrocious living for animals crammed by the thousands into Concentrated Animal Feeding Operations (CAFOs), and the most gravely invasive industry on the planet. Over 97% of the animals people eat come from this brutal, industrialized system.

And the reality is clear: even if you think you’re getting your meat from somewhere more humane, you are still normalizing the exploitation that inevitably results in the most abusive commodification of animals. You cannot claim “personal choice” when that choice requires a victim to suffer and die.

The consensus across mainstream medicine is absolute. A well-planned vegan way of eating is:

• Healthy and nutritionally adequate for every single stage of life—including pregnancy, infancy, childhood, adolescence, and for competitive athletes.
• Clinically proven to reduce the risk of major chronic killers, including atherosclerotic cardiovascular disease, type 2 diabetes, certain cancers, and obesity.

The Institutional Consensus Includes:

• The American Academy of Pediatrics

“Multiple experts have concluded independently that vegan diets can be followed safely by infants and children without compromise of nutrition or growth and with some notable health benefits.”

• The American Heart Association

Whether you’re considering eating less meat or giving it up entirely, the benefits are clear: less risk of disease and improved health and well-being. Consuming less meat decreases the risk of:

• Heart disease.
• Stroke.
• Obesity.
• High blood pressure.
• High cholesterol.
• Type 2 diabetes.
• Many cancers.

• The American Cancer Society

Follow a healthy eating pattern at all ages

• A healthy eating pattern includes:
• Foods that are high in nutrients in amounts that help you get to and stay at a healthy body weight
• A variety of vegetables – dark green, red, and orange, fiber-rich legumes (beans and peas), and others
• Fruits, especially whole fruits in a variety of colors
• Whole grains

• The American Diabetes Association

This plant-forward way of eating is associated with improved health outcomes and decreased risk for a variety of chronic diseases. Specifically, plant-based diets have been linked to a decreased risk for developing kidney disease in people with type 2 diabetes, and decreased risk of mortality in people with chronic kidney disease.

• The Academy of Nutrition and Dietetics

“It is the position of the Academy of Nutrition and Dietetics that, in adults, appropriately planned vegetarian and vegan dietary patterns can be nutritionally adequate and can offer long-term health benefits such as improving several health outcomes associated with cardiometabolic diseases.”

• The National Academy of Sciences

“statistically significant reductions in trimethylamine N-oxide, low-density lipoprotein cholesterol, and body weight” “bean intake can reduce the risk of chronic disease—including cardiovascular disease—as well as overall mortality”

• The 2025 Dietary Guidelines Advisory Committee Scientific Report

Replacing saturated fats with unsaturated fats is associated with reduced risk of cardiovascular disease It also concluded that strong evidence demonstrates that diets lower in saturated fatty acids and cholesterol during childhood results in lower levels of total blood and low-density lipoprotein (LDL) cholesterol throughout childhood, particularly in male children The Committee’s findings reinforce the recommendations in the current (2020-2025) Dietary Guidelines to limit total saturated fat intake to less than 10 percent of calories per day starting at age 2 by replacing it with unsaturated fat, particularly polyunsaturated fats. Evidence indicates that when reducing butter, processed and unprocessed red meat, and dairy, substitution or replacement with a wide range of plant-based food sources, including plant-based protein foods (e.g., beans, peas, and lentils), whole grains, vegetables, or monounsaturated fatty acid (MUFA)- and PUFA-rich vegetable oils and spreads, is associated with cardiovascular disease risk reduction. Fruits, vegetables, and grains are complementary food options between ages 6 and 24 months that are not associated with unfavorable outcomes related to growth or risk of obesity, based on the Committee’s systematic reviews The Committee did, however, identify supporting evidence from food pattern modeling analyses to explore potential modifications to the 2020 HUSS that simultaneously modify at certain calorie levels: (1) Vegetable subgroups, specifically to increase Beans, Peas, and Lentils and decrease Starchy Vegetables while keeping Total Vegetables in the same quantities; and (2) reduce Total Protein Foods by reducing Meat, Poultry and Eggs. The Committee also proposes reorganizing the order of the Protein Foods Group to list Beans, Peas, and Lentils first, followed by Nuts, Seeds, and Soy products

Because a whole-food, plant-based diet is fully capable of optimizing human health, eating animals is an elective lifestyle preference. It is a non-essential behavior.

Hear from World Renowned Doctors and Scientists

⚖️ The Critical Asymmetry of the Debate

When critics of plant-based diets find themselves backed into a corner by the medical consensus, they almost always pivot to a hyper-focus on methodology. They will spend paragraphs trying to hand-wave away massive cohort studies by shouting about “healthy-user bias,” “correlation is not causation,” or “self-reported food questionnaires.”

But notice what they never do.

They never provide a single piece of scientific justification establishing meat as a biological necessity for human health.

Pointing out that nutritional science is complex does not magically turn beef into a health food, nor does it create a biological requirement where none exists. Human metabolism is highly flexible, but flexibility is not a mandate to cause harm.

🛑 The Heavy Price of an Elective Choice

Once an action transitions from a biological necessity to an elective choice, you are legally, logically, and morally required to account for the consequences of that choice.

If you choose to consume animal products when perfectly viable, disease-preventing plant alternatives are readily available, you are actively voting for and supporting:

1. The Real-World Human Cost

The meat industry directly relies on some of the most violent, dangerous, and psychologically damaging occupations on the planet. Slaughterhouse workers suffer from disproportionately high rates of physical injury, severe psychological trauma, and Post-Traumatic Stress Disorder (PTSD) due to the repetitive nature of industrial killing.

2. Atrocious Living and Dying Conditions

Tens of billions of sentient, highly aware animals are subjected to lifelong confinement, mutilation without anesthesia, and industrial slaughter every year. This isn’t a “value system” argument; it is a documented, institutionalized reality of Concentrated Animal Feeding Operations (CAFOs). Watch: Why Humans Became Comfortable With Cruelty | Jim Mason Watch: What They Were Caught Doing | Alka Chanda

3. A Gravely Invasive and Destructive Industry

Animal agriculture is a leading driver of global environmental degradation. It is a wildly inefficient mechanism for feeding a growing population, utilizing roughly 80% of global agricultural land while yielding a meager 20% of global caloric intake. The environmental externalities are staggering:

• Massive deforestation and desertification to grow pesticide-laden monocrops for livestock.
• Millions of gallons of toxic waste stored in anaerobic lagoons that spew ammonia and hydrogen sulfide, poisoning surrounding human communities.
• Severe aquatic dead zones caused by agricultural runoff.
• The catastrophic acceleration of antibiotic-resistant “superbugs” bred in crowded, unhygienic factory farms.

Watch Vegan Footprint: Killing for Food

📢 To Those Eating Animals: Provide the Justification

If you are an omnivore or a proponent of meat-heavy elimination diets, the rules of the debate have fundamentally changed.

We no longer need to spend our time proving that our lifestyle is viable. Decades of peer-reviewed data, clinical trials, and global health guidelines have already done that for us.

Instead, the question belongs to you:

What is your definitive, clinical, scientific justification for consuming animal products, given the immense, measurable destruction it inflicts upon human health, human workers, sentient animals, and the biosphere?

Until you can provide data proving that meat is an absolute requirement for human survival, your critiques of plant-based data are nothing more than a smokescreen. The scale has tipped. The science is settled. The burden of proof is yours.

Share this graphic and post on your social channels to help flip the script on traditional dietary debates! Let us know your thoughts in the comments below.

Read more...

Say the word “vegan” anywhere on the internet and someone will arrive with a combine harvester. Crops kill animals too, the charge goes: mice shredded in the header, rabbits under the tires, voles poisoned in the granary. So your hands aren’t clean either. In its upgraded form, the argument goes further. One large grass-fed steer feeds a person for months and dies once, while a field of wheat kills small animals by the dozen; therefore the beef eater, not the vegan, holds the true least-harm diet.

Unlike most gotchas, this one has an academic pedigree. It was published in a peer-reviewed journal by a serious scientist, and it deserves what every claim in this series gets: a fair hearing, followed by arithmetic. What makes this one unusual is that the fair hearing was already conducted, in the same journal and the same year, and the argument lost. Twenty-odd years later it still circulates as if it had won.

Where the crop deaths argument comes from

Who started the claim that vegans kill more animals?

In 2003, Steven Davis, an animal scientist at Oregon State University, published “The Least Harm Principle May Require That Humans Consume a Diet Containing Large Herbivores, Not a Vegan Diet” in the Journal of Agricultural and Environmental Ethics [1]. His model was straightforward. Drawing on the sparse literature available, he estimated that crop production kills about 15 wild animals per hectare per year, through tillage, harvest, and the predation that follows once cover is stripped away, while ruminant pasture kills about half that, 7.5 per hectare. Run all 120 million hectares of US cropland as crops for a vegan food supply and you get roughly 1.8 billion wild animal deaths; run half of it as pasture for grazing ruminants and the total drops to about 1.35 billion. His conclusion: a diet built around large grass-fed herbivores kills fewer animals than a vegan one.

Credit where it is due. Davis engaged the least-harm principle seriously instead of dismissing it, published where animal ethicists could answer him, and put his numbers on the table. The numbers are the interesting part, because they are the part that flipped.

The correction that flipped it, published the same year

Do vegans really kill more animals than meat eaters?

Gaverick Matheny’s rebuttal, “Least Harm: A Defense of Vegetarianism from Steven Davis’s Omnivorous Proposal,” ran in the same journal that same year [2]. It identified the load-bearing error in a single sentence. Davis compared deaths per hectare when the question is deaths per consumer, and a hectare of crops feeds far more people than a hectare of pasture.

The gap is not subtle. A hectare of cropland can produce on the order of 1,000 kg of plant protein; producing the same protein from grass-fed beef takes roughly ten hectares. So even granting Davis his own field-death rates, 15 per crop hectare and 7.5 per pasture hectare, the per-person accounting inverts his conclusion. Measured against the 20 kg of protein an adult needs in a year, Matheny found a vegan diet kills about 0.3 wild animals, a lacto-vegetarian diet about 0.39, and Davis’s proposed omnivorous diet about 1.5: a vegan diet works out to roughly 0.3 wild animal deaths per person per year, against 1.5 for Davis’s proposed diet. Five times fewer [2]. Matheny noted two further problems, namely that Davis counted deaths but ignored the quality of the lives lost, and ignored the animals prevented from existing; but the denominator error alone settles the arithmetic. Andy Lamey’s follow-up analysis added that Davis’s field-death rates were themselves likely overestimates, in part because they counted animals taken by predators after harvest as though the harvester had killed them [3].

The argument’s best case, the single large pasture-raised ruminant that its defenders always reach for, loses five to one using its own numbers. That correction has now been in print for more than two decades. The Australian remake of the argument, published by Mike Archer in 2011, in which grain kills more animals than rangeland beef, built partly on poisoning campaigns during mouse-plague years, recycles the same denominator problem with shakier data, and met the same fate in the literature [4].

How many animals actually die in crop harvest?

Nobody has a reliable count, and that includes both sides

Here is the part both sides should sit with, because intellectual honesty is the price of admission to this series. In 2018, Bob Fischer and Andy Lamey published “Field Deaths in Plant Agriculture” in the same journal [4], and their central finding is uncomfortable for everyone’s infographics: nobody has a defensible count.

The canonical figures trace back to a handful of small studies, extrapolated far beyond what they measured. The most famous of them, a study of wood mice in English grain fields, actually found that the harvest itself killed very few mice; the population crash afterward came mostly from predation once cover was stripped away, and from mice simply dispersing to the field margins [5]. Animals flee combines more often than they die under them. Mortality varies enormously by crop, region, season, and method, and deaths from predation-after-harvest raise a genuine question of attribution, since the owl was going to eat some mouse somewhere. Fischer and Lamey’s conclusion was blunt: field-death totals could be substantially lower than the numbers circulating in these debates, and confident per-hectare figures, whether deployed by carnivore memes or by vegan counter-memes, outrun the evidence [4].

So let us not pretend to decimal places. The honest position is that the field-death numbers are soft. The conclusion, however, is not, because it never depended on the counts. It depends on the multiplier.

The multiplier: whoever eats animals orders the field twice

Why eating meat means more crops are harvested, not fewer

Livestock do not photosynthesize. Whatever dies in the production of crops, a food system that routes crops through animals multiplies it, because trophic conversion is a losing game at every step. The FAO’s own feed analysis found livestock consume one third of global cereal production and require about 3 kg of human-edible feed per kg of meat [6]. Feed crops absorb 36% of the world’s crop calories and 53% of its plant protein, and only about 12% of those feed calories ever return to a human plate [7]. US beef converts 3% of its feed calories into meat [8]. By some estimates, roughly two-thirds of the calories grown on US cropland are fed to animals rather than people [7]. However many field animals die per harvested acre, the animal-product plate commissions more harvested acres per person than the plant-based one, and then adds the slaughterhouse on top.

Which foods cause the most animal deaths per calorie?

Put the two categories together, field deaths plus slaughter deaths per calorie delivered, and the ledger orders itself. The most-cited tabulation, built on USDA production data within the Davis and Matheny framework [9], puts chicken around 251 deaths per million calories, eggs around 92, beef around 29, pork around 18, and milk around 5, against roughly 2 for plant foods. Treat the decimals as estimates; the ordering is robust, because it follows from body size and feed conversion, not from any particular mouse study.

Notice the ranking, because it embarrasses everyone’s intuitions. Chicken is the deadliest food on the table, over a hundred times the plant baseline, precisely because the bodies are small and each meal costs a life-fraction of many animals plus the feed acres that raised them. Beef, the crop-deaths argument’s champion, performs best among the meats for the same reason in reverse, one large body feeding many meals, and it still runs roughly fifteen times the plant baseline per calorie, and five times per capita in Matheny’s correction. The argument’s best case is the plant eater’s worst case, and it isn’t close.

How many animals does one person’s diet kill in a year?

Per-calorie tables are useful for comparing foods, but most people want the personal number. Here it is, and it is the most concrete way to state the whole argument.

Counting only land animals, government slaughter data divided by population puts the average American’s diet at roughly 25 to 30 animals a year, and the composition surprises people: nearly all of them are chickens, because a chicken yields so few meals. Pigs, cattle, and turkeys together account for barely one animal a year. A vegan’s figure for this column is zero. Confidence: solid, since it comes straight from USDA and FAO slaughter counts.

Now add the animals almost every version of this debate forgets. Once fish and shellfish are converted from tonnage into individual animals, the most detailed US per-capita compilation puts the average American’s annual total at roughly 400 to 580 animals, the large majority of them fish and shrimp, with wild fish caught for fishmeal and killed as bycatch making up much of the remainder. A separate global analysis reaches the same conclusion by a different route. A vegan’s slaughter figure is still zero. Confidence: medium. These are estimates built on government tonnage data using published mean-weight methods, not headcounts, and the range is wide for good reason.

So the honest headline depends entirely on whether fish are counted, and the difference is more than tenfold. State it both ways or not at all: about 25 to 30 land animals a year, or roughly 400 to 580 animals counting aquatic life, against essentially zero for someone eating plants. Note what that does to the arithmetic of the gotcha. The crop-deaths argument is a dispute about a handful of field mice. It is being deployed against a difference of several hundred animals a year.

And that table is the conservative comparison, because it counts only two of the nine columns below: field deaths and direct slaughter. It leaves out predator control, disease culls, on-farm mortality, pesticide wildlife kills, bycatch, dead zones, and habitat loss, every one of which falls on the animal side. The honest per-calorie picture is worse for meat than even these numbers show; the table simply happens to be the one slice that can be drawn apples-to-apples.

The full ledger the gotcha never adds up

Field deaths and slaughter are only two columns. The real account has at least nine, and the crop-deaths argument goes silent on most of them. What follows is that ledger, built wherever possible on government data and peer-reviewed journals rather than advocacy tallies. Where a number is soft, it says so, because a ledger that flags its own weak entries is the one that survives a hostile reader.

A word on what this can and cannot be. Nobody has published a single integrated body count across all of these columns, and the units do not cleanly add; a slaughtered chicken, a bycaught dolphin, and a dead zone measured in square miles are not the same kind of thing. So this is not one headline number. It is a survey of every column, each with its best available figure, showing that on all of them the animal-farming plate carries more.

1. Direct slaughter (solid)

More than 80 billion land animals are killed for food every year, about ten for every human alive [19]. To that, add farmed fish: an estimated 78 to 171 billion finfish killed annually, plus farmed shrimp and prawns in the hundreds of billions [13]. The layer industry culls 6 to 7 billion male chicks on their first day of life as standard practice, a soft but widely accepted industry-derived figure.

2. Wild capture (solid on the totals, softer per species)

Between 1.1 and 2.2 trillion wild fish are pulled from the ocean in an average year [13], roughly half of them ground into fishmeal and oil to feed other farmed animals, most of it to farmed fish but a substantial share to weaning pigs and day-old chicks [12]; so a chicken sandwich and a salmon fillet both carry ocean deaths upstream. Sharks are killed at a rate that rose from about 76 to 80 million a year between 2012 and 2019, roughly a quarter of them threatened species [20]. And a landmark analysis put global marine bycatch, the dolphins, turtles, seabirds, and unwanted fish discarded dead, at about 40% of the reported catch under its definition of unused or unmanaged catch [11]; the figure is definition-dependent and higher than the FAO’s narrower discard estimate, so treat it as an upper-range marker rather than a precise share.

3. Predator and “pest” control (solid, government data)

The USDA’s Wildlife Services program kills wild animals largely on behalf of livestock producers. In 2023 it killed 375,045 native animals, including 68,562 coyotes, 24,603 beavers, 430 black bears, 469 bobcats, 305 gray wolves, and 235 mountain lions [10]; counting invasive species, the program’s total runs to roughly 1.4 million animals a year. An earlier accounting that folds in native birds killed under blanket depredation orders reaches about 2.6 million total and 1.5 million native for 2018. It is the same program cast in a wider net, and the scope difference is why the two years look so far apart. A 2024 investigation found roughly 11,000 of these animals were killed at locations with no recorded livestock damage at all [21]. This entire column is charged to the animal-farming plate and counted by almost no one. The beaver number stings most: those are the animals that rebuild the wetlands and water tables plant agriculture would leave alone.

4. Disease-control depopulation (solid, government data)

When avian flu or swine fever reaches a facility, the response is mass killing. The current US H5N1 outbreak has killed over 185 million birds since February 2022, across all fifty states [22]; the 2014 to 2015 outbreak took another 50.5 million. In China, African swine fever killed or culled an estimated 143 million pigs in 2018 and 2019, cutting the national herd by roughly 40% [23]. That total blends disease deaths, government culling, and panic liquidation, so it is not all regulators killing healthy animals; but every death was a product of raising pigs by the hundreds of millions in the first place. These culls exist only because animals are farmed; the column does not exist on the plant side.

5. On-farm mortality before slaughter (soft, and invisible in every official count)

A large share of farmed animals die before they ever reach the kill floor, and those deaths appear in no slaughter statistic. In the US alone, over 500 million broiler chickens die on the farm each year, from disease, from heart failure driven by the breed’s engineered growth rate, and from culling, before the survivors are trucked to slaughter [24]. That total comes from an advocacy compilation of USDA production data rather than a peer-reviewed source, so treat it as indicative; peer-reviewed cohort work puts first-week mortality near 1.4% and later mortality around 3%. Piglet pre-weaning death, transport losses, and dead-on-arrival birds add still more. The number is soft because the industry does not publish a clean total; but it is real, it is enormous, and it is counted nowhere.

6. Pesticide wildlife kills (soft, and mostly charged to feed crops)

Agricultural pesticides in the US were estimated decades ago to kill on the order of 67 million birds a year [25], a figure later authors consider conservative, and one that predates the neonicotinoid seed treatments now standard on corn and soy. Because roughly two-thirds of US crop calories are grown to feed livestock, most of that acreage, and most of the chemical load on it, belongs to animal agriculture. This is the connection the companion pesticide post documents in federal law: the legal glyphosate ceiling on animal feed runs 1,500 to 4,000 times the ceiling on the produce sold to people [26]. The bird-kill figure is old and imprecise, so flag it as such; the direction is not in doubt.

7. Aquatic dead zones and manure spills (habitat destruction, not a clean body count)

Every summer, fertilizer and manure washing out of the Midwest feed-crop belt feed an oxygen-starved dead zone in the Gulf of Mexico, 4,402 square miles in 2025, and 6,705 in 2024 [27]. NOAA measures the area of habitat rendered uninhabitable, not a body count, so this belongs in the ledger as documented ecosystem destruction rather than a number of corpses. The acute kills are more countable: in 1995 a single ruptured hog lagoon sent some 25 million gallons of waste into North Carolina’s New River, and hurricanes Floyd and Florence later flooded lagoons across the state’s east into its rivers [14]. Present the spills as documented events; treat any specific fish-kill tally as a reported estimate unless a state agency published it.

8. Habitat and biodiversity loss (solid on drivers, directional on the toll)

This is the largest column and the hardest to reduce to a number. Cattle pasture drives about 41% of tropical deforestation, and agriculture as a whole drives 90 to 99% of it [15]. Farming is the single biggest documented threat to species on the IUCN Red List. And the cumulative result is a biosphere in which farmed animals make up roughly 60% of all mammal biomass, while every wild mammal on Earth combined, every elephant, whale, deer, and mouse, is down to about 4% [16]. You cannot put a clean death count on habitat conversion; but every displaced, starved, and never-born wild animal behind those figures sits on the animal-farming side of the ledger.

9. Human deaths (solid, and rarely counted as part of the toll at all)

Food-related air pollution, much of it ammonia off manure and feed fertilizer, is estimated to cause about 12,700 US deaths a year attributable to animal-based foods, against 3,200 for plant-based [28]; by that accounting a gram of beef protein carries roughly the air-pollution mortality of 200 grams of protein from beans or grains. Antibiotic resistance was linked to 1.27 million deaths directly and 4.95 million associated worldwide in 2019 [29], and livestock consume about 73% of the world’s antibiotics [30]; though use-share is not death-share, and the post should not multiply the two. And in the pandemic’s first year, at least 269 workers died at the five largest US meatpackers [31]. Diet-attributable cancer and heart-disease deaths belong to the health post, and are contested enough to keep separate.

Nine columns. Plant agriculture registers meaningfully in exactly one of them, field deaths, and posts the lowest number there too. Every other column is animal agriculture’s alone.

Intention matters, just not the way they think

Is an accidental death the same as a deliberate one?

Suppose the numbers were even. They are not, but suppose. There would still be a difference between deaths that occur as a mitigable side effect of harvesting food and deaths that are the product. Field mortality is what Fischer and Lamey call a historically contingent problem: it falls with better harvest technology, timing, and practice, and above all it falls with acreage, which is exactly what a plant-based food system delivers: roughly 75% less agricultural land to feed everyone [17]. The kill floor cannot be mitigated out of animal farming; it is the point of animal farming. Veganism never claimed to be a zero-harm diet. The standard has always been avoiding exploitation and killing as far as possible and practicable [18], and every practicable improvement points the same direction: fewer acres, fewer blades, no slaughter line.

Take the argument seriously and it converts you

What would you eat if you actually minimized animal deaths?

Here is the test that separates a sincere argument from a conversation-ender. Anyone who genuinely ranked their diet by animal deaths per calorie would act on the nine-column ledger above. They would quit chicken and eggs first, since those top every count; treat beef as damage control at fifteen times the plant baseline; and land, finally, on grains, fruits, vegetables, and legumes, the bottom of every column, including the one the argument is named after. Followed all the way down, the crop-deaths argument is an argument for veganism with extra steps. The people who deploy it almost never eat as though they believe it, which tells you what it is actually for. It was never a way of counting deaths. It is a way of ending conversations about them.

The bottom line

The field-death numbers are soft; the structure is not. Whatever the true count per harvested acre, the plate with animal products on it commissions more harvested acres, then adds more than 80 billion slaughtered land animals a year, over a trillion wild-caught fish, the predator-control kills, the disease culls in the hundreds of millions, the half-billion broilers that die before slaughter, the bycatch and the fishmeal, the lagoons and the dead zones, a wild world compressed to 4% of mammal biomass, and thousands of human deaths from the air it fouls and the drugs it burns through. The plant-based plate carries the smallest share of the one column the gotcha points at, and essentially none of the other eight.

For the acre-by-acre production math, see Two Acres. For the feed statistic this argument usually travels with, see The 86% Meme, Debunked. For the chemical load those extra acres carry, and the farmworkers who absorb it, see The Pesticide Argument. For what the freed land becomes, see Rewilding the Plate.

Related reading in this series

• Two Acres: What Beef Produces vs. What Plants Can: the acre-by-acre math behind the multiplier, roughly 500,000 calories of beef versus 20 to 32 million calories of plants from the same land.
• The 86% Meme, Debunked: the “livestock eat food we can’t” statistic examined with the very study it cites.
• The Pesticide Argument the Carnivore Movement Can’t Answer: the other half of the crop-harm objection, where pesticides are actually applied, and who absorbs them.
• Rewilding the Plate: the regenerative-grazing story audited, and what returns when the cattle come off.

References

1. Davis, S. L. (2003). The least harm principle may require that humans consume a diet containing large herbivores, not a vegan diet. Journal of Agricultural and Environmental Ethics, 16(4), 387–394.
2. Matheny, G. (2003). Least harm: a defense of vegetarianism from Steven Davis’s omnivorous proposal. Journal of Agricultural and Environmental Ethics, 16(5), 505–511.
3. Lamey, A. (2007). Food fight! Davis versus Regan on the ethics of eating beef. Journal of Social Philosophy, 38(2), 331–348.
4. Fischer, B., & Lamey, A. (2018). Field deaths in plant agriculture. Journal of Agricultural and Environmental Ethics, 31(4), 409–428.
5. Tew, T. E., & Macdonald, D. W. (1993). The effects of harvest on arable wood mice. Biological Conservation, 65(3), 279–283.
6. Mottet, A., de Haan, C., Falcucci, A., Tempio, G., Opio, C., & Gerber, P. (2017). Livestock: On our plates or eating at our table? A new analysis of the feed/food debate. Global Food Security, 14, 1–8.
7. Cassidy, E. S., West, P. C., Gerber, J. S., & Foley, J. A. (2013). Redefining agricultural yields: from tonnes to people nourished per hectare. Environmental Research Letters, 8(3), 034015.
8. Shepon, A., Eshel, G., Noor, E., & Milo, R. (2016). Energy and protein feed-to-food conversion efficiencies in the US and potential food security gains from dietary changes. Environmental Research Letters, 11(10), 105002.
9. Animal Visuals (2009). Number of animals killed to produce one million calories in eight food categories. An advocacy-site tabulation built on USDA data within the Davis and Matheny framework; treat point estimates as approximate.
10. Center for Biological Diversity summary of USDA APHIS Wildlife Services data for 2023; see also the annual APHIS Program Data Report tables.
11. Davies, R. W. D., Cripps, S. J., Nickson, A., & Porter, G. (2009). Defining and estimating global marine fisheries bycatch. Marine Policy, 33(4), 661–672.
12. Cashion, T., Le Manach, F., Zeller, D., & Pauly, D. (2017). Most fish destined for fishmeal production are food-grade fish. Fish and Fisheries, 18(5), 837–844.
13. Mood, A., & Brooke, P. Estimates of global numbers of fish caught annually (fishcount.org.uk); peer-reviewed version: Estimating global numbers of fishes caught from the wild annually from 2000 to 2019, Animal Welfare, 33, e6.
14. Burkholder, J., et al. Impacts of industrial animal production on rivers and estuaries, American Scientist (documenting the 1995 New River hog lagoon failure); and NOAA Gulf of Mexico hypoxia monitoring.
15. Pendrill, F., et al. (2019). Agricultural and forestry trade drives large share of tropical deforestation emissions. Global Environmental Change, 56, 1–10, as compiled in Our World in Data, “Drivers of Deforestation”; and Pendrill, F., et al. (2022). Disentangling the numbers behind agriculture-driven tropical deforestation. Science, 377(6611). The 40% versus 41% variation across sources is rounding within the same dataset.
16. Bar-On, Y. M., Phillips, R., & Milo, R. (2018). The biomass distribution on Earth. PNAS, 115(25), 6506–6511.
17. Poore, J., & Nemecek, T. (2018). Reducing food’s environmental impacts through producers and consumers. Science, 360(6392), 987–992.
18. The Vegan Society. Definition of veganism (established 1944).
19. FAO. FAOSTAT: livestock primary, annual global slaughter statistics.
20. Worm, B., et al. (2024). Global shark fishing mortality still rising despite widespread regulatory change. Science, 383(6679), 225–230.
21. NPR Investigations (2024). How a federal program kills wildlife for private interests.
22. USDA APHIS. Highly Pathogenic Avian Influenza detections in commercial and backyard flocks, February 2022 to present.
23. USDA Economic Research Service. How China’s African swine fever outbreaks affected global pork markets; and FAO situation updates on African swine fever in Asia and the Pacific.
24. Animal Equality (2024). Mortality in broiler chickens, an advocacy compilation of USDA NASS production data; peer-reviewed mortality rates in The sustainability gap of US broiler chicken production, Royal Society Open Science.
25. Pimentel, D., et al. (1992). Environmental and economic costs of pesticide use. BioScience, 42(10), 750–760.
26. US EPA. Glyphosate tolerances, 40 CFR § 180.364; see the companion post, The Pesticide Argument the Carnivore Movement Can’t Answer.
27. NOAA. Gulf of Mexico hypoxic zone monitoring: 4,402 square miles in 2025; 6,705 square miles in 2024.
28. Domingo, N. G. G., et al. (2021). Air-quality-related health damages of food. PNAS, 118(20), e2013637118.
29. Murray, C. J. L., et al. (2022). Global burden of bacterial antimicrobial resistance in 2019. The Lancet, 399(10325), 629–655.
30. Van Boeckel, T. P., et al. (2017). Reducing antimicrobial use in food animals. Science, 357(6358), 1350–1352. Livestock account for about 73% of global antimicrobial use; use-share is not death-share.
31. US House Select Subcommittee on the Coronavirus Crisis (2021). Data on coronavirus infections and deaths among meatpacking workers.
32. Sethu, H. How many animals does a vegetarian save? Counting Animals. A detailed independent compilation of USDA and FAO production data into per-capita death counts, including fish and shellfish; an advocacy-adjacent analysis rather than a peer-reviewed one, so treat the totals as well-documented estimates.
33. Animal Charity Evaluators. Effects of Diet Choices (2021). A global estimate of animals spared per plant-based year, built on FAO data across farmed land vertebrates, farmed fishes, and wild-caught fishes.
34. Archer, M. (2011). Ordering the vegetarian meal? There’s more animal blood on your hands. The Conversation. The Australian restatement of the least-harm argument, addressed in Fischer and Lamey (2018).

Read more...

Ingredients

• One can of black beans
• 1/2 cup creamy peanut butter
• 1/2 cup maple syrup
• 1/4 cup almond flour
• 1/4 cup cocoa powder
• 1 and 1/2 tbsp unsweetened applesauce. I usually just blend a chopped p apple with some lemon
• 1/4 teaspoon baking powder
• 1/4 teaspoon baking soda
• 3/4 of a cup vegan chocolate chips, at least
• 1/3 cup chopped walnuts

Bake at 350 for 25 minutes. Let cool for an hour unless you like them really hot and gooey like I do. Just look at that gooey, magnificent chocolatey delicacy! Read more...

Health information, not medical advice. Supplementation decisions belong between you and a clinician who understands plant-based nutrition. This is a long, layered piece: the two-minute answer is right here at the top, and it gets progressively more technical the further you read.

If you want the whole omega-3 picture in one place, this walks from the plain-language verdict down through the biochemistry, the trials, the risks, the measurement problem, and the expert disagreement. Jump to any section using the links, or just start reading.

The short answer: do vegans need to supplement DHA and EPA?

Is a DHA supplement necessary on a healthy vegan diet?

If you are a healthy adult eating a well-planned whole-food vegan diet, there is no good evidence that you need to supplement DHA/EPA. There is also no good evidence that a small dose hurts you. This is a genuinely low-stakes decision, and anyone telling you otherwise in either direction is going beyond the data.

That is the honest summary. Here is the slightly longer version.

What about omega-3 for the heart on a vegan diet?

You almost certainly do not need a supplement for your heart. The big long-chain omega-3 supplement trials failed to move hard outcomes. Plant omega-3 (ALA) has its own favorable outcome data. And vegans already have excellent cardiovascular profiles to begin with. For the bigger picture on dietary fat and heart health, see our deeper dives on dietary fat, cholesterol, and seed oils.

What about omega-3 for the brain on a vegan diet?

Nobody actually knows, in either direction. The supplement trials also failed for the brain, including a 2026 trial that proved the DHA reached the brain and still found no benefit. But no one has ever run a proper long-term study on the brains of healthy long-term vegans, so “we do not know” is the truthful answer, not “you are fine” and not “you are at risk.”

What actually matters more than DHA?

B12 (non-negotiable), vitamin D, iodine, sleep, exercise, not smoking, blood pressure, and overall diet quality. Every serious dementia-prevention framework ranks these far above omega-3. If you are spending worry-energy on DHA while your B12 status is unverified, you have the priorities backwards. Start with our full guide to vitamin B12, then the broader vegan nutrition overview.

Who has the strongest case for taking DHA anyway?

• Pregnant and breastfeeding women. This is the one place near-consensus exists.
• Infants and young children. See our guide to raising children vegan.
• APOE4 carriers. Altered brain DHA handling.
• People with a confirmed very low Omega-3 Index (under about 3%). Not because harm is proven, but because you are outside the range that all the reassuring vegan data was built on.
• Known FADS1/FADS2 variants that impair conversion.
• Older adults with cognitive concerns or family history.

If you skip the supplement, what should you do instead?

1. Eat ALA daily. One tablespoon of ground flaxseed or chia covers the baseline recommendation. Walnuts and hemp seeds work too. Flax and chia have the best ALA-to-omega-6 ratio.
2. Grind your flax. Whole flaxseeds pass through you largely intact.
3. Do not drown everything in omega-6 oils. ALA and linoleic acid (LA) compete for the same enzymes. You do not need to fear omega-6; it is essential and the “ratio panic” is overblown (more below). But cooking everything in sunflower, corn, safflower, or generic “vegetable oil” works against you. Olive, avocado, or no added oil are better defaults. More in our seed oils article.
4. Do not megadose ALA either. There is some evidence that very high ALA can compete with itself at a later step in the pathway.* Moderate and consistent beats heroic.
5. Lock down B12. This one is not optional, and it is plausibly implicated in the one piece of evidence that makes vegans look bad on brain outcomes (see the cohort section).

* Am I eating too much ALA from whole foods? Almost certainly not. This concern is aimed at people taking flaxseed oil by the tablespoon on top of everything else, not at people eating whole foods. A typical day looks like: about 1.6 to 2.4 g ALA from a tablespoon of ground flax, about 2.5 g from a palm of walnuts, and roughly 0.3 to 0.8 g of background ALA from greens, soy, beans, and other seeds. That totals roughly 4 to 5.5 g/day, about 3 to 4 times the Adequate Intake (1.6 g/day for men, 1.1 g/day for women): comfortably normal, and squarely inside the studied range. By contrast, the trials that raised the self-competition question generally used 6 to 15 g/day, usually as oil, and Klein 2025 gave about 3 g/day of supplemental flaxseed oil on top of habitual diet for nine months and still raised erythrocyte DHA by 12.8 to 26%, with conversion unaffected by LA or AA status. The mechanism is real but the threshold is high: both Delta-6 desaturase steps use the FADS2 enzyme, so at some intake the first step could theoretically crowd the last, but that is a hypothesis, and the more likely explanation for the DHA plateau is simply that the final synthesis step is tightly regulated regardless of how much substrate you supply. The genuinely useful takeaway: if you eat 4 to 5 g of ALA daily and your index is still low, substrate is not your bottleneck, conversion is, and eating more flax will not fix that. That points at FADS genotype as the thing worth checking. One optional refinement: walnuts carry roughly 4 times more linoleic acid than ALA (about 10.7 g LA vs 2.5 g ALA per ounce) while flax and chia run the other way, so if you ever wanted to nudge the balance you could weight slightly more toward flax and chia. But since LA status did not affect conversion in Klein 2025, file this under marginal, not “fix this,” and do not let it scare you off walnuts, which are an excellent whole food.

Why the answer is “probably not”

Has the study that would settle this actually been done?

No, and this is the uncomfortable truth both sides skip. The study that would settle it: a long-duration randomized trial of algal DHA versus placebo, in long-term well-planned vegans, with cognitive and brain-imaging endpoints, controlling for B12 status (homocysteine and methylmalonic acid, not just serum B12), sleep, exercise, social engagement, and APOE genotype. That study does not exist and probably never will, because dementia has a decades-long latency and the event rate is too low to power a feasible trial. So everything below is inference from adjacent evidence.

What do the randomized controlled trials show? (Tier 1, strongest evidence)

This is the strongest evidence type, and it points the same direction over and over: no cognitive benefit from supplementation in the populations tested.

• AREDS2 (Chew, JAMA 2015): about 3,500 older adults, 1 g/day EPA+DHA, 5 years. No effect on cognition.
• VITAL-Cog (Kang, 2022): 3,424 plus 794 adults 60+, 840 mg/day, 2 to 3 years. Essentially flat (annual global cognitive change of about minus 0.01 SD units, 95% CI minus 0.02 to 0.003).
• MAPT (Andrieu, Lancet Neurology 2017): 3 years, 800 mg DHA plus 225 mg EPA. Omega-3 alone was null.
• ADCS-DHA (Quinn, JAMA 2010): 402 patients with mild-to-moderate Alzheimer’s, 2 g/day DHA, 18 months. No slowing.
• OPAL (Dangour, AJCN 2010): 867 healthy older UK adults, 2 years. Null.
• Shinto (JAMA Network Open 2024): 102 adults 75+ with low omega-3 status, about 1.65 g/day, 3 years. Primary endpoint null, with a possible signal only in APOE4 carriers.
• PreventE4 (Yassine, eBioMedicine 2026): 365 at-risk older adults, low DHA intake, 47% APOE4 carriers, 2 g/day algal DHA, 2 years. Null despite proven brain delivery.

Why is the PreventE4 trial such a big deal?

Because it closed a loophole every earlier null trial left open. The old objection was: sure, blood levels went up, but did the DHA actually reach the brain? PreventE4 made cerebrospinal fluid target engagement its primary endpoint. The CSF DHA/AA ratio rose with a between-group difference of 0.19 (95% CI 0.16 to 0.21), roughly a 17% increase in CSF DHA, and it happened equally in APOE4 carriers and non-carriers. Participants’ Omega-3 Index climbed from about 4.9% to about 11%.

And then nothing happened. Hippocampal volume did not differ (left minus 0.26%, 95% CI minus 1.43 to 0.91; right plus 0.32%, minus 0.47 to 1.11). Cortical thickness did not differ. The RBANS cognitive score showed a treatment difference of 0.22 (95% CI minus 1.05 to 1.49); both groups improved about 2.7 points from practice effects. The authors concluded that despite biochemical target engagement, no differences in cognition or brain structure were observed over 24 months, and that future research should prioritize brain DHA metabolism over further supplementation trials. Lead investigator Hussein Yassine put it plainly in press coverage: their results do not support fish oil supplements as a preventive measure against Alzheimer’s.

Note what this did to the “dietary DHA cannot reach the brain” argument. The supplement was standard algal DHA, triglyceride form, and it demonstrably raised CNS DHA in humans. The strong version of the transport-form claim (that triglyceride DHA has virtually no effect on brain DHA) was mouse data. In humans at 2 g/day it clearly does something. The corollary cuts the other way too: if DHA got into the CNS and nothing happened, that is an argument against supplementation being useful, not an argument that dietary DHA is inert.

Do vegans and vegetarians have worse heart or brain outcomes? (Tier 2, cohorts)

On the cardiovascular side, the picture is reassuring with one asterisk. In EPIC-Oxford (Tong, BMJ 2019), vegetarians had 22% lower ischaemic heart disease (HR 0.78, 95% CI 0.70 to 0.87) but 20% higher total stroke (HR 1.20, 1.02 to 1.40), about 3 extra strokes per 1,000 over 10 years, mostly haemorrhagic. The associations did not attenuate after adjustment, and the authors did not attribute the stroke signal to omega-3; low LDL and total cholesterol is the leading candidate, with B12 also in the mix. Crucially, this stroke signal is not replicated: the Taiwanese Tzu Chi cohort (Chiu, Neurology 2020) found the opposite, with lower stroke across the board (overall HR 0.52), and a 2022 systematic review found no overall vegetarian-stroke association.

On the brain side, the cohorts mostly favor plant-based eating. Tzu Chi (Tsai, 2022) found reduced dementia risk in vegetarians (HR 0.671). Adventist Health Study-1 (Giem 1993) found meat eaters roughly twice as likely to have dementia. Our full breakdown of that population is in the Adventist Health Studies article, and the broader longevity picture is in what the longest-living people eat and our look at whether the Blue Zones are debunked.

What about the Adventist Health Study-2 neurological finding?

This is the strongest single piece of evidence against the “vegans are fine” position, and it should not be waved away. AHS-2 (Abris, AJCN 2024) found vegetarians and vegans had lower all-cause, cardiac, diabetes, kidney, and infectious mortality, but among those aged 85 and older, higher neurological mortality: 17% higher stroke death, 13% higher dementia death, 37% higher Parkinson’s death.

How to weigh it honestly. It is a large, well-conducted cohort of health-conscious, mostly B12-aware, long-term plant-eaters, which makes it the most relevant data we have. Against that: it is observational, the confidence intervals are wide, it is a subgroup analysis in the very old, and one detail undercuts the DHA interpretation directly. Pesco-vegetarians, who eat fish and therefore get preformed DHA, showed no protective benefit either. If low DHA were driving the signal, the fish-eaters should have been protected. They were not. The B12 alternative explanation is live: AHS-2’s own data showed 15.2% of vegans had B12 intake below the RDA, serum B12 is a poor marker compared to MMA and homocysteine, and supplement potency is genuinely unreliable. The bottom line: a real signal that deserves attention, the best argument the other side has, and one that does not cleanly implicate DHA.

Has anyone actually studied vegan brains directly? (Tier 3)

Once. Lloyd-Wright 2005 (UK doctoral work in the EPIC cohort) found vegan men with low EPA/DHA showed no cognitive impairment across 13 cognitive tests and the same mean IQ by National Adult Reading Test compared to omnivores, with B12 status not changing the result. A small companion RCT gave vegan men 200 mg DHA for 12 weeks; one of 13 tests reached significance, which the authors did not consider meaningful. It is a dissertation, small, and short, but it is the closest thing that exists to the study people keep saying does not exist, and it points the reassuring direction. One lead author (Sanders) later concluded that current evidence is insufficient to warrant advising vegans and vegetarians to supplement with EPA or DHA for cardiovascular prevention.

What does ALA do on its own? (Tier 4)

This gets skipped constantly and should not be. Naghshi (BMJ 2021), a dose-response meta-analysis, found dietary ALA associated with reduced all-cause, cardiovascular, and coronary heart disease mortality. Zamroziewicz (2017/2018) found that ALA and its immediate downstream products SDA and ETA, but not EPA or DHA, were linked to fluid intelligence and gray matter preservation in the frontoparietal cortex. The Cardiovascular Health Study (Virtanen 2013) linked plasma ALA to less brain atrophy and dietary ALA to fewer white matter abnormalities. And the 2026 UK Biobank early-onset dementia analysis (Sala-Vila, Harris et al.; n=217,122) found higher plasma total omega-3 associated with 36% lower early-onset dementia risk (HR 0.64, 95% CI 0.47 to 0.87), with the “non-DHA” fraction (ALA + EPA + DPA) significant across more quintiles than DHA alone. That last one supports the plant-omega-3 position, but it is observational, comes from a group that sells omega-3 testing, and “non-DHA” is a subtraction estimate.

What do the major health authorities say? (Tier 5)

The 2025 Academy of Nutrition and Dietetics vegetarian position paper states that EPA and DHA can be obtained through vegan algal supplements if desired or recommended, although this may not be a requirement for health or disease prevention, and notes that isotope studies suggest ALA conversion may be sufficient to maintain adequate DHA through plant-sourced ALA alone. WHO’s 2019 dementia-risk guidelines do not recommend polyunsaturated fatty acid supplementation. The Alzheimer’s Association says there is insufficient evidence to recommend DHA for treating or preventing Alzheimer’s. And the Yu 2020 umbrella review (243 observational studies, 153 trials) placed EPA/DHA in the weak-evidence bucket while education, homocysteine, midlife hypertension, diabetes, and obesity got the strongest ratings.

The biochemistry: how ALA becomes DHA

Why are only ALA and LA “essential” fatty acids?

Your body cannot make omega-3s from scratch. It lacks the Delta-12 and Delta-15 desaturase enzymes that place double bonds at the n-3 and n-6 positions. Plants have them; we do not. That is why ALA (18:3n-3) and LA (18:2n-6) are the only two truly essential fatty acids, a point we cover more broadly in our guide to essential nutrients. Everything else in the omega-3 family is built from ALA through alternating desaturation (adding a double bond) and elongation (adding two carbons).

What is the actual ALA to DHA conversion pathway?

The chain runs: ALA (18:3n-3), then via Delta-6 desaturase (FADS2) to SDA (stearidonic acid, 18:4n-3), then via elongase (ELOVL5) to ETA (20:4n-3), then via Delta-5 desaturase (FADS1) to EPA (20:5n-3), then via elongase (ELOVL2) to DPA n-3 (22:5n-3), then elongated to a 24-carbon intermediate, desaturated again by FADS2, and finally shortened back to 22 carbons by one round of peroxisomal beta-oxidation to yield DHA (22:6n-3).

Two features of this pathway matter enormously and almost never get explained. First, the last step is genuinely strange: DHA is not made by simply elongating DPA. The chain is elongated to 24 carbons, desaturated again by the same Delta-6 enzyme used at step one, then shortened back to 22 carbons by beta-oxidation in the peroxisome. This is the Sprecher pathway, and it is why DHA synthesis is inefficient and tightly regulated: it requires routing the molecule through a different organelle. Second, the same FADS2 enzyme is needed twice, at step one and at the final desaturation. That is the mechanistic basis for the claim that too much ALA might impair DHA synthesis by saturating the enzyme at the first step.

Does omega-6 block omega-3 conversion?

Less than the internet suggests. The omega-6 pathway (LA to GLA to DGLA to arachidonic acid) runs on the same FADS1, FADS2, and ELOVL5 enzymes, which is the basis of the concern. But conversion of LA to AA is genuinely poor, roughly 0.2 to 2.7%. Rett and Whelan 2011 found that decreasing dietary LA by up to 90% showed no significant correlation with tissue AA (p=0.39), and increasing LA up to six-fold likewise showed no correlation (p=0.72): tissue AA is remarkably resistant to dietary LA. The FAO 2010 report concluded there is no rationale for a specific omega-6-to-3 or LA-to-ALA ratio recommendation if absolute intakes are adequate. That said, some studies do show that lowering LA while keeping ALA up raises EPA and DHA. The honest position: the ratio is not the crisis it is made out to be, absolute ALA intake matters more, and there is no reason to fear omega-6 as a class. Get your ALA, do not cook everything in high-LA oils, stop there. We cover this fully in our seed oils and dietary fat articles.

Is it true that only 1% of ALA converts to DHA?

The figure is real but it is used misleadingly. The actual numbers (Burdge, Williams, Baker): men convert ALA to EPA at roughly 8% and to DHA anywhere from under 0.1% to 4%; women convert to EPA up to about 21% and to DHA up to about 9%, a difference attributed to estrogen and progesterone promoting synthesis while testosterone inhibits it. Infants convert better than adults.

Why the percentage is the wrong question: a tracer percentage tells you what fraction of one labeled dose ended up as DHA. Most ALA is beta-oxidized for energy or stored, which is not a failure, it is what fat is for. The percentage says nothing about whether total-body synthesis meets tissue demand. Domenichiello, Kitson and Bazinet (2015) argued formally that oral tracer studies likely underestimate synthesis, because labeled ALA is absorbed into tissues and adipose before reaching the liver where conversion happens; using steady-state infusion methods, synthesis rates come out substantially higher. Do the arithmetic: brain DHA consumption is about 3.8 mg/day, adult adipose stores an estimated 20 to 50 g of DHA, so even a fraction-of-a-percent conversion rate produces enough. The valid counter-argument: producing enough for the brain’s minimum is not the same as optimal tissue status, and “adequate” here means absence of overt deficiency, which is a low bar.

Do the intermediates matter, or can I just take DHA directly?

This is the “whole pie versus one slice” question. The fair version: some intermediates probably do have independent value. SDA and ETA showed up associated with fluid intelligence and gray matter preservation while EPA and DHA did not, in Zamroziewicz’s work. DPA n-3 is increasingly studied on its own. ALA has independent outcome data. That is a decent argument for making sure you eat ALA rather than relying on a capsule alone. It is not a good argument that taking preformed DHA is harmful. And the mirror claim, that “taking DHA only gives you DHA,” is false in practice, though the mechanism is subtler than usually stated. Compound-specific isotope analysis (Metherel and Bazinet, 2019) shows DHA supplementation reliably raises plasma EPA, but the extra EPA does not carry the isotopic fingerprint of the DHA, so it is not true backward conversion; the likeliest explanation is that DHA spares existing EPA from being used up. Either way, taking DHA does raise your EPA level, so the strict one-way-pathway argument does not deliver the conclusion its advocates want.

Where DHA lives in the body and what it does

Where is DHA concentrated in the body?

DHA is not spread evenly. In most tissues it is a minor component, roughly 1 to 5% of total fatty acids. Then there are two spectacular exceptions. In the rod outer segments of the retina, DHA is 50 to 60% of total fatty acid content, the highest concentration anywhere in the body; the photoreceptor disk membranes are 80 to 90% phospholipid with only 8 to 10% cholesterol, unusually fluid, and DHA is what makes that possible. In the brain, DHA is over 90% of the omega-3 PUFAs and roughly 10 to 20% of total brain lipids, concentrated in gray matter and within that in synaptic membranes. Arachidonic acid, an omega-6, is the second most abundant brain PUFA at about 20% of neuronal fatty acids, worth remembering when people frame this as omega-3 good, omega-6 bad. Sperm is also DHA-rich. Heart, liver, muscle, adipose, and blood cells are relatively low.

How much DHA does the brain actually use per day?

This reframes everything. Umhau et al. (Journal of Lipid Research 2009) used carbon-11 DHA PET imaging in living humans and found net brain DHA incorporation of 3.8 plus or minus 1.7 mg/day, whole-brain turnover of 0.076% per day, and a half-life of DHA in the human brain of approximately 2.5 years. Your brain consumes under 4 mg of DHA per day, and what is in there now takes years to turn over. Against 20 to 50 g stored in adipose, that reservoir alone could supply the brain for well over a decade before accounting for any ongoing synthesis.

Is DHA recycled in the body?

Yes, and heavily. When a membrane phospholipid is remodeled, phospholipase A2 cleaves DHA from the sn-2 position; it is not excreted but re-esterified back into new phospholipids by acyltransferases. It is a largely closed local loop. In the retina it is tighter still: shed photoreceptor outer segments are phagocytosed by retinal pigment epithelium cells, and the DHA-rich phospholipids from degraded phagosomes are immediately reintroduced to the pathway. Practically, this means tissue DHA is not something you burn through and must constantly replace from diet, and it means any trial shorter than a few years is fighting the kinetics: one honest explanation for why the RCTs keep coming up null.

What does DHA actually do in cells?

Six double bonds make DHA extraordinarily flexible, which lets membranes deform (important for vesicle fusion at synapses and disk stacking in photoreceptors) and changes how embedded receptors and ion channels behave, underlying effects on neurotransmission including dopamine signaling. DHA is also the precursor to specialized pro-resolving mediators (neuroprotectin D1, D-series resolvins, maresins) that actively resolve inflammation rather than merely suppressing it, and to the endocannabinoid synaptamide. It also acts as a ligand for nuclear transcription factors including PPARs.

How does DHA get into the brain, and does the LPC form matter?

This is where the interesting fight is. Mfsd2a is a transporter at the blood-brain barrier that carries DHA in lysophosphatidylcholine (LPC) form, not as free fatty acid (Nguyen, Nature 2014); humans with MFSD2A loss-of-function mutations develop lethal microcephaly. So LPC transport is unambiguously real and important, especially in development. The strong claim (Sugasini and Subbaiah) is that dietary LPC-DHA enriches brain DHA efficiently whereas free and triglyceride DHA at equivalent dose have virtually no effect, which is the basis for arguing that standard algal or fish-oil supplements cannot help the brain.

Why the strong claim does not hold up as stated: it is mouse and rat data, with no completed human cognitive-outcome trial (two are underway, expected around 2029). Bazinet’s own lab found that plasma non-esterified DHA, not LPC-DHA, is the major pool supplying the rat brain, and a 2025 study from that group failed to reproduce brain DHA increases from dietary phospholipid carriers in mice. PreventE4 raised human CSF DHA using triglyceride-form algal DHA. And the specific claim that plasma LPC-DHA can be as high in someone with a low Omega-3 Index as a high one has no human confirmation; the closest human data (Ly et al., J Lipid Res 2023) found plasma LPC 22:6 did not rise with high-dose supplementation even as the index did, which is consistent with LPC-DHA being separately regulated but does not establish the cross-sectional claim. What survives is narrower but still important: the Omega-3 Index is not a validated proxy for brain DHA, so inferring brain deficiency from a low index is unproven.

Why do Alzheimer’s autopsies not consistently show lower brain DHA?

If low DHA drove Alzheimer’s, AD brains should be DHA-depleted. Some studies find that; many do not, including Fraser 2010, the largest such autopsy series, and Skinner 1993 and Igarashi 2011. Cunnane (2013) called this out directly and concluded the literature on DHA in human plasma or brain neither supports nor refutes the putative link between lower DHA and AD risk. Possible explanations, none settled: low blood DHA in AD may be a consequence of neurodegeneration rather than a cause (reverse causation); blood DHA may simply mark overall diet quality; what is depleted may be oxidized DHA products rather than total DHA (AD autopsies do consistently show elevated lipid peroxidation, and lower magnesium); or brain DHA may be defended homeostatically until very late. What AD autopsies do consistently show points toward protecting DHA from oxidation rather than adding more.

Does cerebrospinal fluid reflect brain DHA?

Partly. CSF is a much better proxy than red blood cells because it is on the brain side of the barrier, but it is not neuronal membrane content. DHA in CSF has crossed into the CNS compartment and is equilibrating there; it does not prove incorporation into synaptic phospholipids. So a critic can fairly say PreventE4 showed CNS delivery without showing tissue incorporation. That is a legitimate gap and the last remaining refuge for the transport-form argument, but it is much narrower than “dietary DHA cannot reach the brain,” and it does not rescue the case for supplementing, since the null functional result is the harder problem for that side.

The measurement problem: the Omega-3 Index

What is the Omega-3 Index actually validated for?

The Omega-3 Index is the percentage of EPA plus DHA in red blood cell membrane fatty acids. It was developed by William S. Harris, who also founded OmegaQuant, which sells the test, worth knowing when you weigh claims about what it means. Proposed categories: desirable 8 to 12%, intermediate 4 to 8%, undesirable below 4%. It is genuinely validated for one thing: correlation with coronary heart disease mortality across cohort studies, and it predicts all-cause mortality in the Framingham Offspring cohort. As an epidemiological risk marker in general, largely omnivorous populations, it has real support.

Does the Omega-3 Index reflect brain DHA?

It has never been validated for that, and this is the crucial distinction that gets silently elided. Evidence it is a poor brain proxy: the AHS-2 brain sub-study (Loong, 2023) found EPA, DHA, and the index correlated with white matter volume but none correlated with hippocampal volume or frontal cortical thickness, and across six cognitive tests DHA had no association with any. Add the autopsy literature above, the rodent finding that brain DHA correlated negatively with the erythrocyte index, and an infant study where RBC DHA explained only about a quarter of the variance in brain growth. So a “brain target” of 6% or 7% is an extrapolation from a cardiovascular-risk biomarker, not a validated threshold.

Is there really a 4.4 threshold for brain aging?

This number gets quoted a lot, so it deserves a close look. It comes from Tan, Harris, Beiser et al. (Neurology 2012), a cross-sectional analysis of 1,575 Framingham participants averaging 67 years old. The authors did use the word “threshold,” and they did report that the lowest quartile had smaller total brain volume and worse performance on several cognitive tests, with the discussion framing it as roughly two years of extra structural brain aging.

But three things are usually left out. First, the design is a single snapshot, and the authors state plainly that it “precluded examination of the relationship between RBC omega-3 PUFA levels and the development of clinical dementia.” It cannot show a rate of change or establish causation. Second, RBC DHA was not linearly related to any brain MRI measure; the “threshold” is a dichotomous comparison of the bottom quartile against the top three, and 4.4 percent is simply the 25th-percentile cutoff of that sample, not a breakpoint derived from any spline or curve-fitting analysis. Third, it has never been independently replicated as a threshold value. The pattern that the lowest quartile fares worst does recur (Satizabal 2022, Pottala 2014), but no study has validated 4.4 as an inflection point.

Two popular embellishments are also worth flagging. The vivid “about a teaspoon less brain matter” line appears in neither the paper nor the associated press release. And the claim that an elderly brain “ends up smaller than that of two- to three-year-old children” misstates Courchesne 2000, which documents decline from an adolescent peak, not shrinkage below toddler volume.

None of this means the association is fake. It means the number is a quartile cutoff from one cross-sectional study, not a validated danger line, and it should not be used to make anyone panic.

What is a typical Omega-3 Index for a vegan?

Sarter 2014/2015 found vegans averaged 3.7%, with 64% below 4% and 27% below 3%, on an average ALA intake of 3.4 g/day (more than double the AI). Notably the vegan index was close to omnivore US soldiers at about 3.5%, so “low vegan” is really “low non-fish-eater.” AHS-2 found vegan values around 4%, with ALA and LA intake not associated with the index. EPIC-Norfolk (Welch 2010) found the precursor-product ratio was higher in non-fish-eaters, evidence of upregulated conversion, though the vegan subgroup there was very small.

Can eating ALA raise your Omega-3 Index?

Modestly and slowly. Lane 2022 found ALA supplementation does not reliably increase the index while algal EPA/DHA does. But Klein 2025 (nine months, about 3 g/day flaxseed oil) did raise erythrocyte DHA by 12.8 to 26% across all diet groups, and Ezaki 1999 saw no change at 3 months but plus 21% at 10 months. So duration matters, and ALA can move DHA, just usually not enough to take a 2% index to 8%.

What Omega-3 Index should a vegan aim for?

Here is the honest impasse: because RBC concentrations are not a validated brain marker, there is no defensible brain-based target. Eight percent has cohort support for cardiovascular risk in omnivores. Any brain target you see quoted is an extrapolation. And for vegans specifically, no target has ever been tested against outcomes. If you set a numeric goal, be clear you are doing it on cardiovascular or precautionary grounds, not because a brain threshold exists.

How much EPA and DHA does it take to raise the index?

Walker et al. (AJCN 2019) pooled 14 trials into a predictive model. To reach a mean index of 8% in 13 weeks with triglyceride-form EPA+DHA: about 2,200 mg/day from a 2% baseline, about 1,500 mg from a 4% baseline, about 750 mg from a 6% baseline. Ethyl-ester form needs roughly 40% more. Two caveats: these doses far exceed the 250 mg most plant-based clinicians recommend and sit above the 1,000 to 1,500 mg threshold where atrial fibrillation risk becomes measurable, and the lifetime dose needed to maintain 8% is much lower (average Japanese intake is 800 to 1,000 mg/day with a population mean index above 8%). Flock 2013 found dose alone explained 68% of the response variance; body weight, baseline, age, activity, and sex together explained only about 10%. In Sarter’s vegan trial, about 250 mg/day moved the group mean from 3.1% to 4.8% in four months.

Risks of supplementing

Does omega-3 supplementation cause atrial fibrillation?

This is the best-established risk, and it is dose-dependent. Gencer et al. (Circulation 2021), pooling 7 cardiovascular outcome trials, found omega-3 increased AF risk by about 25% overall, dose-dependent, with roughly 1,000 mg/day raising risk about 12%. A 2025 updated meta-analysis (34 trials, about 114,326 participants) found significant AF risk only in the high-cardiovascular-risk, high-dose (over 1,500 mg/day) subgroup, with a pooled OR of 1.48 (95% CI 1.21 to 1.81) and an absolute risk difference of 0.8%; the other three subgroups were non-significant. UK Biobank (Chen 2024) found regular fish-oil use in people without cardiovascular disease showed no primary prevention benefit and increased AF and stroke risk. Practical translation: at 200 to 500 mg/day you are well below where this shows up; at 2,000-plus mg/day, the dose you would need to take a 2% index to 8%, you are in the zone where it does.

Does omega-3 help if you already have atrial fibrillation?

Commonly believed, but the evidence does not support it. A meta-analysis of 4 RCTs (1,268 participants) found no effect on preventing AF recurrence (HR 1.13, 95% CI 0.96 to 1.33), and the FORWARD trial (1 g/day for a year in persistent AF) did not reduce recurrent AF and found no difference in the composite of mortality, stroke, MI, embolism, heart failure, or severe bleeding. The one supportive study was a small post-ablation nested case-control, not a randomized trial. There is an observational, EPA-specific signal that EPA correlates inversely with ischemic brain infarcts in AF patients, but AF patients have been underrepresented in trials.

Does DHA raise LDL cholesterol?

Yes, modestly, and it is DHA-specific. Theobald 2004 found about 0.7 g/day DHA raised LDL cholesterol by 7%. Bernstein 2012 (algal oil meta-analysis) found roughly 1.68 g/day raised LDL by about 0.23 mmol/L (7 to 8%) while lowering triglycerides and raising HDL. The ComparED study (2018) found high-dose DHA had more pronounced LDL effects than high-dose EPA. This is partly offset because DHA shifts LDL toward larger, less atherogenic particles, but if your cardiovascular strategy is built around keeping LDL low, as many whole-food plant-based approaches are, it is a genuine consideration at higher doses. See our cholesterol article.

Does omega-3 or fish oil cause prostate cancer?

Contested, not resolved, but weaker than the headlines. Brasky 2013 (SELECT trial) found plasma phospholipid DHA associated with total, low-grade, and advanced prostate cancer (advanced RR 1.48, 95% CI 1.10 to 1.99). But nobody in SELECT was given fish oil, the team collected no fish-intake data, and Brasky himself said they did not know the source. The absolute levels were low (cases 3.66% vs controls 3.52%), not people on fish oil. Reverse causation is plausible: 30 to 40% of men who developed cancer had baseline PSA over 3 versus 7% of controls, and pre-cancerous tissue shows altered fatty acid metabolism. Japanese men eat about 8 times more fish with an index of 8 to 10% yet historically have far lower prostate cancer mortality. The same team had previously reported fish-oil supplement use was not associated with increased risk, dietary intake studies generally show no effect on incidence, and fish intake is associated with reduced prostate cancer mortality. Honest status: an unreplicated biomarker association with a credible reverse-causation explanation, not something that should drive your decision.

Does DHA oxidize and cause harm?

DHA has six double bonds, making it the most peroxidizable fatty acid in the body, and fish oil supplements have been documented to contain oxidized lipids, sometimes exceeding set limits. But the leap from “DHA is peroxidizable” to “DHA supplementation causes oxidative harm in vivo” is mechanistic speculation, the same species of overreach the anti-supplement camp rightly criticizes elsewhere. Practical response: buy fresh, buy from companies that test, store properly, do not megadose, and eat an antioxidant-rich diet, which vegans generally do. High doses (900 mg EPA plus 600 mg DHA and up) may also reduce immune function and increase bleeding time, per the Institute of Medicine and FDA, but neither is relevant at 250 to 500 mg.

Practical supplement questions

Does it matter that algae DHA is triglyceride form and not phospholipid?

Yes for absorption, no for the brain-transport argument. Forms you will encounter: natural triglyceride (in fish and algae), ethyl ester (cheaper, made during concentration), re-esterified triglyceride, free fatty acid, phospholipid (krill, not vegan), and monoacylglycerol. Triglyceride absorbs meaningfully better than ethyl ester, by roughly 50 to 70% in reviews; Schuchardt 2011 found krill phospholipid highest, then triglyceride, then ethyl ester, though differences were not statistically significant. Most algal oil is already triglyceride form, which is good, and Arterburn showed algal-oil capsules and cooked salmon were nutritionally equivalent DHA sources. The important non-answer: better systemic absorption is not the same as better brain delivery, and PreventE4 already showed triglyceride-form algal DHA reaches human CSF. So buy triglyceride over ethyl ester for value, but do not imagine you are solving the brain-transport question by doing so. The tip that matters more than form: take it with a fat-containing meal.

Is sublingual or liquid omega-3 better absorbed than capsules?

No good evidence supports sublingual for omega-3. Fatty acids are absorbed via the intestinal lymphatic system in chylomicrons, which is not something the sublingual mucosa does well for a large lipid; sublingual makes sense for B12, not for omega-3. Liquid versus capsule makes no meaningful absorption difference, though liquid is usually cheaper per milligram and oxidizes faster once opened, so refrigerate it.

How do I compare omega-3 products with different doses and ratios?

Read the EPA plus DHA content, not the oil content. A “1,000 mg algae oil” softgel might contain 300 mg of actual EPA plus DHA. Find milligrams of EPA plus DHA per serving, divide by capsules per serving to get milligrams per capsule, then divide product price by total milligrams to get cost per milligram, and compare that across products. On ratio: most algal products are DHA-dominant, some DHA-only, a few balanced. Given that several lines of evidence point toward EPA being at least as relevant as DHA and that DHA is the one raising LDL and carrying the AF and prostate signals, a product containing both is probably a more sensible default than DHA-only. Look for third-party testing, a published oxidation value, minimal fillers, opaque packaging, and a reasonable expiry. Tank-grown algae avoids the ocean-contaminant problem entirely, which is the strongest argument for algae over fish oil regardless of everything else here.

How much DHA and EPA should a vegan take?

There is no evidence-based answer, because the outcome that would define “enough” has never been measured in this population. What exists: common plant-based clinical recommendations of 200 to 300 mg/day (Brenda Davis; Michael Klaper; Jack Norris suggests 200 to 300 mg every 2 to 3 days as insurance); Greger at 250 mg/day; Joel Kahn at 250 mg combined; Fuhrman aiming for an index of 7 to 9% (which means well over 1 g/day from a low baseline); and about 200 mg preformed DHA as the general pregnancy guideline. The rough consensus among people who recommend it at all is 250 mg/day, which sits well below any risk threshold, and Sarter showed that dose moves a vegan group mean from 3.1% to 4.8% in four months.

The expert disagreement: eight voices, one unsettled question

This topic has attracted an unusually wide spread of credentialed opinion, and laying the voices side by side is more useful than picking a favorite. They disagree mostly about two things: how far above a 4 percent omega-3 index you should try to get, and whether observational biomarker data or randomized outcome data should drive the decision. Money matters too, so conflicts are flagged throughout. Here is the spectrum, from “it’s a scam” to “vegans clearly need it”: Geoff Palmer, Tim Radak, the Sherzais, Michael Greger, Philip Calder, Simon Hill, Rhonda Patrick, and Joel Fuhrman.

Is Geoff Palmer right that DHA supplements are a scam?

Geoff Palmer, founder of the plant-based sports-nutrition brand Clean Machine, has become the most prominent voice arguing that DHA supplements, fish and algae alike, are “100 percent a scam” and that vegans should take none. He has no scientific or clinical credentials, and there is a conflict worth stating plainly up front: Clean Machine sells Ahiflower oil, a plant ALA and SDA product that is a direct commercial substitute for the DHA supplements he attacks, and his reference document thanks an executive of the company that is Ahiflower’s exclusive supplier. That does not make him wrong, but it is the same kind of conflict this article flags for Fuhrman and Patrick, and it is larger than either.

What he gets right, and where even that gets slippery. Two kernels are real. First, the mechanism of retroconversion: Palmer says the ALA-to-DHA pathway runs one way and that DHA does not convert backward into EPA, and on the narrow enzymatic point the best human evidence (Metherel and Bazinet’s isotope work) agrees. But watch what he builds on it. His actual words are “if you take DHA, that’s all you get,” and that outcome claim is false: in those same studies, supplementing DHA raised plasma EPA by roughly 130 percent. The pathway doesn’t reverse, but your EPA level climbs anyway, because DHA spares existing EPA from being used up. So he is right about the plumbing and wrong about the water: taking DHA does not leave your EPA untouched. Second kernel, cleaner: he is right that the red blood cell omega-3 index does not measure brain DHA, a point the dolphin data and PreventE4’s own spinal-fluid design both support. On that one he is ahead of the sloppier pro-supplement talking points.

Where he overshoots. On almost every claim he starts from a real citable kernel and then sprints past it into an absolute, commercially convenient conclusion his own cited researchers do not endorse. “The brain can’t use dietary DHA at all” rests on mouse data from a lab with a competing-product interest, and PreventE4 showed triglyceride algae DHA does reach the human central nervous system. “90 percent of conversion is invisible to blood tests” is his number, not a finding in any paper. His prostate-cancer claim misstates the study: the 71 percent figure was for high-grade cancer incidence, not death, and his proposed membrane-fluidity mechanism appears nowhere in the literature and is contradicted by studies showing DHA suppresses androgen-receptor signaling. “Inuit were dying at 35 because of DHA” inverts the history: their short lifespans were driven by infection, accident, and infant mortality, and life expectancy rose as the marine diet was abandoned.

The dangerous one, with an important update. In his Plant Based News interview Palmer claimed B12 is “found in plants” and called the B12 requirement “garbage,” citing a hydroponic-lettuce study in which lettuce absorbed B12 only because it was artificially injected into the growing medium, at an absorption rate of hundredths of a percent. That is a horticultural proof of concept, not a dietary source. But in a later long-form debate with the dietitian Dr. Matthew Nagra, Palmer sharply walked this back, saying “please don’t listen to any influencer, me or anybody else, that says you don’t need to take B12,” calling that message “reckless” and “dangerous,” and reframing his point as a speculative theory about ancestral soil rather than current advice. Take the corrected version as his real position: vegans should supplement B12. The walkback matters, and it is a pattern, because Palmer himself concedes he overstates.

What happened when a dietitian debated Palmer directly?

The most useful test of Palmer’s case is the roughly five-hour debate in which Dr. Matthew Nagra, an evidence-focused plant-based dietitian, went through the claims one by one. Two things stand out. First, Palmer opened by conceding the game: he admitted he says “does” when he should say “may,” that his case is “theoretical and mechanistic” because “the data doesn’t exist,” and that he states things “as conclusions” while really raising “questions.” That is a remarkable admission from someone who told a larger audience that DHA is “100 percent a scam.” It means his strongest claims are, by his own account, hypotheses dressed as findings.

Second, the specifics did not survive scrutiny. On the central “DHA blunts EPA and causes harm” claim, Nagra showed Palmer was stitching together separate studies with different doses, different comparators, and different populations, then treating the contrast as if a single trial had pitted EPA against EPA-plus-DHA, which no study in his stack actually did. When Palmer leaned on a 37 percent “increased risk” figure, Nagra pointed out it was not statistically significant, with a confidence interval running from a 15 percent reduction to a doubling, meaning the data could not even establish the direction of the effect. On oxidation, Palmer’s key citation was a paper literally titled “Why Fish Oil Fails,” which turns out to be authored by a supplement marketer with no research credentials and to have since been retracted, and Nagra made the same point this article makes about the rest of Palmer’s mechanistic case: a laboratory oxidation finding is not a human outcome, and if isolated mechanisms settled things you could equally “prove” that exercise is bad because it too transiently raises LDL’s susceptibility to oxidation. Nagra’s closing framing was that Palmer’s position is grounded in mechanism and possibility, his own in outcome data, and that when the questions lack clear answers the honest move is caution, not certainty.

Net verdict: Palmer is a useful corrective on two narrow technical points and wrong, overstated, or, by his own admission, unfalsifiable on the headline claims that drive his “scam” conclusion. He walks the most dangerous one (B12) back when pressed, which is to his credit but also tells you how much weight the original framing could bear. None of his named researchers, including Radak, endorse the leap from “the biology is more complex than fish-oil marketing suggests” to “take no DHA.”

What does Dr. Tim Radak say about vegans and DHA?

Radak (DrPH, MPH, RDN; public health nutritionist, former Director of Nutrition at the Physicians Committee for Responsible Medicine; plant-based advocacy background, no commercial product) lands on “probably not needed.” His strongest contribution is the most rigorous treatment of confounding in this literature: during the decades these cohorts ran, people were told to swap red meat for fish, so a high omega-3 index often marks the person who followed health advice, and the bottom quintile often marks a poor overall diet. He documents dozens of uncontrolled variables and correctly notes WHO and the Alzheimer’s Association do not recommend supplementation. Where he is weak: asymmetric skepticism. He forensically dismantles the one positive brain-volume trial (Witte 2014) while citing small, old, weak studies on his own side without similar scrutiny, leans hard on a tiny vegan subgroup in Welch 2010, and makes a hundred-fold units error on the AHS-2 adipose DHA figure (writing 12% where the study says 0.12%). He also indulges the same mechanistic overreach he criticizes, with the “DHA auto-oxidizes and causes harm” material. His actual bottom line is moderate: an algae supplement could be considered, the research does not suggest benefit or need, except for FADS carriers. That is “not needed,” not “avoid.”

What does Dr. Joel Fuhrman say about vegans and DHA?

Fuhrman (MD) says vegans need it. Note the conflict: he sells both an EPA/DHA product and the omega-3 index test, and the article recommending both links to his own store, a more concrete conflict than Radak’s advocacy background. His strongest contribution is the AHS-2 mortality data, and he is right that the vegan index sits around 4% and that ALA does not reliably raise it. Where he is weak: he essentially does not engage the null RCT literature (AREDS2, VITAL, MAPT, Cochrane were all available; his January 2026 article predates PreventE4 but not those), calls the mechanistic link “undeniable,” describes a vegan brain as “stripped of its primary defense,” dismisses the B12 explanation for AHS-2 in one sentence, and cites a Parkinson’s/pesticide study whose protective factor was actually ALA, the plant omega-3, to argue for EPA/DHA. His 7 to 9% target is a goal he concedes may be unattainable.

Why is DHA not in Dr. Greger’s Daily Dozen?

Because the Daily Dozen is a whole-food checklist, and supplements live on a separate list. This is commonly misread as Greger not recommending DHA. He does: his Optimum Nutrition Recommendations include 250 mg/day of pollutant-free (yeast- or algae-derived) long-chain omega-3s, alongside B12, vitamin D, and iodine. Ground flaxseed is in the Daily Dozen and covers ALA. His stated rationale is specifically brain, explicitly not heart: in his own words this would be predominantly for brain health, not heart health, since the long-chain omega-3s have disappointingly not reduced cardiovascular risk. He cites Sarter’s finding that 250 mg/day moved vegans from 3.1% to 4.8% and frames it as precaution: our bodies can elongate ALA, but can they make enough for optimal brain health, and until we know more, take 250 mg. That is arguably the most defensible of the four positions: explicitly precautionary rather than evidence-of-benefit, honest about the cardiovascular failure, and at a dose below any risk threshold.

Does Greger’s brain-atrophy argument still hold up?

Partly, and it is worth separating the pieces. His argument runs: low omega-3 status correlates with brain atrophy, there is a threshold around 4.4, most vegans fall below it, one trial showed supplementation slows atrophy, therefore take 250 mg.

The correlation link holds. Framingham, WHIMS, and the 2022 midlife Framingham analysis all find the lowest omega-3 group faring worst on brain measures. The vegan-status link holds too, straight from Sarter. Those parts are accurate.

The two weakest links are the threshold and the trial. As covered above, 4.4 is a quartile cutoff rather than a validated breakpoint. And the single supporting RCT, Witte 2014, has never been replicated in the decade-plus since, despite being the linchpin of the causal claim. Meanwhile the evidence has moved against that link: AREDS2, MAPT, VITAL-Cog, and Shinto 2024 were all null on cognition or brain imaging, and PreventE4 in 2026 pushed participants from an index of about 4.9 all the way to about 11, proved the DHA reached the central nervous system, and still found no difference in hippocampal volume, cortical thickness, or cognition.

One honest caveat rescues his specific claim from outright falsification. PreventE4 participants started at about 4.9, which is above his 4.4 line, so the trial tested whether pushing already-adequate people higher helps, not whether rescuing below-threshold people helps. No trial has ever tested that second question with brain-structure endpoints. So his hypothesis is best described as unfalsified but also unsupported by any randomized evidence, and the honest framing of a 250 mg dose is insurance against a documented low biomarker, not a proven way to protect your brain.

What do the Sherzais, the vegan neurologists, recommend?

Dean and Ayesha Sherzai are neurologists who co-direct an Alzheimer’s prevention program, wrote The Alzheimer’s Solution, and raised two children vegan from conception. Because they are plant-based brain specialists, their position carries particular weight, and it is notably measured. They are food-first: eat omega-3-rich whole foods like chia and flax, cut the saturated and omega-6 fat that impairs conversion, and supplement algae-based DHA optionally, mainly at high-stakes life stages. As Dean Sherzai put it, “eat foods rich in omega-3s like chia and flax seeds and you should be fine, but even then, if you’re worried, especially in certain times of life like developing brain, children, pregnancy, or aging brain, take a supplement, algae-based.” Their book names a figure of at least 250 mg DHA per day for those who choose to supplement. Ayesha Sherzai supplemented during pregnancy but currently takes only B12 plus a careful diet.

Two things make them useful in this lineup. First, their own peer-reviewed systematic review (2023) is the most sober document any plant-based advocate here has produced: it found “inconsistent effects” of supplementation and a likely “threshold effect” in which people already meeting their needs gained nothing, with a trend toward benefit only in those already declining. Second, they have no supplement line and no testing-company sponsorship that I could find, which sets them apart from Fuhrman. They frame Alzheimer’s as largely modifiable by lifestyle and treat any single supplement as minor next to diet pattern, blood pressure, glucose, and sleep. Their practical stance sits almost exactly alongside Greger’s: a small algal dose as reasonable insurance, prioritized for pregnancy and the aging brain, not a mandate.

What does Professor Philip Calder say?

Philip Calder is one of the most-cited academic lipid scientists in the world and a useful non-vegan, non-commercial anchor. He is careful with the word “essential”: strictly, only ALA and linoleic acid are essential, because vegan populations “eat zero EPA and DHA and they live,” so DHA is best called conditionally essential, an optimization rather than a survival requirement. He is candid that ALA-to-DHA conversion is low (under about 5 percent), that high omega-6 intake worsens it, and that you cannot reach the EPA and DHA levels associated with benefit by eating ALA alone. His practical read: preformed EPA and DHA are helpful, most people don’t get enough, and if you don’t eat fish you should both mind your omega-6 intake and consider a supplement, aiming for a modest lifelong intake rather than short high doses.

What makes Calder valuable is his evenhandedness on the surrounding fights. He pushes back hard on omega-6 fear, noting that people with the highest blood linoleic acid have the lowest cardiovascular and diabetes risk, and he cautions that raising omega-3 by slashing omega-6 is not automatically a net win because omega-6 does essential work too. He also frames omega-3 as “a nutrient, not a drug,” arguing the null supplement trials asked too much: a small dose, added late, on top of multiple medications, over a few years, cannot mimic a lifetime of adequate status. That framing is the single best explanation on offer for why the observational and trial evidence point different directions, and it applies to the vegan question directly.

What does Dr. Rhonda Patrick say about omega-3?

Patrick (PhD, biomedical science) is the non-vegan voice here, and including her is useful precisely because she is working from the same biomarker literature as the plant-based advocates but reaching a much more aggressive conclusion. Her position: aim for an Omega-3 Index of 8 to 11 percent, which she describes as the range with the greatest health benefits.

What she gets right deserves credit. Her treatment of supplement forms is accurate and matches the independent evidence: triglyceride and phospholipid forms need pancreatic lipase, ethyl esters need an additional enzyme and therefore a fatty meal, and free fatty acid forms need no enzymatic digestion at all. She reports the Walker 2019 dosing model correctly. Her coverage of FADS and ELOVL genetics, sex differences in conversion, and the finding that the index rises with age is all sound. She reports the atrial fibrillation dose split honestly rather than burying it, noting that under 1 gram raised risk 12 percent while over 1 gram raised it 49 percent. And notably, she concedes that the Omega-3 Index may not be the best marker for the brain, citing work suggesting plasma BDNF better reflects brain concentrations. That is a real point of agreement with the plant-based skeptics.

Where the argument strains. Her page is built around roughly ten pull-quotes from William Harris, who developed the Omega-3 Index and owns the company that sells the test, which is a heavier reliance on a single commercially interested source than any of the vegan authors. The 8 to 11 percent target is sourced to a 2004 paper proposing the index in the first place, which is thin support for calling it robust. And there is an unresolved internal tension: she targets 8 to 11 percent, correctly notes this requires 1,750 to 2,500 mg per day, and correctly notes that over 1 gram per day raises atrial fibrillation risk by 49 percent, without reconciling the three. Her weakest material is the COVID section, which leans on a 101-person trial reporting a survival difference too large to be plausible and on computer modeling of the viral spike protein.

Why does Rhonda Patrick recommend fish over algae supplements?

She does acknowledge algae. She notes algal sources suit people who avoid fish and cites the systematic review showing algal supplementation raises DHA and the omega-3 index in vegetarians. But the framing subordinates it: the article states it focuses primarily on marine-derived omega-3s, and the main comment on algal products is that they provide lower concentrations than similarly priced fish oil. Her enthusiasm goes toward salmon roe as a concentrated phospholipid source.

The reason matters, and it produces the most interesting collision in this whole debate. Her stated rationale is that APOE4 carriers respond well to DHA from fish but less well to most supplements, because fish provides DHA in phospholipid form, which determines whether it becomes free DHA or LPC-DHA for transport across the blood-brain barrier. She cites her own 2019 hypothesis paper for this.

That is the same mechanistic argument the plant-based side uses, aimed in the opposite direction. The vegan version says the LPC transport route means ordinary supplements cannot enrich the brain, so rely on ALA and endogenous conversion. Patrick’s version says the LPC transport route means non-phospholipid supplements cannot enrich the brain, so eat fish and roe. Identical premise, opposite conclusion. That alone should make anyone cautious about how much weight the mechanism can carry.

And PreventE4 undercuts both versions, hers quite directly. The trial gave triglyceride-form algal DHA to a cohort that was 47 percent APOE4 carriers. Her hypothesis predicts carriers should be the group that fails to get it into the brain. Instead, cerebrospinal fluid DHA rose essentially equally in carriers and non-carriers, with an interaction p-value of 0.71. The transport-form part of the hypothesis did not survive contact with the trial. Her page was last updated in July 2026, about a month after PreventE4 published, and does not address it.

What does Simon Hill (Plant Proof) recommend?

Simon Hill is a plant-based nutrition communicator whose position is the most pro-supplementation of the plant-based voices here, closer to Harris and Patrick than to Greger or the Sherzais. He targets an omega-3 index of 8 to 12 percent and recommends 1 to 1.5 grams a day of combined DHA and EPA, algae oil preferred, in triglyceride form, taken with a meal, with a baseline index test before starting and a retest after four months. He argues that vegans and vegetarians can run an index as low as 2 to 3 percent, that ALA from flax and chia is unlikely to be sufficient for most people, and that the atrial fibrillation signal is real but dose-dependent and outweighed, for people without cardiovascular disease, by the cardiovascular benefit he infers from the VITAL low-fish subgroup.

He discloses that he has no omega-3 product or testing company of his own, though he has sponsor and affiliate relationships with other supplement brands (among them a prebiotic he co-formulated, plus a plant-based multivitamin and other wellness products), so weight his advice knowing he profits from supplements broadly but not from omega-3 specifically. His dose sits at the high end of what this article’s evidence supports: reaching an 8 percent index from a low baseline genuinely does take the 1 to 2 gram range, but that is also the dose zone where the AFib and LDL trade-offs begin, and where the null cognitive trials found no benefit despite hitting those levels. His “measure and optimize to a target” philosophy is coherent, but it rests on treating the index as a goal worth chasing, which the brain evidence does not establish and the cardiovascular evidence supports only in general, largely omnivorous populations.

What do all these experts agree on?

More than the shouting suggests. Nearly everyone here accepts that the vegan omega-3 index runs around 4 percent, that high-dose omega-3 carries an atrial fibrillation signal, and that the Omega-3 Index is an imperfect marker for the brain specifically. The plant-based experts, even Palmer, agree that B12 is non-negotiable (Palmer being the lone, and dangerous, exception), and that flax, chia, and walnuts belong in the daily diet. Almost no one is telling healthy vegans to eat fish.

The disagreement is really about two things: how far above 4 percent to chase, and whether observational biomarker data or randomized outcome data should drive the call. It sorts the eight voices into three camps. The scam and skeptic camp (Palmer, Radak) weights the trials and the confounding critique and lands on “not needed” or “avoid.” The precautionary camp (the Sherzais, Greger, and Calder in practice) treats a small algal dose as cheap insurance, prioritized for pregnancy and the aging brain, without claiming proven benefit. The optimize-a-target camp (Simon Hill, Patrick, Fuhrman) weights the biomarker data heavily and pushes for an 8-to-12 percent index requiring roughly 1 to 2 grams a day, which is also exactly the dose zone where the AFib and LDL trade-offs appear and where the null cognitive trials found nothing.

The practical distance from the skeptics to the precautionary camp is one small 250 mg capsule. The distance from there to the optimizers is about eight capsules and a different theory of evidence. The real crux, still unresolved, is whether a 4 percent index in an otherwise healthy person is a benign set point or a quiet insufficiency with a decades-long fuse. Nobody has the evidence to close it. The sharpest illustration is watching the same LPC-DHA transport mechanism get deployed by Palmer to argue for flaxseed and by Patrick to argue for salmon roe in the same year, while a large-cohort neurologist like Gary Fraser reports an unpublished signal of higher neurological risk in the very oldest vegans that no one can yet explain.

What would actually settle this

What study would answer the vegan DHA question for good?

A long-duration randomized trial of algal EPA/DHA versus placebo in long-term well-planned vegans, with cognitive and brain-structural endpoints, controlling for B12 status (homocysteine and MMA, not just serum B12), sleep, exercise, social engagement, and APOE genotype. It will probably never happen: dementia latency is measured in decades, early-onset dementia is too rare to power, the vegan population is small, and nobody has a commercial incentive to fund a trial whose likely result is “the cheap capsule does not matter.” What is actually in progress: LPC-DHA cognition trials (University of Cincinnati and others, expected around 2029), the first real human test of the transport-form hypothesis, plus continued follow-up of the AHS-2 and UK Biobank cohorts. What would change the picture: a positive LPC-DHA trial, replication of the AHS-2 neurological signal in an independent vegan cohort with B12 properly measured, or a trial in genuinely low-status individuals showing benefit.

The takeaway

So, bottom line, do healthy vegans need to worry about DHA?

If you eat a well-planned whole-food vegan diet, get your ALA from flax, chia, and walnuts, do not drown everything in omega-6 oils, keep your B12 verified with MMA and homocysteine rather than just serum B12, exercise, sleep, stay socially and cognitively engaged, and keep your blood pressure and metabolic health in good shape, you are doing far more for your brain and heart than any omega-3 capsule has ever been shown to do in a randomized trial.

If you want to take 250 mg of algal EPA/DHA anyway as cheap insurance against a question nobody has answered, that is a defensible choice with a very small downside. If you do not, that is also defensible, and the evidence is arguably on your side. What is not defensible is the confidence on either end: the people telling you DHA is dangerous and you should avoid it, and the people telling you your brain is being stripped of its defenses. Both are running well past what anyone actually knows. Talk to a physician or registered dietitian who understands plant-based diets, especially if you are pregnant or planning to be, know your APOE status, are over 65, or have a confirmed very low index.

For the rest of the picture on eating this way, see our guides to vegan nutrition, B12, dietary fat, cholesterol, seed oils, plant toxins and antinutrients, raw diets, losing weight, and raising children vegan. And if you came here through the ethics-and-environment side of the conversation, our pieces on crop deaths, the pesticide argument, the “livestock eat food we can’t” claim, regenerative farming, rewilding, coconut oil vs animal fat, the Adventist Health Studies, the longest-living people, the Blue Zones, the Saul Newman Blue Zones fact-check, and the meat and bowel cancer link go deeper.

References and further reading

Links open in a new window. Where a study was discussed above, the primary source, DOI, or PubMed record is linked here.

Supplementation trials: brain and cognition

• Chew EY, et al. AREDS2 cognitive outcomes. JAMA 2015. PubMed
• Kang JH, et al. VITAL cognitive ancillary. Alzheimer’s & Dementia TRCI 2022. PubMed
• Andrieu S, et al. MAPT trial. Lancet Neurology 2017. PubMed
• Quinn JF, et al. ADCS DHA trial in Alzheimer’s. JAMA 2010. PubMed
• Dangour AD, et al. OPAL trial. Am J Clin Nutr 2010. PubMed
• Shinto L, et al. Omega-3 and white matter, JAMA Network Open 2024. PubMed
• Yassine HN, et al. PreventE4 (CNS target engagement of high-dose DHA). eBioMedicine 2026 / SSRN preprint. SSRN
• Sydenham E, Dangour AD, Lim WS. Omega-3 for prevention of cognitive decline. Cochrane 2012. PubMed
• Witte AV, et al. Long-chain omega-3 and brain function/structure. Cereb Cortex 2014. PubMed

Cardiovascular outcome trials and reviews

• Abdelhamid AS, et al. Omega-3 for cardiovascular disease. Cochrane 2020. PubMed
• Manson JE, et al. VITAL trial (marine n-3). N Engl J Med 2019. PubMed
• ASCEND Study Group. n-3 fatty acids in diabetes. N Engl J Med 2018. PubMed
• Naghshi S, et al. Dietary ALA and mortality, dose-response meta-analysis. BMJ 2021. PubMed

Atrial fibrillation

• Gencer B, et al. Omega-3 supplements and atrial fibrillation risk. Circulation 2021. PubMed
• Kowalski C, et al. Omega-3 and AF recurrence, meta-analysis. 2018. PMC
• Kowey PR, et al. FORWARD trial (n-3 PUFA for AF). 2010. PubMed

LDL cholesterol and lipid effects

• Theobald HE, et al. Low-dose DHA raises LDL. Am J Clin Nutr 2004. PubMed
• Bernstein AM, et al. Algal oil DHA meta-analysis. J Nutr 2012. PubMed
• Allaire J, et al. ComparED study (DHA vs EPA, LDL). J Clin Endocrinol Metab 2018. PubMed

Prostate cancer

• Brasky TM, et al. Plasma phospholipid fatty acids and prostate cancer in SELECT. J Natl Cancer Inst 2013. PubMed
• McCarty MF, DiNicolantonio JJ. Omega-3 and prostate cancer: examining the pertinent evidence. Mayo Clin Proc 2014. Mayo Clinic Proceedings
• Dietary omega-3 and prostate cancer progression, PLCO analysis 2025. PMC

Vegetarian and vegan cohorts

• Tong TYN, et al. EPIC-Oxford, heart disease and stroke over 18 years. BMJ 2019. PubMed
• Chiu THT, et al. Vegetarian diet and stroke, Tzu Chi cohort. Neurology 2020. PubMed
• Tsai J, et al. Vegetarian diet and dementia, Tzu Chi cohort. Nutrients 2022. Nutrients
• Abris GP, et al. Cause-specific mortality in vegetarians, AHS-2. Am J Clin Nutr 2024. PubMed
• Harris WS, Pawlak R, et al. Dietary and erythrocyte PUFAs in AHS-2 diet groups. PLEFA 2025. PubMed
• Welch AA, et al. EPIC-Norfolk, precursor-product ratio of ALA to long-chain n-3. Am J Clin Nutr 2010. PubMed
• Loong S, et al. Omega-3, cognition, and brain volume in AHS-2 older adults. Brain Sci 2023. Brain Sciences

Conversion, tracer studies, and the pathway

• Domenichiello AF, Kitson AP, Bazinet RP. Is DHA synthesis from ALA sufficient to supply the adult brain? Prog Lipid Res 2015. PubMed
• Barcelo-Coblijn G, Murphy EJ. ALA and its conversion to longer chain n-3. Prog Lipid Res 2009. PubMed
• Baker EJ, et al. Metabolism and functional effects of plant-derived omega-3 in humans. Prog Lipid Res 2016. PubMed
• Rett BS, Whelan J. Increasing dietary linoleic acid does not increase tissue arachidonic acid. Nutr Metab 2011. PMC
• Klein V, et al. ALA supplementation and long-chain n-3 profiles across diets (NuEva). Front Nutr 2025. Frontiers in Nutrition

Brain DHA, transport, and turnover

• Umhau JC, et al. Imaging incorporation of circulating DHA into human brain. J Lipid Res 2009. Journal of Lipid Research
• Nguyen LN, et al. Mfsd2a is a transporter for DHA. Nature 2014. PubMed
• Guemez-Gamboa A, et al. MFSD2A mutations and microcephaly. Nat Genet 2015. PubMed
• Sugasini D, et al. Dietary LPC-DHA and brain DHA enrichment. Sci Rep 2017. Scientific Reports
• Ly J, et al. Circulating phosphatidylcholines as surrogate biomarkers of the omega-3 index. J Lipid Res 2023. Journal of Lipid Research
• Metherel AH, Bazinet RP, et al. Whole-body DHA synthesis exceeds brain uptake in rats. 2013. PubMed

The Omega-3 Index and dose-response

• Harris WS, Del Gobbo L, Tintle NL. The Omega-3 Index and CHD mortality. Atherosclerosis 2017. PubMed
• Flock MR, et al. Determinants of erythrocyte omega-3 content, dose-response RCT. J Am Heart Assoc 2013. JAHA
• Walker RE, et al. Predicting the effects of supplemental EPA and DHA on the omega-3 index. Am J Clin Nutr 2019. AJCN
• Sarter B, et al. Blood DHA and EPA in vegans, and effects of an algal supplement. Clin Nutr 2015. PubMed

Supplement form and bioavailability

• Schuchardt JP, et al. EPA and DHA incorporation from fish oil vs krill oil. Lipids Health Dis 2011. PMC
• Arterburn LM, Hall EB, Oken H. Distribution, interconversion, and dose response of n-3 fatty acids in humans. Am J Clin Nutr 2006. PubMed

Tissue distribution and roles

• SanGiovanni JP, Chew EY. Omega-3 in the retina. Prog Retin Eye Res (review). PMC
• Lacombe RJS, Bazinet RP. DHA in the brain (structure and function review). 2019. ScienceDirect

Autopsy and the DHA-Alzheimer’s puzzle

• Cunnane SC, et al. Plasma and brain fatty acids in Alzheimer’s disease. 2012/2013. PubMed
• Fraser T, et al. Fatty acid composition of frontal, temporal and parietal cortex in Alzheimer’s. 2010. PubMed

Practitioner positions

• Greger M. Optimum Nutrition Recommendations. NutritionFacts.org. NutritionFacts
• Greger M. Should Vegans Take DHA to Preserve Brain Function? NutritionFacts
• Fuhrman J. Why Vegans Need DHA and EPA. DrFuhrman.com 2026. DrFuhrman.com
• Radak T. Essential fatty acids, diet, and lifestyle factors influencing brain health. radaktim.wixsite.com
• Patrick RP. Omega-3 fatty acids (topic overview). FoundMyFitness. FoundMyFitness
• Patrick RP. Role of phosphatidylcholine-DHA in preventing APOE4-associated Alzheimer’s disease. FASEB J 2019. PubMed
• Sherzai AZ, Sherzai AN, Sherzai D. A systematic review of omega-3 consumption and neuroprotective cognitive outcomes. Am J Lifestyle Med 2023. PubMed
• Sherzai D, Sherzai A. Omega-3s and Brain Health (interview). The Proof with Simon Hill. theproof.com
• Hill S. Omega-3 vs Omega-6: The Ultimate Guide to Essential Fatty Acids (masterclass, with Calder, Fuhrman, the Sherzais, Fraser, Harris). The Proof. theproof.com
• Calder PC, Innes JK. Marine omega-3 fatty acids for cardiovascular health: an update for 2020. Int J Mol Sci 2020. PubMed

Retroconversion, feedback inhibition, and the Palmer claims

• Metherel AH, Irfan M, Klingel SL, Mutch DM, Bazinet RP. Compound-specific isotope analysis reveals no retroconversion of DHA to EPA but substantial conversion of EPA to DHA following supplementation. Am J Clin Nutr 2019. PubMed
• Metherel AH, Bazinet RP. Dietary DHA downregulates liver DHA synthesis by inhibiting eicosapentaenoic acid elongation. J Lipid Res 2024. PubMed
• Ference BA, et al. LDL cause atherosclerotic cardiovascular disease: EAS Consensus Statement. Eur Heart J 2017. PubMed
• Harris WS, Schmitt TL. Unexpected similarity in RBC DHA and AA levels between bottlenose dolphins and humans. PLEFA 2014. PubMed
• Bito T, et al. Production of cyanocobalamin-enriched lettuce grown using hydroponics. J Agric Food Chem 2013. PubMed
• Nagra M, Palmer G. Omega-3 Supplements: Helpful or Harmful? (debate). YouTube
• Peskin BS. Why fish oil fails: a comprehensive 21st century lipids-based physiologic analysis. J Lipids 2014. (Palmer’s oxidation citation; author is a marketer with no research credentials, and the paper was later retracted.) PubMed

The omega-3 index threshold and brain-imaging studies

• Tan ZS, Harris WS, Beiser AS, et al. Red blood cell omega-3 fatty acid levels and markers of accelerated brain aging. Neurology 2012. PubMed
• Satizabal CL, Himali JJ, Beiser AS, et al. RBC omega-3 fatty acids, MRI markers, and cognition in midlife: Framingham. Neurology 2022. PubMed
• Courchesne E, Chisum HJ, Townsend J, et al. Normal brain development and aging: quantitative MR imaging. Radiology 2000. PubMed
• Harris WS, von Schacky C. The Omega-3 Index: a new risk factor for death from coronary heart disease? Prev Med 2004. PubMed
• Harris WS, Pottala JV, Varvel SA, et al. Erythrocyte omega-3 increases and linoleic acid decreases with age: 160,000 patients. PLEFA 2013. PubMed
• Benton D, Donohoe RT, Clayton DE, Long SJ. Supplementation with DHA and psychological functioning of young adults. Br J Nutr 2013. PubMed
• Stonehouse W, Conlon CA, Podd J, et al. DHA supplementation improved memory and reaction time in healthy young adults. Am J Clin Nutr 2013. PubMed

Related reading on VeganLinked

• Vegan nutrition overview
• Essential nutrients on a vegan diet
• Vitamin B12
• Dietary fat
• Cholesterol
• Seed oils
• The Adventist Health Studies
• What the longest-living people eat
• Are the Blue Zones debunked?
• Saul Newman vs the Blue Zones, a fact-check
• Plant toxins and antinutrients
• Raw diets
• Losing weight on a vegan diet
• Raising children vegan
• Crop deaths
• The pesticide argument
• Livestock and food we cannot eat
• Regenerative farming
• Rewilding and regenerative claims
• Coconut oil vs animal fat
• Meat and bowel cancer

Read more...

This piece answers the viral tweet thread point by point, the secondhand version most people actually encounter. If you’d rather have the deeper, narrative take on the man behind the thread, see the companion piece: Saul Newman vs. the Blue Zones: A Fact-Check, which walks through the same evidence as a single argument and traces the one recurring error running through all of Newman’s claims.

Meticulously fact-checked, structured responses to all major pillars of the viral anti-Blue Zones narrative:

1. The Age 90+ Metric: Correcting the 100/110+ supercentenarian mix-up.
2. The Scientific Timeline: Separating the 1999/2004 field data from later monetization.
3. The Level A Verification: Proving Sardinia’s methodology actively hunts and removes its own errors.
4. The U.S. Birth Certificate Artifact: Explaining why a 1915 registration timeline makes U.S. data a calendar inevitability.
5. The Japan Koseki Scandal: Disentangling genealogical registries from the physical resident registries in Okinawa.
6. The Greek Pension Relative Math: Proving Ikaria’s 1.49% vs. 0.33% mainland ratio cancels out national baseline errors.
7. The Loma Linda Fallacy: Dismantling the use of zip-code-wide CDC data to disprove independent, individual lifestyle cohort studies.
8. The Actuarial Reality: Showing how extreme-age tables are mathematically extrapolated, not dictated by supercentenarian lists.
9. The Consulting vs. Infrastructure Split: Acknowledging fair business critiques while insulating the underlying demographic science.
10. The Academic Precedent and Status: Proving longevity skepticism dates back to the 1870s and placing Newman’s still-unpublished preprint in context.

Are the Blue Zones real at all?

“The Blue Zones aren’t real, it’s bad record keeping, pension fraud, and money”

Claim: “The Blue Zones, places where people supposedly live past 100 at extraordinary rates, aren’t real. On what went wrong: bad record keeping, pension fraud, and lots of money.”

Skepticism about extreme old age is fair; demographers have known about record fraud and age inflation since the 1870s. But this critique conflates general database errors with the highly specific, manual verification used for Blue Zones.

The core academic definition of a Blue Zone relies heavily on the proportion of the population reaching age 90, an age cohort where data is vastly more reliable.

Furthermore, the formal scientific defense published in The Gerontologist details how researchers painstakingly cross-checked individual civil, church, and military records to systematically filter out fraud. Meanwhile, Newman’s central Blue Zones paper remains an unpublished preprint seven years after it first appeared in 2019; by his own account (STAT News, May 2026) it is still in review at BMJ Public Health after an unusually protracted nine rounds of peer review, far more than the typical one to three, and it has not passed the peer-reviewed checkpoint the Blue Zones validation cleared long ago. (His separate, peer-reviewed work on late-life mortality plateaus is legitimate and is not at issue here.)

Did the Blue Zones start as a marketing gimmick?

“It started as a 2004 paper then got popular after a National Geographic cover”

Claim: “The Blue Zones started as a 2004 paper then got popular after a National Geographic cover story. They became big business: books, resorts, tours, frozen meals, towns paying huge sums to get certified, and even a Netflix documentary.”

The timeline is right, but the logic is flawed. Dr. Gianni Pes presented the Sardinia data in 1999, and the landmark study was peer-reviewed in Experimental Gerontology in 2004, long before a brand existed.

The longevity cohorts in Okinawa, Nicoya, and Ikaria were independently validated by demographers and medical researchers with zero financial stake in the trademark.

It’s true the concept later became big business: books, resorts, tours, frozen meals, city-certification contracts, and the four-part Netflix docuseries Live to 100: Secrets of the Blue Zones (released August 30, 2023). But all of that came years after the peer-reviewed field data. Dan Buettner commercializing the concept later doesn’t retroactively falsify decades of independent, peer-reviewed demographic field work. A scientific finding isn’t wrong just because someone figured out how to sell books about it.

Is the whole thing just pension fraud and messy records?

“The age records are a mess and pension fraud is rife”

Claim: “The idea is that if we copy the habits of people in the Blue Zones, we’ll also reach 100+. The problem: the age records are a mess, and pension fraud is rife.”

The framing is off. Blue Zones measure the population share reaching age 90, not supercentenarians, so “reach 100+” misstates the actual science.

More importantly, the pension-fraud narrative ignores the rigorous methodology used. In Sardinia, researchers used Level A verification, painstakingly cross-checking every person over 90 against civil registries back to 1866, parish archives from the 17th century, and full village family trees.

This exact process famously caught a false 110-year-old record, where a woman was confused with her younger sister who shared the same name and was born three years later. The researchers published the error and removed the data. A methodology that actively hunts down, detects, and publicizes its own false positives isn’t being fooled by fraud; it’s preventing it.

Do missing birth certificates prove the ages are fake?

“The best predictor of US supercentenarian status is not having a birth certificate”

Claim: “The best predictor of US supercentenarian status is not having a birth certificate. 82% of records predate statewide birth certification; when a state hit full coverage, its count dropped ~80%.”

The numbers are roughly accurate; Saul Newman found a 69–82% drop in U.S. supercentenarian records after state certification. But the conclusion is fundamentally flawed for two reasons.

First, it’s a calendar artifact. The U.S. didn’t establish a nationwide birth-registration area until 1915 (completed in 1933). Because a supercentenarian must be 110+ years old, anyone reaching that age today must have been born before full certification existed. There was no alternative.

Second, and decisively: none of this applies to the Blue Zones. Newman’s birth-certificate finding is strictly about the United States, which has zero demographically validated Blue Zones. The actual zones are in Italy, Japan, Greece, and Costa Rica, each validated using their own independent, localized registries. Even granting his U.S. numbers in full, the critique fails to touch the actual claim.

Did Japan find 230,000 dead centenarians?

“An audit found 82% of Japan’s centenarians, 230,000 people, were dead on paper”

Claim: “Japan found its oldest citizen had been dead 30 years while family cashed his pension. An audit then found 82% of centenarians, 230,000 people, were dead on paper.”

This merges two separate registries and inflates the result. The 234,354 figure came from the koseki, a genealogical family register that is never crossed out when someone dies; those entries represent 0.5% of all births recorded between 1872 and 1910, and the researchers who investigated concluded the impact on Japan’s life-expectancy statistics was effectively nil.

The register that actually pays pensions and counts the living, the resident registry, showed 44,449 living centenarians as of September 1, 2010, all confirmed alive by local officials. When those officials ran nationwide proof-of-life checks, they verified 584 genuinely missing centenarians (271 reported by August 27, 2010, and 313 more by October 2010, per the Ministry’s February 4, 2011 release), an error rate of roughly 1.3%, not 82%. The ministry attributed the gaps to deaths unreported in postwar chaos and to emigration, and contemporary press coverage highlighted absurd outliers, including a family register that still listed a resident as 186 years old.

Critically, those roughly 600 missing centenarians were spread across other prefectures, and zero turned up in Okinawa (0 of 922 centenarians), the only part of Japan that was ever a Blue Zone.

Were Greek centenarians collecting pensions from the grave?

“Greece had 9,000 centenarians and 72% were collecting pensions from the grave”

Claim: “Greece had 9,000+ centenarians on the books; an audit found at least 72% were collecting pensions from the grave.”

The scandal is real but aimed at the wrong data. The 9,000 were centenarians drawing old-age benefits, entries in a pension-payments system, not a validated demographic count. (The “72%” is Newman’s own estimate about those payment rolls, not an audited demographic finding.)

Greece’s Blue Zone is Ikaria, and it was validated in 2008–09, before the bailout audit, using independent sources: national death statistics for 1995–2006, the municipal register, birth records complete from 1913, and interviews with every resident aged 90 and over cross-checked against datable events. It found 124 people aged 90+ alive on the island, 1.49% of the population, nearly five times the 0.33% on the mainland. Because Ikaria is measured against the rest of Greece, any uniform record-keeping problem largely cancels out of the comparison.

Is Loma Linda a fake American Blue Zone?

“Loma Linda became a Blue Zone because an editor asked for one”

Claim: “Loma Linda became a Blue Zone because an editor asked for an American one, and CDC data shows its lifespan is only average.”

The origin story is true; Buettner told The New York Times he included Loma Linda because his editor said he needed to find America’s blue zone. But the CDC comparison tests the wrong population.

Loma Linda isn’t one of the four zones validated by demographic criteria; those are Okinawa, Sardinia, Ikaria, and Nicoya. The underlying research is a cohort study of Seventh-day Adventists that predates Buettner by decades: Fraser and Shavlik followed 34,192 California Adventists from 1976 to 1988 and found life expectancy at age 30 was 7.28 years higher for the men and 4.42 years higher for the women than for other white Californians.

A city-wide average can’t rebut that, because Adventists are only a subset of the city’s residents, roughly 9,000 of some 23,000, and the study tracked people, not zip codes.

There’s a deeper point here that turns Loma Linda from Newman’s easiest target into the strongest evidence in the whole field. Every other validated zone entangles diet with everything else about an isolated place: genetics, climate, a thousand years of shared geography. Loma Linda doesn’t: the Adventists live in the same country as their neighbors, shop at the same supermarkets, and use the same medical system, so what differs is mainly what they do. That makes the Adventist Health Study a rare natural experiment, and the longevity gradient it reveals is dietary: Adventist men outlive other Californians by 7.3 years, and vegetarian Adventist men by 9.5. Newman attacked the freeway-exit packaging; the cohort underneath it is the cleanest data in longevity science. We work that cohort in detail in two companion pieces: What the Longest-Living People Actually Eat, which traces the full plant-forward dose-response gradient, and What the Adventist Health Studies Actually Tell Us, which is an honest accounting of what that mortality data can and can’t prove.

Does this data make your insurance cost more?

“Extreme old-age data sets your pension and insurance rates, so you pay more”

Claim: “Extreme old-age data is used to model lifespan, which sets pension rates and insurance pricing, so if models assume you live longer, you pay more.”

Wrong on the mechanism and half-wrong on the direction. Life tables aren’t built from supercentenarian records: the Social Security actuaries extrapolate at extreme old age precisely because the data there is sparse or of questionable quality, and pricing runs on experience data covering millions of lives, mostly between 65 and 95. The entire validated global supercentenarian database launched with 672 people, far too few to move an aggregate table.

And the economics don’t run one way: if a model assumes you live longer, life insurance gets cheaper, not more expensive, because you’re less likely to die during the policy term. Longer assumed lifespans do raise the cost of annuities and pensions, but those are priced from pensioner experience, not from a list of 110-year-olds.

Isn’t Blue Zones just selling common-sense advice?

“Communities paid millions and the advice is just free common-sense stuff”

Claim: “Communities paid millions to get certified, and most Blue Zone advice is just free common-sense stuff public health has said for years.”

Both halves have merit, and this is the strongest part of the thread, but it isn’t a debunk of the demography. The certification-fee criticism lands (one North Carolina community, Brevard, reportedly committed around $3 million over three years). The daily-drinking point lands too: recent meta-analyses argue that the apparent health benefit of moderate drinking largely reflects “abstainer bias,” the false appearance of benefit created by putting former and sick ex-drinkers in the non-drinker comparison group, though that reversal is still debated among alcohol researchers.

Where the claim slips is “it’s just free advice”: the Blue Zones Project product isn’t advice, it’s built-environment and policy work, road diets, complete-streets ordinances, tobacco policy, school walking routes. Whether cities should pay for that is a fair question, especially since most of the reported outcomes come from the company’s own materials. That’s a critique of a business, not evidence the zones are fake.

Who is Saul Newman and did he uncover the problems?

“I spoke to the researcher who uncovered all the problems, Saul Justin Newman”

Claim: “On my podcast I spoke to the researcher who uncovered all the problems with the Blue Zones, Saul Justin Newman.”

This is the emotional core of the thread, so it’s worth being precise about who Newman is and where his Blue Zones work actually stands.

First, the skepticism he channels is real and old, and the Blue Zones researchers agree with it. Systematic age-exaggeration has been documented since the 1870s, when William Thoms (Human Longevity: Its Facts and Its Fictions, 1873) showed that almost every celebrated “ultra-centenarian” claim collapsed under documentary scrutiny, and laid down validation rules demographers still use. The most famous modern “longevity hotspots” that turned out to be false were demolished the same way: Zhores Medvedev’s 1974 work dismantled the Caucasus and Abkhazia longevity claims, and Richard Mazess and Sylvia Forman’s 1979 study in the Journal of Gerontology found systematic age-exaggeration after age 70 in Vilcabamba, Ecuador, with not a single verified centenarian (the oldest resident was 96). The entire point of modern Blue Zone validation is to survive the Thoms, Medvedev, and Mazess test, not to ignore it.

Second, Newman’s status. His headline Blue Zones paper has been a preprint on bioRxiv since 2019 (updated in 2020 and again in 2024) and, as of 2026, still has not been published in a peer-reviewed journal. Newman says it has been through nine rounds of review at BMJ Public Health without a decision (reported by STAT News, May 2026). Whatever the reason for the delay, the fact stands: the specific claim that the Blue Zones are an artifact has not cleared peer review, while the validation of the four zones has (Austad and Pes, The Gerontologist, December 2025).

Third, and this is where honest framing matters, Newman is not a crank. His separate work correcting a Nature paper on the limits to human lifespan was peer-reviewed and stands, and his broader critique of extreme-age data won him the first-ever Ig Nobel Prize in Demography in 2024. The distinction is simple: his rigorous, published mortality-statistics work is one thing; his unpublished blanket claim that the Blue Zones are fake is another. Citing the first to lend authority to the second is the actual sleight of hand.

What does the Gerontologist defense actually say?

“The December Gerontologist defense is worth reading, their strongest work is Sardinia”

Claim: “A few people flagged the December Gerontologist defense, worth reading. Their strongest work is Sardinia, back to 1866 and church archives.”

It’s fair to flag it, but the defense is far from Sardinia-only. All four demographic zones are backed by independent, highly robust data structures:

• Nicoya, Costa Rica: A continuous birth registry since 1883, national ID numbers that encode the exact volume and line of a citizen’s birth record, and a death registry the UN classified as complete since 1961. Late or suspicious registrations are systematically excluded.
• Okinawa, Japan: Directly answers the “burned records” objection. Twenty microfilm copies of the historical birth registers survived WWII, allowing exact reconstruction. A targeted audit of 52 centenarians matched stated ages in 94.2% of cases, with zero systematic age inflation found.

Most importantly, the paper concedes realities a marketing whitewash never would. It openly states that Okinawa no longer qualifies as an active Blue Zone due to lifestyle shifts, and that Nicoya’s longevity zone has shrunk to about a quarter of its original size. A scientific defense that transparently reports the decline of its own subject isn’t hiding anything.

And once the zones are established as real, the question worth asking is what the people in them actually ate, because that’s the part with a practical takeaway. An analysis of 154 dietary surveys across the five zones found that roughly 95% of what the longest-lived residents ate came from plants: beans above all, plus whole grains, vegetables, greens, tubers, and fruit, with meat treated as a rarity. Traditional Okinawans are the starkest case, drawing the majority of their calories from sweet potatoes. The demographic fight is about whether the zones are real; the more useful conversation is about the pattern they share. We take that up in full in What the Longest-Living People Actually Eat.

Related reading

Saul Newman vs. the Blue Zones: A Fact-Check takes the same evidence as a single narrative argument aimed at Newman himself. What the Longest-Living People Actually Eat climbs the full hierarchy of diet-and-longevity evidence, from population patterns to identical-twin trials. What the Adventist Health Studies Actually Tell Us is an honest look at what the Loma Linda cohort, the one Blue Zone built on modern vital records, can and can’t prove.

Sources: Austad and Pes, “The validity of Blue Zones demography: a response to critiques,” The Gerontologist (Dec 2025); Saito, Yong and Robine, “The mystery of Japan’s ‘missing centenarians’ explained,” Demographic Research (2012); Rosero-Bixby, Demography (2008) and Demographic Research (2023); Fraser and Shavlik, “Ten Years of Life: Is It a Matter of Choice?,” Archives of Internal Medicine (2001); Willcox et al., Journals of Gerontology Series A (2008); Thoms, Human Longevity (1873); Mazess and Forman, Journal of Gerontology (1979); Medvedev (1974); U.S. Social Security Administration, Office of the Chief Actuary, Actuarial Study No. 120; Newman, bioRxiv preprint 704080 (2019, rev. 2024); STAT News (May 2026).

Read more...

A recent livestream interview, billed as an exposé of “the great supplement scam,” made a series of claims about vitamin B12 that deserve a full, careful, permanent debunking. The guest, a 45-year vegan author, told viewers that B12 was essentially invented in a Merck laboratory, that deficiency symptoms are really chemical poisoning misattributed to a vitamin, that the cyanide in cyanocobalamin should scare you, that he hasn’t had a blood test in 50 years and knows he’s healthy because he feels great, and (most dangerously) that vegans should not only skip B12 supplements but actively seek out unfortified nutritional yeast to avoid the B12 that’s been added to foods.

This post takes those claims apart one at a time, using the full hierarchy of scientific evidence: consensus statements, systematic reviews and meta-analyses, randomized controlled trials, cohort studies, case series, and the underlying biochemistry. It also addresses the conspiracy framing head-on, because the “follow the money” story collapses under one inconvenient, well-documented fact: the single biggest customer of the vitamin B12 industry is animal agriculture itself [29, 30]. And it closes with exactly what the evidence says vegans should do, covering doses, forms, and testing, so that nobody reading this ever has to gamble.

One thing before we start. I’m writing this as an advocate for plant-based eating. The people who most need this information are vegans, and the people most harmed by the video’s advice are vegans, including, in the worst documented cases, vegans’ babies [13]. Debunking B12 denial isn’t an attack on veganism. It’s a defense of it.

The claims on the table

Distilled from the interview, here is what was asserted:

1. Vegans don’t need vitamin B12, and B12 deficiency isn’t real. The symptoms attributed to it are actually caused by environmental chemical exposure.
2. B12 was “invented” or fabricated in a Merck laboratory as part of a profit scheme.
3. The fact that B12 is the one vitamin you can’t get from fruits and vegetables is suspicious, and is evidence of an agenda to push people toward meat, dairy, and pills.
4. Cyanocobalamin contains cyanide, so taking it means knowingly consuming poison.
5. Vitamins in general are “refined petroleum,” petrochemicals one step removed from pharmaceuticals.
6. The speaker’s own 45 years of unsupplemented health (no doctor visits, no blood tests, a “biological age of 22” from balance and grip-strength tests) proves supplementation is unnecessary.
7. Healthy centenarians in places like Okinawa have “terrible” blood levels yet thrive, so blood markers are meaningless.
8. Follow the money: the pharmaceutical-supplement complex profits from the B12 narrative, and vegan doctors who recommend it have been “too gullible.”
9. Practical advice given: don’t supplement, and buy unfortified nutritional yeast specifically to avoid added B12.

Every one of these fails. Let’s establish how we’ll know that, and then go claim by claim.

First, the ground rules: how the hierarchy of evidence works

What is the hierarchy of scientific evidence?

Science doesn’t treat all information equally. At the bottom of the evidence pyramid sit anecdotes, testimonials, and expert opinion: “I’ve been fine for 45 years.” One step up is mechanistic reasoning and lab findings. Then case reports and case series, which document individual patients. Then observational studies: cross-sectional surveys and prospective cohorts following thousands of people over years. Above those sit randomized controlled trials (RCTs), where an intervention is tested against a placebo with neither participants nor researchers knowing who got what. At the top are systematic reviews and meta-analyses that pool every available trial, and the consensus statements of expert bodies that weigh all of it together.

The reason the pyramid is shaped that way is simple: each step up removes more ways of fooling yourself. Anecdotes can’t distinguish luck from causation. Observational studies can’t fully separate correlation from confounding. RCTs can. Meta-analyses check whether RCTs agree with each other.

Why one person’s testimony isn’t proof

Notice what the interview’s entire case rests on: one man’s anecdote, the bottom rung, plus a conspiracy theory whose only function is to explain away every rung above it. That’s not an alternative body of evidence. It’s a request to exempt one belief from evidence entirely.

The case for B12, by contrast, is supported at every level of the pyramid, and we’ll climb the whole thing.

What vitamin B12 actually is, in ninety seconds

What does vitamin B12 do in the body?

Vitamin B12 (cobalamin) is a cobalt-centered molecule made only by certain bacteria and archaea, not by plants, and not by animals [1, 27, 28]. In human biochemistry it is a cofactor for exactly two enzymes, and both matter enormously [1, 2].

Methionine synthase converts homocysteine into methionine, feeding the methylation reactions that maintain DNA, proteins, lipids, and neurotransmitters. When B12 runs short, homocysteine (a metabolite associated with stroke, cognitive decline, and cardiovascular risk) backs up in the blood [22, 23], and DNA synthesis in rapidly dividing cells falters. In the bone marrow, that produces the hallmark megaloblastic anemia: oversized, malformed red blood cells, plus fatigue, glossitis, and low blood counts [1, 2].

Methylmalonyl-CoA mutase handles a step in fatty-acid and amino-acid metabolism. When B12 is missing, its substrate spills over as methylmalonic acid (MMA), which is why elevated MMA is the single most specific laboratory fingerprint of B12 deficiency [1, 2].

What are the symptoms of B12 deficiency?

B12 also maintains the myelin sheath insulating your nerves. Deficiency causes numbness and tingling in the hands and feet, balance problems, and, if untreated, subacute combined degeneration of the spinal cord, with damage visible on MRI [18], progressing in advanced cases to vision loss, psychiatric symptoms, and dementia [1, 2]. Two facts make this especially unforgiving: the neurological damage can occur without any anemia to warn you, and past a certain point it can be permanent [1, 2].

How long does it take to become B12 deficient?

Now the kinetic detail that explains almost everything about this controversy: your liver stores roughly 1 to 5 milligrams of B12 against a daily requirement of about 2.4 micrograms, a thousand-to-two-thousand-day buffer, stretched further by efficient recycling through bile [2]. The U.S. National Institutes of Health states plainly that deficiency symptoms can take years to appear after intake stops [2]. Keep that number in mind. It is the reason “I feel fine” is exactly what a person sliding toward deficiency is expected to say, right up until they don’t.

Claim 1: “Vegans don’t need B12”

Let’s climb the pyramid.

What do health authorities actually say about vegans and B12?

The U.S. Institute of Medicine set the adult RDA at 2.4 micrograms per day in 1998 [3]. The German-Austrian-Swiss (D-A-CH) nutrition societies raised their reference value to 4.0 micrograms per day in 2019 after reviewing newer biomarker data [4]. The Academy of Nutrition and Dietetics, the largest body of nutrition professionals in the world, reaffirmed in its current position paper, updated in 2025, that appropriately planned vegan dietary patterns can be nutritionally adequate and offer long-term health benefits, and that vegans need a reliable source of B12 from fortified foods or supplements [5]. (The 2025 update scoped its formal position statement to adults; it simply did not undertake to review children and pregnancy this cycle, which is a limit on what the paper assessed, not a reversal of the long-standing conclusion, echoed by the NIH and by vegan dietitians below, that well-planned vegan diets are appropriate across the lifespan with a B12 source in place.) The NIH’s Office of Dietary Supplements lists vegans and the infants of unsupplemented vegan mothers among the groups at defined risk of deficiency [2]. VeganHealth.org, run by vegan registered dietitians as a project of Vegan Outreach, calls B12 “arguably the most important nutrient in vegan nutrition” and describes the necessity of supplementation as the overwhelming consensus of both mainstream nutrition science and vegan health professionals [33]. In 2001, an open letter titled “What Every Vegan Should Know About Vitamin B12” was signed by a long list of vegan health professionals and vegan organizations saying the same thing [34]. These are not meat-industry fronts. They are the vegan movement’s own institutions.

What do the systematic reviews show?

The definitive modern review of B12 deficiency, Green and colleagues in Nature Reviews Disease Primers, lays out the biochemistry, epidemiology, and clinical course in exhaustive, multi-author detail [1]. A systematic review by Pawlak and colleagues of serum B12 across vegetarian and vegan populations found high deficiency prevalence in every age group studied among those not using a reliable source, in some cohorts affecting the large majority [20]. Rizzo and colleagues’ review of B12 among vegetarians reaches the same conclusion and details assessment and supplementation strategy [19].

Is there randomized controlled trial evidence for B12?

This is where the “it’s all confounded” excuse dies. In a 12-week randomized, double-blind, placebo-controlled trial in vegans, a B12-fortified toothpaste (B12 and literally nothing else, against an identical placebo paste) significantly improved serum B12 and holotranscobalamin, and lowered MMA [16]. The same design in older adults produced the same result [17]. A 2018 Cochrane systematic review of RCTs found high-dose oral B12 comparable to intramuscular injections at normalizing status in deficient patients [6], building on the landmark 1998 randomized trial by Kuzminski and colleagues [7]. Add a single micronutrient, change nothing else, and deficiency markers normalize against placebo. That is causation, demonstrated.

What do studies of real vegans find?

One of the earliest studies of vegans, from the U.K. in 1955, documented overt deficiency among unsupplemented vegans, including nerve damage and dementia [33]. In the decades since, study after study of vegans not using supplements or fortified foods, including raw-food vegans specifically [15], macrobiotic communities, and long-term vegans in multiple countries, finds depressed serum B12, elevated MMA, and elevated homocysteine, while vegans who supplement show healthy values, frequently better than meat-eaters’ [19, 20, 33].

The medical literature also contains a long, grim catalog of individual unsupplemented vegans, and especially breastfed infants of unsupplemented or under-supplemented vegan mothers, presenting with severe deficiency: developmental regression, brain atrophy, failure to thrive [2, 13]. These are the cases the “I feel fine” anecdote conveniently never has to meet.

Are there any reliable plant sources of B12?

There are no reliable plant sources of B12, a conclusion reached not by fiat but by decades of analytical chemistry showing that the “B12” detected in spirulina and most algae is predominantly pseudo-B12, an analogue inactive in humans [27], and that fermented foods, soil-dusted produce, and intestinal bacteria (which produce B12 in the colon, downstream of the ileal absorption site) cannot be counted on [27, 28, 33]. VeganHealth maintains an encyclopedic, referenced page testing every proposed plant source, including nori, chlorella, tempeh, duckweed, and organic produce, and the pattern is consistent: unreliable, inactive, or unproven [33].

Is B12 well-researched, or is the science thin?

This isn’t a thin literature being over-read. B12 is among the most exhaustively researched nutrients in the history of science: a century of continuous investigation, four Nobel laureates across two prizes [32], and a body of work anyone can sample by typing “cobalamin” into PubMed and watching tens of thousands of papers come back. It has its own Cochrane review [6], its own commissioned Nature Reviews Disease Primer [1], its own Institute of Medicine reference-intake process [3], and an NIH-commissioned international consensus review devoted to nothing but the fine points of measuring its biomarkers [39]. Fields this mature are not where century-long errors of the kind the interview alleges go to hide; they are precisely where such errors get found. So the claim on the table isn’t that nobody has checked whether humans need B12. It’s that almost no nutrient in history has been checked harder.

Every rung of the pyramid, one answer. This is what “settled” looks like.

Claim 2: “B12 was invented in a Merck laboratory”

Was vitamin B12 invented by a pharmaceutical company?

This confuses isolating a molecule with inventing one, and gets the history backwards.

The story doesn’t start with Merck. It starts with pernicious anemia, a disease that was, as the name records, essentially a death sentence. Patients wasted away with anemia and neurological collapse, and no treatment existed. In the 1920s, George Minot and William Murphy discovered that feeding patients large amounts of liver reversed the disease, work that (with George Whipple’s) won the 1934 Nobel Prize in Physiology or Medicine [32]. Something in liver was keeping people alive; for two decades, chemists raced to find out what.

Who actually discovered vitamin B12?

In April 1948, two teams isolated the red crystalline compound independently, publishing within a single week of each other: Rickes, Folkers, and colleagues at Merck in the United States in Science, and E. Lester Smith at Glaxo in the United Kingdom in Nature [31]. Nature’s contemporaneous editorial marveled that the twenty-two-year quest opened by Minot and Murphy was ending with single doses around 15 micrograms doing what pounds of liver had done, and within months Randolph West demonstrated that injecting just 5 to 10 micrograms of the pure substance produced a near-maximal blood-cell recovery in pernicious anemia patients [31]. Dorothy Hodgkin then solved the molecule’s structure by X-ray crystallography in 1956, among the most complex ever determined at the time, work central to her 1964 Nobel Prize in Chemistry [32].

So: a fatal disease, a liver cure, a twenty-year international hunt, two independent teams, four Nobel laureates across two prizes, and a molecule that bacteria have been synthesizing for on the order of billions of years. Merck no more invented B12 than Galileo invented the moons of Jupiter. The claim in the interview, sourced to a recent anti-B12 book, that scientists were just numbering vitamins arbitrarily (“we’ll make B1, make it 12”) isn’t an alternative history. It’s not history at all.

One final inversion worth savoring: far from being a weapon against veganism, the isolation of B12 is arguably what made lifelong veganism medically realistic, a point the vegan dietitians at VeganHealth make themselves [33]. The U.K. Vegan Society was founded in 1944; the one nutritional question hanging over the whole project was answered four years later, by the very discovery this narrative teaches vegans to resent.

Claim 3: “Deficiency symptoms are really chemical poisoning blamed on B12”

Is B12 deficiency actually chemical poisoning in disguise?

This claim has the virtue of being cleanly falsifiable, and it has been falsified about as thoroughly as anything in clinical medicine.

B12 deficiency isn’t a vague syndrome pinned on people; it has a specific, multi-marker laboratory fingerprint: low serum B12 and holotranscobalamin, elevated methylmalonic acid, elevated homocysteine, macrocytic red cells, hypersegmented neutrophils on a blood smear, and characteristic spinal-cord changes on MRI in advanced neurological cases [1, 2, 18]. No pattern of “chemical poisoning” produces that fingerprint.

And here is the decisive test: give the patient a few micrograms of cobalamin, change nothing else about their diet, environment, or “toxin exposure,” and every element of the fingerprint reverses. The blood counts recover. MMA falls. Homocysteine falls. In the placebo-controlled trials cited above, that reversal happens against a placebo group whose environment is identical [6, 16, 17]. Pernicious anemia patients have been pulled back from the edge of death by B12 alone since 1948, exactly as liver extract did before that [31]. If the symptoms were caused by petrochemical exposure, twenty micrograms of a single vitamin would not cure them. It does. Case closed, not rhetorically, but experimentally.

The interview’s related move, “people are poisoned by chemicals and the government blames a deficiency,” also has the incentive structure exactly backwards. A deficiency diagnosis leads to a pennies-per-week generic vitamin. If institutions wanted a narrative that maximized dependence and revenue, an untreatable environmental-poisoning story would serve far better than one solved by the cheapest supplement in the store.

Claim 4: “It’s the only vitamin not in plants, so it’s suspicious”

Why is B12 the one vitamin not found in plants?

The explanation is not a conspiracy. It’s microbiology, and it’s been understood for decades.

B12 is synthesized exclusively by certain bacteria and archaea [1, 27, 28]. Plants neither make it nor need it, which is why it isn’t in your fruit trees. Animals don’t make it either; they accumulate it. Ruminants like cows carry fermentation vats (rumens) full of B12-producing microbes, which need dietary cobalt to do the job. Pigs, chickens, and primates historically got theirs from soil, insects, water, and fecal contamination of food [28, 33]. Pre-industrial humans picked up traces the same unglamorous ways: unwashed produce, untreated water, and generally living in closer contact with microbial filth than any of us would now accept.

Did modern sanitation cut off our B12 supply?

Modern sanitation, meaning chlorinated water, washed produce, and food-safety standards, closed that pipeline. That trade was overwhelmingly worth it; clean water has saved more lives than any medicine. But it means that in the modern world, everyone’s B12 traces back to a managed microbial source: either bacteria cultured in fermentation tanks and put into your supplement or fortified food directly, or bacteria in and around farmed animals, increasingly propped up (as we’ll see under Claim 8) by supplements added to the animals’ feed [29, 30].

So the “suspicious” fact dissolves into an ordinary one: B12 is not an animal nutrient that vegans are missing. It is a bacterial nutrient that every animal on Earth, humans included, must obtain from microbes one way or another. Vegans simply take the direct route.

Notice that this single fact also detonates two other claims from the interview. Supplemental B12 cannot be “refined petroleum” (Claim 5 below), because it is produced by fermentation. And it requires no animal exploitation whatsoever, which will matter when we get to the ethics.

Claim 5: “Cyanocobalamin is cyanide, so they’re telling us there’s poison in it”

How much cyanide is actually in a cyanocobalamin tablet?

Time for arithmetic, which the fear-mongering never survives.

Cyanocobalamin is a large molecule with a mass of about 1,355 daltons, of which the cyanide group contributes about 26, roughly 2 percent by weight. A large 1,000-microgram tablet therefore contains about 20 micrograms of cyanide; the standard 50-microgram daily maintenance dose contains about 1 microgram. Your body routinely detoxifies cyanide via the enzyme rhodanese into thiocyanate and excretes it, and the trace amounts in everyday plant foods that contain cyanogenic compounds (flaxseed, almonds, lima beans, cassava) comfortably exceed what’s in a B12 tablet. Twenty micrograms of cyanide is, quite literally, less than you’d get from a pinch of ground flax.

Isn’t B12 used as the antidote for cyanide poisoning?

It is, and that’s the deeper irony. Cobalamin binds cyanide so avidly that a B12 form, hydroxocobalamin, is the modern hospital antidote for acute cyanide poisoning, given intravenously at five grams: five thousand times the size of a large 1,000-microgram supplement tablet. And note what that antidote reaction produces. Per the drug’s own FDA labeling, hydroxocobalamin binds cyanide to form nontoxic cyanocobalamin, which the body then simply excretes in urine. The molecule the interview frames as a poison-delivery device is, in actual emergency medicine, the safe end-product of cyanide detoxification, what your body makes cyanide into on the way out the door.

Even the interviewer, to her credit, noted mid-conversation that plant-based physicians like Dr. Michael Klaper have explained the cyanide content is far too small to matter. That’s correct. The video acknowledges the debunk and keeps the fear anyway, which tells you the fear was never really about the chemistry.

Are there people who should avoid cyanocobalamin?

Two honest nuances, because airtight means acknowledging real edge cases. In kidney failure, where clearance of cyanide metabolites is impaired, methylcobalamin or hydroxocobalamin are reasonable alternatives to high-dose cyanocobalamin, in consultation with a physician [23, 33]. And it has been speculated, though not confirmed, that heavy smokers, whose cyanide load is already elevated, might prefer non-cyano forms [35]. For everyone else, cyanocobalamin is the best-studied, most stable, cheapest form, and the NIH notes no evidence that absorption differs between forms [2, 11]. Precision is not scandal.

Claim 6: “Vitamins are refined petroleum”

Is vitamin B12 made from petroleum?

Whatever one thinks of other supplements, this is flatly, checkably false for B12, and B12 is the supplement under discussion.

Vitamin B12 cannot be practically synthesized chemically at commercial scale. Its total chemical synthesis, completed in 1972 by the collaborating teams of Robert Woodward at Harvard and Albert Eschenmoser at ETH Zürich, took almost twelve years and more than a hundred chemists (91 postdoctoral researchers and 12 doctoral students from 19 nations), and stands in chemistry lore as the monument that proved nobody would ever make B12 that way for production. Instead, all commercial B12 is produced the way nature produces it: by bacterial fermentation, using cultured microorganisms such as Propionibacterium freudenreichii, Pseudomonas denitrificans, and Ensifer adhaerens [29, 30]. European regulators’ safety dossiers for B12 describe exactly this: cyanocobalamin “produced by fermentation,” purified to better than 99 percent [30].

In other words, the one supplement the evidence says vegans need is, of all things, a purified product of bacterial culture, closer kin to the microbes in your sauerkraut than to anything from an oil refinery. The petroleum story isn’t an exaggeration of the truth. It’s the opposite of it.

Claim 7: “I’m 72 with a biological age of 22, never tested, never sick, so B12 is a scam”

This is the emotional core of the video, so let’s give it a full and fair autopsy.

Why doesn’t “I’ve been vegan 45 years and I’m fine” prove anything?

One person’s testimony cannot distinguish “B12 is unnecessary” from “this individual is lucky, unusual, or not yet symptomatic.” That’s not an insult; it’s the entire reason evidence hierarchies exist. And it comes wrapped in textbook survivorship bias: the vegans who followed this path and developed anemia, neuropathy, or worse are not hosting livestreams about their vitality. Some of them are in the case-report literature instead [2, 13, 33]. VeganHealth even maintains a page on the rare, genuinely documented long-term exception, precisely because outliers exist and precisely because outliers don’t set guidance, any more than a chain-smoking 95-year-old debunks oncology [33].

Can you really coast for decades without B12?

Recall that the liver banks a multi-year supply, recycled efficiently through bile [2]. And by his own telling, his history isn’t even 45 unsupplemented years: after his first 15 to 20 years without, he took weekly methylcobalamin for a stretch of years and tapered off only about a decade before the interview, while also spending two months every year at a health institute. A previously supplemented adult coasting on hepatic stores, roughly a decade into depletion, feeling fine? That is not a refutation of B12 science. It is a prediction of B12 science, the quiet middle of the curve, before the part nobody wants.

Can grip strength or “biological age” tests detect B12 deficiency?

No. “Never tested” is not evidence of health; it’s the absence of evidence, engineered. Balance time, push-ups, grip strength, and sit-to-stand are perfectly nice functional measures, and they measure precisely none of the things B12 deficiency degrades first. Serum B12, MMA, and homocysteine are cheap, specific, and validated [1, 2]. Refusing them while claiming certainty about what they’d show is not confidence. It’s a coin flip with the results face-down, and because B12 neuropathy can begin subtly and become irreversible without ever causing anemia [1, 2], it’s a coin flip where one side can’t be un-flipped.

What about healthy old people in Okinawa with “terrible” blood levels?

The claim arrives unsourced, and the traditional Okinawan diet includes fish and pork, so whatever it shows, it isn’t about veganism. What the actual gerontological literature shows is the opposite direction of concern: roughly one in five American adults over 60 has at least one abnormal B12 biomarker [26], deficiency in older adults runs anywhere from 3 to 43 percent depending on the definition [2], and low B12 status with elevated MMA and homocysteine tracks with worse cognitive and vascular outcomes, not secret vitality [22, 23].

None of this requires doubting that the man feels wonderful. It requires only the one thing the whole video refuses: distinguishing between feeling fine and being fine, a distinction B12, with its years-long fuse and its silent, sometimes permanent neurological endgame, was practically designed to exploit.

Claim 8: “Follow the money”

Fine. Let’s follow all of it, in three steps.

Who profits from telling vegans to take B12?

Almost nobody, which is the problem with the theory. Cyanocobalamin is off-patent, generic, and produced at industrial scale by fermentation. A full year of the standard vegan regimen, one 2,000-microgram tablet a week, costs a few dollars, less than a single trip to a juice bar. There is no moat, no monopoly, no recurring-prescription lock-in; the FDA-recognized forms are made by many competing manufacturers worldwide. If a shadowy complex needed a profit engine, it could hardly have designed a worse one than the cheapest vitamin on the shelf. Meanwhile (and this is said without malice, only symmetry) the interview itself promotes books, a paid festival, and retreat appearances. If we are auditing incentives, we audit everyone’s.

The strongest, most persistent voices telling vegans to supplement B12 are vegan institutions: the 2001 open letter signed by vegan health professionals and vegan organizations [34], VeganHealth.org (a project of the nonprofit Vegan Outreach, which explicitly declines to recommend brands) [33], the dietitians who authored the Academy’s position paper [5], and plant-based physicians such as Greger, Klaper, and Barnard, who are famously skeptical of the supplement industry in general and carve out B12 as the exception the evidence forces [35]. Their motive is transparent: every preventable deficiency case becomes an “I was vegan until it wrecked my health” story, and those stories drive people back to animal products. Vegan advocates push B12 because they are protecting vegans and veganism, not a pharma balance sheet.

How much B12 goes to animal agriculture?

Here is the fact that turns the whole conspiracy inside out. World production of the pure vitamin has grown to roughly 100 tonnes a year, and the peer-reviewed breakdown of that market puts the animal-feed sector at about 55 percent of sales, more than food and pharmaceutical uses combined [29]. European food-safety regulators maintain approval dossiers for fermentation-produced cyanocobalamin explicitly as a nutritional feed additive “for all animal species” [30]. The reason is structural: pigs and poultry cannot synthesize B12 and, raised in confinement away from soil, insects, and manure-borne microbes, they receive it directly in their vitamin premix.

Why do farmers give cattle cobalt supplements?

Ruminants like cattle get the precursor instead. Cobalt is routinely supplemented in feed and pasture so their rumen microbes can build the B12 for them, a practice dating to 1935, when mysterious fatal wasting diseases of livestock in Australia and New Zealand (“bush sickness,” “coast disease,” “enzootic marasmus”) were traced to cobalt-deficient soils and cured with trace cobalt. The same disease had baffled farmers worldwide under a dozen local names, from “cobalt pine” in Scotland to “salt sickness” in Florida. And since a ruminant’s only known biological use for cobalt is building cobalamin, ruminant cobalt deficiency simply is B12 deficiency: emaciated, dying sheep recovered dramatically on tiny oral doses of cobalt, an accidental continent-scale demonstration of this vitamin’s essentiality in mammals, run on livestock decades before anyone had a culture war to fight about it. All of this predates internet veganism by the better part of a century.

Is it true that 90 percent of B12 supplements go to livestock?

No, and it’s worth saying so plainly, because airtight cuts both ways. That figure circulates widely in vegan spaces, but it has no primary source; it appears to trace to an uncited 2013 newspaper op-ed and has been repeated ever since without one. The documented figure is the roughly 55 percent of direct B12 sales [29], plus the near-universal cobalt supplementation of ruminants, which doesn’t show up in B12 sales statistics at all. “A majority, plus the cobalt pipeline” is the defensible claim, and it is entirely sufficient.

Because look at what it means. If B12 were an invention designed to smear veganism as deficient, the conspirators forgot to tell the meat industry, which quietly operates the largest B12 supplementation program on the planet just to keep its products B12-positive. For the typical modern omnivore, meat is a B12 supplement with extra steps and a middleman with a heartbeat. Vegans who take the tablet aren’t falling for the system. They’re skipping it.

“Nobody in the Zen monastery took supplements”: the naturalism trap

Can a natural diet require supplements?

The interview leans repeatedly on an appeal to nature: a divine universe wouldn’t require pills; monks never supplemented; our bodies recycle what they need if our consciousness is right.

But “natural” is a description, not a credential. The natural human condition also featured child mortality near half, endemic intestinal parasites, and pernicious anemia as an untreatable death sentence [31, 32]. Our ancestors’ B12 pipeline was environmental grime (untreated water, unwashed roots, microbial contamination), an unreliable source even then, and one that modern sanitation has rightly closed [28, 33]. VeganHealth addresses this squarely in an essay titled “Can a Natural Diet Require Supplements?”: the question isn’t whether supplementation is natural, it’s whether the nutrient is necessary, and it is [33].

Nor is B12 unique in this. Modern omnivorous life is quietly fortified everywhere you look: iodine in salt (ending endemic goiter), vitamin D in milk (ending rickets), folic acid in flour (preventing neural-tube defects), and B12 itself in the feed of the animals people eat [29, 30]. Nobody argues that iodized salt disproves the human diet. A vegan taking a weekly chewable is doing exactly what the whole modern food system already does, just honestly, and without the animal.

Can gut bacteria make enough B12 for us?

They genuinely do make B12, but in the colon, downstream of the ileum where B12 is absorbed. It leaves the body unused [33]. This isn’t a hypothesis; it’s anatomy, and it was tested directly in the raw-vegan trial described below, where probiotic supplements failed to fix B12 status while tablets and fortified yeast succeeded [15].

B12 deficiency is not a “vegan disease”: the omnivore numbers

How common is B12 deficiency in meat eaters?

Here is the part that dissolves the “B12 exposes veganism’s weakness” framing entirely, because deficiency is common among people who eat meat every day.

Per the NIH’s summary of national survey data (NHANES 2007–2018), about 3.6 percent of all U.S. adults are outright B12 deficient and about 12.5 percent are insufficient. In a population where vegans are a small minority, that means the overwhelming majority of those deficient Americans are omnivores [2, 25]. Raise the cutoff to the level many clinicians treat as the gray zone and, in earlier NHANES data, 26 percent of adults fell below it [2]. Authoritative reviews put low or marginal B12 status at up to 40 percent of Western populations [1, 2]. Among older adults it’s worse: depending on the definition, 3 to 43 percent of community-dwelling seniors are deficient, roughly one in five Americans over 60 has at least one abnormal B12 biomarker [26], and atrophic gastritis, which cripples absorption of food-bound B12 specifically, affects 8 to 9 percent of people over 65 [2].

Why does the government tell everyone over 50 to take B12?

Because of that last mechanism. Since 1998, the Institute of Medicine has advised that everyone over 50, steak lovers included, get most of their B12 from fortified foods or supplements, because the crystalline form absorbs even when food-bound B12 no longer does [2, 3]. Read that again: mainstream nutrition policy tells half the adult lifespan of omnivores to do exactly what vegans are told to do. If needing supplemental B12 “disproves” a diet, it disproves being over fifty.

The pattern holds all the way down. Pernicious anemia, an autoimmune malabsorption disease with nothing to do with diet, remains the most common cause of clinically evident B12 deficiency in the world [2]. Metformin, the first-line diabetes drug taken by tens of millions, depletes B12, and so do the proton-pump inhibitors and acid blockers in half of America’s medicine cabinets [2].

Does eating meat actually protect against B12 deficiency?

Less than you’d think. In the Framingham Offspring cohort of 2,999 ordinary, overwhelmingly omnivorous Americans, 39 percent had plasma B12 below 258 pmol/L, 17 percent below 185, and 9 percent below 148, with little difference between age groups [24]. Deficiency prevalence was roughly half as high among regular fortified-cereal eaters as among those who ate none (12 versus 23 percent), and similarly split between the highest and lowest dairy consumers (13 versus 24 percent), but showed no significant differences across meat-intake tertiles [24]. The authors’ conclusion was that supplements, fortified cereal, and milk appear protective against low B12. Eating animals every day did not protect these people; the fortification infrastructure did.

Is B12 insufficiency common in pregnancy, even for meat eaters?

Yes, and this is where the “vegan problem” framing falls apart most completely, because the stakes are highest exactly where the diet matters least. A 2016 systematic review and meta-analysis by Sukumar and colleagues, pooling dozens of studies, found vitamin B12 insufficiency in roughly one in five pregnant women in the first trimester (21 percent), 19 percent in the second, and 29 percent by the third [51]. The authors stated it plainly: B12 insufficiency during pregnancy “is common even in nonvegetarian populations,” and B12 levels fall as pregnancy progresses [51]. (In fairness, the same review found no consistent link between that insufficiency and low birth weight, so this is about maternal status, not a proven birth-weight harm.) The lesson isn’t that vegan pregnancy is uniquely risky. It’s that pregnancy raises B12 demand for everyone, which is why a reliable source matters for every pregnant woman, and simply matters most for one following a plant-based diet.

So the honest picture is this: B12 adequacy in the modern world is a human engineering problem, solved for everyone, omnivores very much included, by fortification and supplementation running quietly in the background. Vegans aren’t the exception to the system. They’re just the ones who can see it.

What actually happens to vegans who don’t supplement

What percentage of vegans are B12 deficient?

Since the video’s thesis stands on one unsupplemented vegan feeling great, it’s worth stating what the record shows about unsupplemented vegans as a group.

The earliest cohort data, a 1955 study of British vegans, already documented overt deficiency, including nerve damage and dementia, in the pre-supplement era [33]. Modern systematic reviews of serum B12 in vegetarians and vegans find high deficiency prevalence in every age group among those without a reliable source, ranging in some studied groups to a large majority [19, 20]. The single most cited snapshot comes from the EPIC-Oxford cohort: among 689 British men, 52 percent of vegans were vitamin B12 deficient (below 118 pmol/L), versus 7 percent of vegetarians and exactly one of 226 omnivores. Tellingly, serum levels showed no association with years on the diet, consistent with stores depleting silently on their own schedule [37].

Do raw vegans get enough B12 from fermented foods?

No. A two-year Finnish study of raw “living food” vegans, whose participants ate roughly two kilograms of fermented foods a day on the theory that fermentation and friendly gut bacteria would supply their B12, watched status deteriorate slowly but consistently in six of the nine followed [33, 38]. Studies of non-supplementing vegans consistently show elevated homocysteine, the stroke- and dementia-associated metabolite, where supplementing vegans show healthy levels [22, 23, 33].

Two studies deserve special attention from the raw-food and fruit-festival community the video was addressing. Donaldson studied 49 followers of a mostly raw vegan diet (the Hallelujah Acres program, mostly two to four years in) and found 76 percent with serum B12 below 300 pg/mL. In follow-up, sublingual B12 tablets and fortified nutritional yeast corrected the elevated methylmalonic acid, while probiotic supplements did not [15], simultaneously demonstrating the fix and closing the “my gut bacteria will handle it” exit. And Crane’s study of vegans found that chewing a B12 tablet raised serum levels dramatically where swallowing the identical tablet whole barely moved them, and that the non-responders responded promptly once they started chewing [14].

What happens to babies of B12-deficient vegan mothers?

At the bottom of this literature sit the cases that should end the debate for anyone with a conscience: severely deficient breastfed infants of vegan mothers, including a mother who believed she was covered because she took a multivitamin containing B12 [13]. Infant B12 deficiency can cause developmental regression, failure to thrive, and even measurable brain atrophy [2, 13, 49, 50]. There is a genuinely hopeful wrinkle here, and it’s worth stating precisely: in case reports, that cerebral atrophy has substantially reversed on brain imaging after B12 treatment [49, 50]. But imaging reversal is not the same as full recovery. The same case literature, and authoritative reviews, warn that long-term cognitive, motor, and language deficits often persist, and that outcome depends heavily on how early treatment begins [1, 49]. Which is exactly the point: caught early, much can be undone; caught late, the damage can be permanent. These children did not choose a philosophy. They inherited an information environment, the very one videos like this pollute.

People really do differ: the heterogeneity question, taken seriously

The most sophisticated version of the anti-supplement argument isn’t “B12 is fake.” It’s “people vary, so maybe some of us absorb better, recycle better, or host bacteria that cover us.” That question deserves a straight answer, because the variation is real, measured, and, properly understood, cuts the opposite way from how it’s used.

Is there a gene that affects B12 levels?

Yes, several. Genome-wide association studies have mapped common variants that shift B12 status. The strongest hit is FUT2, the classic “secretor” gene: people homozygous for the non-secretor variant run roughly 10 to 25 percent higher total serum B12, a finding replicated across European, Chinese, and Indian populations, with further loci identified in TCN1, CUBN, and MUT [40]. There is now an entire curated database of B12-associated genomic variants [41].

Two things follow. First, an amusing nuance: when researchers split serum B12 into its two carrier fractions, the famous FUT2 effect ran through holo-haptocorrin, the biologically inert fraction, while the bioactive holotranscobalamin was untouched [48]. The best-known “B12 gene” mostly moves the part of the number that doesn’t feed your cells. Second, genetic variation also runs in the unlucky direction: mutations in the intrinsic-factor gene (GIF) and in CUBN (Imerslund–Gräsbeck syndrome) cause deficiency in people eating meat at every meal [40]. Heterogeneity is not a one-way ticket to safety.

Can some people just absorb or recycle B12 better?

In rate, yes; in direction, no. Intrinsic factor capacity spans a spectrum, from normal abundance through age-related atrophic gastritis down to pernicious anemia’s total loss [1, 2]. Fractional absorption genuinely adapts: at tiny intakes the intrinsic-factor system captures around half of what passes through, versus roughly one percent of a large dose [2, 10]. That is efficient, but efficiency cannot conjure micrograms that were never swallowed. And enterohepatic recycling, meaning reclaiming the B12 your bile secretes, differs between people; classic work by Victor Herbert estimated that a highly efficient recycler with full stores could take on the order of decades to run dry, versus a few years for someone who recycles poorly [33, 44]. This is the honest explanation for the rare, genuinely documented long-term unsupplemented exception [33], and, notably, for the interview’s own protagonist. Variance in the speed of depletion is fully compatible with, indeed predicted by, everything in this article. What it does not do is reverse the arrow.

Do Indian vegetarians prove humans can adapt to make their own B12?

The best card in this deck is a real Nature paper: in 1980, Albert, Mathan, and Baker showed that Pseudomonas and Klebsiella cultured from the small intestines of healthy southern Indian subjects could synthesize assay-active B12 in vitro, and proposed this might explain why some Indian vegetarians resisted overt deficiency [42]. Cited fairly, it’s intriguing. Read fully, it dissolves. The finding was test-tube synthesis by organisms from people carrying heavy small-bowel colonization, itself a product of high-microbial-exposure environments, not something a sanitized Westerner can will into existence.

The modern microbiome literature then delivers the twist: people with heavy small-intestinal bacterial loads tend to have lower B12 status, because bacteria bind cobalamin with an affinity rivaling intrinsic factor’s, and the great majority of gut microbial species consume corrinoids rather than donate them [43]. That’s why small intestinal bacterial overgrowth is a recognized cause of B12 deficiency [2, 43]: when microbes do live upstream of your absorption site, they mostly steal. And the population-level check is brutal for the adaptation story. Indian lacto-vegetarian populations, the very groups the hypothesis was built around, show some of the highest measured rates of B12 deficiency and elevated homocysteine in the world [33, 45]. As for the colon, where our resident bacteria really do make B12, it sits downstream of the ileal receptors, and while researchers have recently probed possible trace colonic absorption with isotope-labeled B12, nothing found changes the bottom line: no non-supplementing vegan population anywhere has been shown to maintain adequate status from internal synthesis [33]. There is no documented route by which colonic B12 meaningfully “finds its way up.”

What does all this variation mean for me personally?

It explains the variance: why one unsupplemented vegan crashes in two years and another coasts for twenty, why exceptions exist, why your friend’s anecdote and the case reports can both be true. But you cannot feel your FUT2 genotype, your intrinsic-factor output, or your biliary recycling efficiency from the inside. The only way to know which end of the distribution you occupy is the blood panel. That converts heterogeneity from an argument against supplementing into the strongest possible argument for testing: if you sincerely believe you’re the exception, the scientific response is a serum B12 and MMA test, not a vow of confidence. And the cost asymmetry does the rest of the work. Being wrong about needing a supplement costs pennies; being wrong about being special can cost your spinal cord.

The strongest counter-evidence on record: an honest audit

Is there any study showing humans don’t need B12?

A document that intends to be airtight has to do the thing its opponents never do: go looking for the best evidence against its position and print what it finds. So, how would one assess the contrary literature? Search the systematic reviews, the Cochrane library, and the position statements of every nutrition body on Earth for any conclusion that humans don’t need B12, that deficiency isn’t a real clinical entity, or that vegans can safely skip a reliable source. The result of that search is: nothing. No such review exists. No such trial exists. No retraction, no replication crisis, no dissenting expert body. The most recent regulatory movement went the other direction, with the German-speaking nutrition societies raising their reference intake in 2019 [4]. The essentiality of B12 has the same evidentiary status as the essentiality of iron.

What does exist, meaning the studies that genuinely show “conflicting results” and that a sharp critic could wave, falls into three clusters, each of which turns out to be about a different question. Here they are, at full strength.

Why did the big B-vitamin heart trials fail?

Through the 2000s, a series of large RCTs (HOPE-2, NORVIT, SEARCH, VITATOPS and others) gave B vitamins to tens of thousands of mostly B12-replete cardiovascular patients. Homocysteine fell on schedule; heart attacks mostly didn’t [22]. These null results are real, and they killed the strong version of the homocysteine-CVD hypothesis, though the meta-analytic picture retains a modest stroke reduction, clearest in low-folate populations [22]. But notice what these trials tested: whether lowering homocysteine in non-deficient people prevents cardiovascular events. That is not the question “do humans need B12,” any more than a failed trial of extra water for marathon performance would show hydration is a myth. Not one of these trials enrolled deficient people and asked whether repletion fixes deficiency, the question that was settled in 1948 and re-confirmed by every placebo-controlled repletion trial since [6, 7, 16, 17].

Is it true that a trial found B12 caused harm?

Yes, in one specific population, and it’s worth knowing exactly what it showed. The single most legitimate “B12 caused harm” result in the literature is the DIVINe randomized trial: 238 patients with diabetic kidney disease given high-dose B vitamins (folic acid 2.5 mg, B6 25 mg, and a full 1,000 mcg of cyanocobalamin daily) or placebo for three years. The vitamin group’s kidney function declined faster (a GFR drop of 16.5 versus 10.7) and vascular events roughly doubled, with the harm concentrated in participants whose kidney function was already below half of normal [46]. This is a real randomized finding and this article does not shrink from it. It is precisely why the dosing guidance in the protocol below routes people with kidney disease away from high-dose cyanocobalamin and toward methyl- or hydroxocobalamin under a physician’s care [23, 46], the same conclusion drawn by the trial’s own investigators [23]. But look at what DIVINe is and isn’t: megadoses, in renal failure, of the one form whose cyanide-handling requires working kidneys. It is a boundary-condition finding about form and dose in a sick subgroup, and the fact that an RCT found that boundary and the guidance absorbed it is the system functioning, not the edifice cracking. Nobody in DIVINe was deficient; nothing in DIVINe suggests deficiency is fictional.

Does B12 cause lung cancer?

The study anti-supplement critics cite most is Brasky and colleagues’ 2017 analysis of the VITAL cohort (77,118 people, 808 lung cancers), which found that men, not women, taking high-dose individual B12 supplements (over 55 mcg per day averaged across a decade, and not multivitamins) had roughly double the lung-cancer risk, with the association concentrated in smokers and absent for adenocarcinoma, the subtype least tied to smoking [47]. The authors themselves called it a concern worthy of further evaluation rather than proof, and published critiques note the self-reported exposure, the inability to exclude confounding by smoking intensity, and contradicting studies, including meta-analytic evidence pointing the protective direction for B6 [47]. Set against it stands the randomized evidence: a meta-analysis of 18 RCTs in nearly 75,000 people found B-vitamin supplementation, including B12 up to 2,000 mcg daily, had little or no effect on cancer incidence, cancer death, or total mortality [2]. The practical residue of this cluster is modest and reasonable: don’t megadose recreationally, and if you smoke, the most important B12-related intervention available to you is quitting. What the cluster cannot carry is the load the anti-B12 narrative piles on it, since a study premised on B12’s biological potency cannot double as evidence that B12 is inert or unnecessary.

And that is the pattern across all three clusters, which is worth stating plainly because it is the answer to “are there studies that make this questionable.” Every genuinely rigorous “contrary” result in the record concerns supplementing already-replete people for chronic-disease prevention, or form and dose in special populations. None concerns whether B12 is essential, whether deficiency is real, or whether vegans need a source. The anti-B12 position has no cluster of its own; its citations are books, podcasts, and testimony. Meanwhile, examining the opposition’s best evidence didn’t weaken this article’s recommendations, it sharpened them, adding the kidney caveat and the no-megadosing note. An argument that improves when fed its strongest opposition is the kind you can freeze.

The single most harmful sentence in the video

Should vegans buy unfortified nutritional yeast?

Absolutely not, and among everything said in the interview, this piece of practical advice stands out as uniquely damaging: not merely declining supplements, but deliberately buying unfortified nutritional yeast to avoid the B12 added to foods.

Understand what this means. Nutritional yeast contains no B12 of its own. Saccharomyces cerevisiae is a fungus and cannot synthesize cobalamin; only fortified brands contain it, which is why labels must be checked [2, 33]. Fortified foods are the quiet safety net that partially protects many vegans who never think about B12 at all. The video’s advice cuts that net on purpose, and because of the multi-year storage buffer, its casualties won’t surface until long after the livestream is forgotten, when the connection will be invisible to everyone, including the victims. Some of those victims, statistically, will be children and breastfed infants. There is no charitable reading of health advice whose harm is engineered, by biochemistry rather than intent, to arrive anonymously and years late.

So what should you actually do? The airtight protocol

Everything below reflects the convergent recommendations of the NIH [2], the Cochrane evidence on oral therapy [6], VeganHealth’s dietitian-authored guidelines [33], and Dr. Greger’s regimen from his B12 video series [35]. It is boring, cheap, and it works.

How do I fix a B12 deficiency?

If you’ve had no reliable B12 source for more than a few months, replenish first: VeganHealth suggests 2,000 mcg once daily for two weeks to refill stores [33]. If you’ve been formally diagnosed as deficient, the treatment literature uses 1,000 mcg per day for one to four months depending on how low you started, before dropping to maintenance [35]. High-dose oral therapy corrects deficiency as effectively as injections for most people, a finding once called one of medicine’s best-kept secrets [8], confirmed in a landmark randomized trial in which daily oral dosing actually outperformed monthly injections on the blood markers [7], in a Cochrane review [6], and in subsequent clinical commentary [9]. The reason is that at large doses about 1 percent absorbs by passive diffusion, with no intrinsic factor required [2, 10].

How much B12 should a vegan take per day?

Maintenance dosing with cyanocobalamin, by situation:

• Adults (roughly 18–65), including pregnancy and breastfeeding: 50 mcg once daily, or 2,000 mcg once weekly [33, 35]
• Older adults (65+): higher daily dosing. VeganHealth suggests 500 mcg per day, and Greger’s updated chart uses 1,000 mcg per day [33, 35]. This is evidence-based, not arbitrary: see the trials below.
• Infants (from about 6 months, once solids begin): 5 mcg per day [21]
• Children 4 to 10: 25 mcg per day, roughly half the adult dose [21]
• Children 11 and up: adult dosing, 50 mcg per day or 2,000 mcg per week [21]
• Pregnancy and breastfeeding: 50 mcg per day or 2,000 mcg per week, ideally split into two half-doses to improve absorption, plus B12 status checks through the pregnancy [21, 35]
• Kidney disease: avoid high-dose cyanocobalamin, and discuss methylcobalamin (typically 1,000–2,000 mcg per day) with your physician [23, 33, 46]

Why do people over 65 need such a high dose?

Because absorption declines with age, and the dose needed to actually normalize the biochemistry turns out to be far higher than most supplements provide. Two trials pin this down. In a dose-finding study of older adults, researchers tested 25, 100, and 1,000 mcg per day: the 25 and 100 mcg doses lowered but did not normalize methylmalonic acid in most participants, and only the 1,000 mcg dose reliably brought it into the normal range [52]. A second, larger randomized trial in older people with mild deficiency tested doses from 2.5 up to 1,000 mcg and found that 647 to 1,032 mcg per day produced 80 to 90 percent of the maximum achievable reduction in methylmalonic acid [53]. That’s the evidence behind the 1,000-mcg-a-day recommendation after 65: not a guess, but the dose the trials show is actually needed.

Why are supplement doses so much higher than the RDA?

Because intrinsic factor, the transport protein that handles efficient absorption, saturates at about 1 to 2 mcg per sitting and needs roughly 4 to 6 hours to reset [2, 33]. Infrequent large doses work only via that roughly 1 percent passive-diffusion route (absorption is about 2 percent at 500 mcg and 1.3 percent at 1,000 mcg [2, 10]), so the math demands more, less often.

Can I get enough B12 from fortified foods instead?

Yes, with attention to timing. Small fortified doses ride the intrinsic-factor system, so they must be spread across at least three servings a day, spaced 4–6 hours apart, together supplying roughly double the daily value. Greger’s rule of thumb is about 190 percent of the DV across three servings [35]. Fortified nutritional yeast is the classic vehicle, and these serving sizes meet the target for several common brands [35]:

• Lotus (Australia and New Zealand): 1 tsp, three times daily
• Marigold Engevita (UK): 2 tsp, three times daily
• Anthony’s Premium (Canada): 1 tsp, three times daily
• Hoosier Hill Farm (Canada): 1 tsp, three times daily

Always check the label, because fortified is the operative word, and store B12 foods and supplements away from light, which degrades cobalamin [33].

Cyanocobalamin or methylcobalamin: which is better?

Use cyanocobalamin. It is the most stable form, the form behind virtually the entire treatment-trial literature, and the NIH notes no evidence that absorption differs between forms [2]. The “coenzyme” forms marketed as premium, methylcobalamin and adenosylcobalamin, are not likely to be superior [11]. Methylcobalamin is notably light-sensitive and less stable, and in one small report even 2,000 mcg per day of it failed to normalize status in one of three vegans tested [35]. Your body converts stable cyanocobalamin into both active coenzyme forms exactly where it needs them [2, 11]. The methyl-B12 premium is, as VeganHealth puts it, largely supplement-industry marketing [33], a delicious irony given which side of this debate cries “follow the money.” (The kidney-disease and possibly heavy-smoker exceptions above are the caveats [23, 33, 35].)

Should I chew or swallow my B12 tablet?

Chew or dissolve B12-only supplements. Your salivary glands secrete haptocorrin, a B12-binding protein that escorts the vitamin safely through digestion [2], which is the plausible mechanism behind Crane’s striking result that chewers’ levels soared while whole-tablet swallowers’ didn’t, until they too began chewing [14, 35].

Is a multivitamin enough for B12?

Not as your sole source. Classic work by Kondo and colleagues showed that other vitamins and minerals in a combined pill can degrade B12 into inactive, potentially transport-blocking analogues [12], and the tragic infant case above involved a mother relying on exactly such a product [13]. A standalone chewable, sublingual, or liquid B12 is the reliable tool [33, 35]. (If you do take a multivitamin for other reasons, swallow it whole rather than chewing [33].)

Does B12 toothpaste work?

Partially. Two double-blind, randomized, placebo-controlled trials found a B12-fortified toothpaste, used two minutes twice daily, improved B12 status on average in vegans [16] and in older adults [17], which is a lovely stand-alone demonstration that adding B12 and nothing else fixes the markers. But it didn’t work in every individual, so it belongs as an adjunct verified by testing rather than a primary strategy [35].

How do I get my B12 tested?

The workup is simple: serum B12, confirmed with methylmalonic acid (the most specific marker) when serum values are borderline, with homocysteine as supporting evidence [1, 2, 33]. Skip spirulina and chlorella in the weeks before testing, since their pseudo-B12 analogues can inflate a serum reading while doing nothing for you [27, 33]. If you’ve gone years unsupplemented, test now: caught early, deficiency is fully and cheaply reversible; caught late, the neurological component may not be [1, 2].

Can you take too much B12?

The Institute of Medicine set no upper intake limit for B12 because of its low toxicity [2, 3]. The notable minor side effect is acne-like breakouts in a small number of people at high doses, which resolve on dose reduction [33]. One last preemption, since it’s the objection most likely to be thrown at this post: headlines linking high blood B12 to cancer come from observational studies of serum levels, which disease itself can elevate, since some illnesses raise the body’s B12-binding proteins independent of intake [33]. The controlled evidence points the other way: a meta-analysis of 18 randomized trials in nearly 75,000 people found B-vitamin supplementation, including B12 at doses up to 2,000 mcg per day, had little or no effect on cancer incidence, cancer deaths, or total mortality [2]. (The VITAL cohort finding on high-dose supplements in male smokers is examined in full in the audit section above [47].)

Don’t some expert bodies advise against vegan diets for kids and pregnancy?

They do, and it’s worth engaging honestly, because a critic will raise it and the resolution actually strengthens the case rather than weakening it. In 2016 the U.S. Academy of Nutrition and Dietetics judged well-planned vegan diets appropriate for all life stages, while the German Nutrition Society (DGE) declined to recommend a vegan diet for pregnant women, nursing mothers, infants, children, or adolescents, specifically because it requires supplementation. A French-speaking pediatric group took a similar cautious line, stressing that on a vegan diet “vitamin B12 and vitamin D supplementation is always necessary” [54].

Two things resolve the apparent conflict. First, the disagreement isn’t about the biology. Every one of these bodies agrees on the underlying facts: an unsupplemented plant-based diet is inadvisable in pregnancy and childhood, and B12 must be supplied. The dispute is narrower and more political, about whether to affirmatively recommend a diet that depends on a supplement to vulnerable groups, not about whether the diet can be made adequate. That is exactly what “well-planned” is doing in the Academy’s phrasing. As one review reconciling the positions put it, everyone agrees the supplement is non-negotiable; they differ only on default messaging [55].

Second, the most-cited dissent has since softened. The DGE revised its position in 2024: rather than advising against vegan diets for these groups, it now states it “cannot make a clear recommendation either in favour of or against a vegan diet” for them, citing limited data and the need for well-informed planning and supplementation. So the strongest version of the “even the experts say no” objection is already out of date. The trajectory of expert opinion is toward the position this article takes: a well-planned, supplemented vegan diet can work at every life stage, and the supplement is the plan working, not the plan failing.

The ethics coda: supplementing is the vegan position

Is taking a B12 supplement un-vegan?

The interview asks, sincerely, why vegans would trust industries that torture animals, and then aims that question at a bacterially fermented vitamin. Point it in the right direction and it answers itself.

Supplemental B12 involves zero animals: cultured microbes in a tank [29, 30]. The B12 in modern animal products, by contrast, is substantially the output of the feed-supplementation pipeline documented above, the same industrial vitamin, routed through a confined animal’s body before reaching the plate [29, 30]. Between a tablet and a slaughterhouse, the tablet is not the ethical compromise. It is the ethical escape.

And there’s the movement-level accounting. Every influential vegan who persuades followers to shun B12 is, on a years-long fuse, manufacturing the future ex-vegans and cautionary headlines that anti-vegan voices feast on, doing the opposition’s work for free, with our own community’s trust as the fuel. The animals are best served by vegans who are visibly, verifiably thriving at year five, year twenty, year forty-five. The full weight of the evidence says a well-planned vegan diet delivers exactly that [5], and well-planned means, at minimum, the one nutrient our modern, sanitized food system no longer provides by accident. Fifty micrograms a day is not a crack in the vegan argument. It’s the keystone.

The book behind the interview

Is “The Enigma of Vitamin B12” credible?

The interview didn’t invent its B12 claims; it was drawing on a specific source, which the host praised on air: a self-published 2025 book, The Enigma of Vitamin B12: Nature’s Only Mistake?, by Anthea V. Hayes. It’s worth addressing by name, because its arguments are the ones circulating, and every one of them has already been answered above.

The book’s marketing frames B12 as “Nature’s only mistake” and asks, with heavy implication, why a healthy diet should require a pill and why the list of deficiency symptoms keeps growing. Laid against the evidence, each beat collapses. B12 isn’t nature’s mistake; it’s a bacterial nutrient that every animal (cow, pig, human) must obtain from microbes, which is why the animals people eat are themselves supplemented or cobalt-fed, per the roughly 55 percent feed-market fact. It wasn’t “invented” or “patented into existence” by a drug company; it was isolated in 1948 to cure a reliably fatal disease, and it’s manufactured today by bacterial fermentation, not a chemistry set. The “why so many symptoms?” rhetorical flourish mistakes a well-mapped two-enzyme biochemistry (DNA synthesis and myelin maintenance, hence the blood and nerve effects) for something suspiciously sprawling. And the implied answer, that you can safely skip it, is precisely the claim that the EPIC-Oxford data (52 percent of vegans deficient), the raw-vegan data (76 percent low), and the infant case reports refute in the only currency that matters: measured deficiency in real people. A book can ask “why is this the one nutrient not in plants?” as though it were a smoking gun. The answer is microbiology, and it has been known for decades. Provocative questions are not evidence, and a compelling narrative is not a controlled trial.

A brief word about the rest of the interview

The B12 claims did not travel alone. The same conversation asserted that contagion does not exist, that childhood vaccines are destroying nervous systems, and that fluoride, chemtrails, and calcified pineal glands explain public gullibility. A full treatment is beyond this post’s scope, but the pattern deserves one sentence: each of those claims is contradicted by evidence at every level of the same hierarchy used here, and when defending an opinion about a vitamin requires denying the germ theory of disease, the opinion is carrying more load than any opinion should. Extraordinary claims stacked on extraordinary claims do not buttress each other. They compound.

The bottom line

Vitamin B12 is a bacterial nutrient every human needs, discovered through a twenty-year scientific effort to cure a fatal disease, produced today by fermentation without a single animal, and required in amounts that cost a few dollars a year. Its deficiency has a specific laboratory fingerprint, reverses with B12 alone in placebo-controlled trials, silently spends down a multi-year bodily reserve before symptoms appear, and can end in irreversible nerve damage, which is precisely why “I feel fine and refuse testing” is the least informative sentence in this debate. Deficiency is common among omnivores, universal policy already tells everyone over 50 to supplement, and the majority customer of the world’s B12 supply is the animal-agriculture industry itself, the fact that turns “B12 is an anti-vegan conspiracy” into an argument that refutes its own premise. Take your cyanocobalamin, 50 micrograms chewed daily or 2,000 weekly, or eat properly spaced fortified foods, get tested if you’ve lapsed, make sure the pregnant and breastfeeding vegans in your life are covered, and let this be the place the “great supplement scam” narrative is finally laid to rest.

Further resources

Dr. Greger’s B12 series and summary at NutritionFacts.org, including the updated recommendations infographic, plus the videos “How Long Does It Take to Become Vitamin B12 Deficient and What Are the Symptoms?”, “The Best Way to Test for Vitamin B12 Deficiency”, “The Best Source of Vitamin B12: Supplements, Shots, or Fortified Foods?”, “How Much Vitamin B12 Do We Need Each Day?”, “Are Large Weekly Doses of Vitamin B12 Safe?”, “The Optimal Vitamin B12 Dosage for Adults”, “The Optimal Vitamin B12 Dosage for Kids, Pregnancy, and Seniors”, and “The Best Type of Vitamin B12: Cyanocobalamin or Methylcobalamin?”

VeganHealth.org’s B12 hub, including “Should I Get My B12 Status Tested?”, “Vitamin B12 in Plant Foods”, “Intestinal Bacteria as a Vitamin B12 Source”, “Can a Natural Diet Require Supplements?”, and “What Every Vegan Should Know About Vitamin B12”.

Editorial note: the quantitative claims in this article, including study populations, percentages, doses, dates, discovery history, and industrial-production figures, were checked against the cited primary sources, abstracts, and regulatory documents in August 2026, prior to publication.

References

1. Green R, Allen LH, Bjorke-Monsen AL, et al. Vitamin B12 deficiency. Nat Rev Dis Primers. 2017;3:17040. PubMed
2. National Institutes of Health, Office of Dietary Supplements. Vitamin B12: Fact Sheet for Health Professionals. ods.od.nih.gov
3. Institute of Medicine, Food and Nutrition Board. Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National Academies Press; 1998. NCBI
4. Ströhle A, Richter M, González-Gross M, et al. The Revised D-A-CH Reference Values for the Intake of Vitamin B12: Prevention of Deficiency and Beyond. Mol Nutr Food Res. 2019;63(6):e1801178. PubMed
5. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian Dietary Patterns for Adults: A Position Paper of the Academy of Nutrition and Dietetics. J Acad Nutr Diet. 2025. Current position paper, approved 2025, in effect through 2032; supersedes Melina V, Craig W, Levin S. Position of the Academy of Nutrition and Dietetics: Vegetarian Diets. J Acad Nutr Diet. 2016;116(12):1970-1980.
6. Wang H, Li L, Qin LL, Song Y, Vidal-Alaball J, Liu TH. Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database Syst Rev. 2018;3:CD004655. PubMed
7. Kuzminski AM, Del Giacco EJ, Allen RH, Stabler SP, Lindenbaum J. Effective treatment of cobalamin deficiency with oral cobalamin. Blood. 1998;92(4):1191-1198. PubMed
8. Lederle FA. Oral cobalamin for pernicious anemia. Medicine’s best kept secret? JAMA. 1991;265(1):94-95. PubMed
9. Alpers DH. Another indication for the use of oral vitamin B-12 supplementation. Am J Clin Nutr. 2018;108(1):1-3. PubMed
10. Carmel R. How I treat cobalamin (vitamin B12) deficiency. Blood. 2008;112(6):2214-2221. PubMed
11. Obeid R, Fedosov SN, Nexo E. Cobalamin coenzyme forms are not likely to be superior to cyano- and hydroxyl-cobalamin in prevention or treatment of cobalamin deficiency. Mol Nutr Food Res. 2015;59(7):1364-1372. PubMed
12. Kondo H, Binder MJ, Kolhouse JF, Smythe WR, Podell ER, Allen RH. Presence and formation of cobalamin analogues in multivitamin-mineral pills. J Clin Invest. 1982;70(4):889-898. PubMed
13. Guez S, Chiarelli G, Menni F, Salera S, Principi N, Esposito S. Severe vitamin B12 deficiency in an exclusively breastfed 5-month-old Italian infant born to a mother receiving multivitamin supplementation during pregnancy. BMC Pediatr. 2012;12:85. PubMed
14. Crane MG, Sample C, Patchett S, Register UD. Vitamin B12 Studies in Total Vegetarians (Vegans). J Nutr Med. 1994;4(4):419-430. Taylor and Francis
15. Donaldson MS. Metabolic vitamin B12 status on a mostly raw vegan diet with follow-up using tablets, nutritional yeast, or probiotic supplements. Ann Nutr Metab. 2000;44(5-6):229-234. PubMed
16. Siebert AK, Obeid R, Weder S, et al. Vitamin B-12-fortified toothpaste improves vitamin status in vegans: a 12-wk randomized placebo-controlled study. Am J Clin Nutr. 2017;105(3):618-625. PubMed
17. Zant A, Awwad HM, Geisel J, Keller M, Obeid R. Vitamin B12-fortified toothpaste improves vitamin status in elderly people: a randomized, double-blind, placebo-controlled study. Aging Clin Exp Res. 2019;31(12):1817-1825. PubMed
18. Briani C, Dalla Torre C, Citton V, et al. Cobalamin deficiency: clinical picture and radiological findings. Nutrients. 2013;5(11):4521-4539. PubMed
19. Rizzo G, Laganà AS, Rapisarda AM, et al. Vitamin B12 among Vegetarians: Status, Assessment and Supplementation. Nutrients. 2016;8(12):767. PubMed
20. Pawlak R, Lester SE, Babatunde T. The prevalence of cobalamin deficiency among vegetarians assessed by serum vitamin B12: a review of literature. Eur J Clin Nutr. 2014;68(5):541-548. PubMed
21. Baroni L, Goggi S, Battaglino R, et al. Vegan Nutrition for Mothers and Children: Practical Tools for Healthcare Providers. Nutrients. 2018;11(1):5. PubMed
22. Hankey GJ. B vitamins for stroke prevention. Stroke Vasc Neurol. 2018;3(2):51-58. PubMed
23. Spence JD. Metabolic vitamin B12 deficiency: a missed opportunity to prevent dementia and stroke. Nutr Res. 2016;36(2):109-116. PubMed
24. Tucker KL, Rich S, Rosenberg I, et al. Plasma vitamin B-12 concentrations relate to intake source in the Framingham Offspring study. Am J Clin Nutr. 2000;71(2):514-522. PubMed
25. Zhou Y, Wang A, Yeung LF, Qi YP, Pfeiffer CM, Crider KS. Folate and vitamin B12 usual intake and biomarker status by intake source in United States adults aged 19 y and older: NHANES 2007-2018. Am J Clin Nutr. 2023;118(1):241-254. PubMed
26. Bailey RL, Carmel R, Green R, et al. Monitoring of vitamin B-12 nutritional status in the United States by using plasma methylmalonic acid and serum vitamin B-12. Am J Clin Nutr. 2011;94(2):552-561. PubMed
27. Watanabe F, Yabuta Y, Bito T, Teng F. Vitamin B12-containing plant food sources for vegetarians. Nutrients. 2014;6(5):1861-1873. PubMed
28. Watanabe F, Bito T. Vitamin B12 sources and microbial interaction. Exp Biol Med (Maywood). 2018;243(2):148-158. PubMed
29. Martens JH, Barg H, Warren MJ, Jahn D. Microbial production of vitamin B12. Appl Microbiol Biotechnol. 2002;58(3):275-285. PubMed
30. EFSA Panel on Additives and Products or Substances used in Animal Feed (FEEDAP). Safety and efficacy of a feed additive consisting of vitamin B12 (cyanocobalamin) produced by fermentation with Ensifer adhaerens for all animal species. EFSA Journal. 2024;22(4):e8752. B12 is authorised as a nutritional feed additive for all animal species; authorised by Commission Implementing Regulation (EU) 2024/1325.
31. Rickes EL, Brink NG, Koniuszy FR, Wood TR, Folkers K. Crystalline vitamin B12. Science. 1948;107(2781):396-397; Smith EL. Purification of anti-pernicious anaemia factors from liver. Nature. 1948;161:638-639; West R. Activity of vitamin B12 in Addisonian pernicious anemia. Science. 1948;107:398.
32. The Nobel Prize in Physiology or Medicine 1934 (Whipple, Minot, Murphy) and The Nobel Prize in Chemistry 1964 (Dorothy Crowfoot Hodgkin). nobelprize.org
33. Norris J. Vitamin B12. VeganHealth.org (Vegan Outreach), including the subpages “Vitamin B12 in Plant Foods,” “Intestinal Bacteria as a Vitamin B12 Source,” “Should I Get My B12 Status Tested?,” “Coenzyme Supplements: Methylcobalamin and Adenosylcobalamin,” “Are Multivitamins an Adequate Source of B12 for Vegans?,” “B12 Status of Raw Foodist Vegans,” “Individual Cases of B12 Deficiency in Vegans,” “Immerman: The Exception,” and “Can a Natural Diet Require Supplements?”
34. What Every Vegan Should Know About Vitamin B12: An Open Letter from Health Professionals and Vegan Organizations (2001). VeganHealth.org
35. Greger M. How Not to Diet. Flatiron Books; 2019; and the NutritionFacts.org vitamin B12 video series and Updated Vitamin B12 Recommendations Infographic.
36. Mariotti F, ed. Vegetarian and Plant-Based Diets in Health and Disease Prevention. Academic Press; 2017.
37. Gilsing AMJ, Crowe FL, Lloyd-Wright Z, Sanders TAB, Appleby PN, Allen NE, Key TJ. Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: results from a cross-sectional analysis of the EPIC-Oxford cohort study. Eur J Clin Nutr. 2010;64(9):933-939. PubMed
38. Rauma AL, Törrönen R, Hänninen O, Mykkänen H. Vitamin B-12 status of long-term adherents of a strict uncooked vegan diet (“living food diet”) is compromised. J Nutr. 1995;125(10):2511-2515. PubMed
39. Allen LH, Miller JW, de Groot L, Rosenberg IH, Smith AD, Refsum H, Raiten DJ. Biomarkers of Nutrition for Development (BOND): Vitamin B-12 Review. J Nutr. 2018;148(suppl 4):1995S-2027S. PubMed
40. Hazra A, Kraft P, Selhub J, Giovannucci EL, Thomas G, Hoover RN, Chanock SJ, Hunter DJ. Common variants of FUT2 are associated with plasma vitamin B12 levels. Nat Genet. 2008;40(10):1160-1162. PubMed. Replications followed in Chinese and Indian populations, with additional loci in TCN1, CUBN, and MUT identified in subsequent GWAS.
41. CobVar: a comprehensive resource of vitamin B12-associated genomic variants. 2025. PMC
42. Albert MJ, Mathan VI, Baker SJ. Vitamin B12 synthesis by human small intestinal bacteria. Nature. 1980;283(5749):781-782. PubMed
43. Degnan PH, Taga ME, Goodman AL. Vitamin B12 as a modulator of gut microbial ecology. Cell Metab. 2014;20(5):769-778. PMC
44. Herbert V. Vitamin B-12: plant sources, requirements, and assay. Am J Clin Nutr. 1988;48(3 Suppl):852-858.
45. Antony AC. Vegetarianism and vitamin B-12 (cobalamin) deficiency. Am J Clin Nutr. 2003;78(1):3-6.
46. House AA, Eliasziw M, Cattran DC, Churchill DN, Oliver MJ, Fine A, Dresser GK, Spence JD. Effect of B-vitamin therapy on progression of diabetic nephropathy: a randomized controlled trial (DIVINe). JAMA. 2010;303(16):1603-1609. PubMed
47. Brasky TM, White E, Chen CL. Long-term, supplemental, one-carbon metabolism-related vitamin B use in relation to lung cancer risk in the Vitamins and Lifestyle (VITAL) cohort. J Clin Oncol. 2017.
48. FUT2 secretor variant p.Trp154Ter influences serum vitamin B12 concentration via holo-haptocorrin, but not holo-transcobalamin, and is associated with haptocorrin glycosylation. Hum Mol Genet. 2017;26(24):4975. Oxford Academic
49. Kocaoglu C, Akin F, Caksen H, Böke SB, Arslan S, Aygün S. Cerebral atrophy in a vitamin B12-deficient infant of a vegetarian mother. J Health Popul Nutr. 2014;32(2):367-371. PubMed
50. Lövblad K, Ramelli G, Remonda L, Nirkko AC, Ozdoba C, Schroth G. Retardation of myelination due to dietary vitamin B12 deficiency: cranial MRI findings. Pediatr Radiol. 1997;27(2):155-158. PubMed
51. Sukumar N, Rafnsson SB, Kandala NB, Bhopal R, Yajnik CS, Saravanan P. Prevalence of vitamin B-12 insufficiency during pregnancy and its effect on offspring birth weight: a systematic review and meta-analysis. Am J Clin Nutr. 2016;103(5):1232-1251. PubMed
52. Rajan S, Wallace JI, Brodkin KI, Beresford SA, Allen RH, Stabler SP. Response of elevated methylmalonic acid to three dose levels of oral cobalamin in older adults. J Am Geriatr Soc. 2002;50(11):1789-1795. PubMed
53. Eussen SJ, de Groot LC, Clarke R, et al. Oral cyanocobalamin supplementation in older people with vitamin B12 deficiency: a dose-finding trial. Arch Intern Med. 2005;165(10):1167-1172. PubMed
54. Lemale J, Mas E, Jung C, Bellaiche M, Tounian P. Vegan diet in children and adolescents. Recommendations from the French-speaking Pediatric Hepatology, Gastroenterology and Nutrition Group (GFHGNP). Arch Pediatr. 2019;26(7):442-450. PubMed
55. Pawlak R. To vegan or not to vegan when pregnant, lactating or feeding young children. Eur J Clin Nutr. 2017;71(11):1259-1262. PubMed. German Nutrition Society (DGE) position: 2016 (advising against vegan diets for these groups), updated 2024 to “no clear recommendation either in favour of or against.”

Read more...

There is a moment, somewhere around the first scan or the first kick, when the question stops being abstract. What am I going to feed this person? Not for a meal. For a childhood. For the body and the brain and the habits they will carry for the rest of their life.

If you are reading this, you have already had the thought that a plant-based start might be one of the best gifts you can give them. I want to tell you plainly: the evidence agrees that it can be, and I want to give you everything you need to do it well, at every stage, without fear and without guesswork. I am not going to oversell it. A vegan diet for a child is not magic, and it is not effortless. It asks for a little more attention than the default. But so does every good thing you will do as a parent, and this one is unusually well rewarded.

Is a vegan diet the best diet for raising children?

I will answer the question directly, because you are probably asking it: yes, I believe it is, and the word doing the work in that sentence is honest, not loud.

Here is the reasoning, laid out so you can decide for yourself. On lowering the risk of the chronic diseases that shorten modern lives, a well-planned plant-based diet sits at or very near the top of everything nutrition science has to offer. I will be straight that it is not the only contender on health alone; the Mediterranean pattern has a deep evidence base too, and it shares most of a vegan diet’s plant-forward DNA. But a family’s diet is never only a health intervention. It is also the daily, three-times-a-day expression of what that family stands for, and on that axis nothing else comes close, because a vegan diet is the only common way of eating that asks no animal to suffer or die for the plate.

Put the two together, among the strongest choices we have for a long, healthy life and the only one that lines up with the compassion most children already feel toward animals, and “the best choice we can make for them” stops being a slogan and becomes a defensible conclusion. What I will not do is tell you that studies have proven vegan children grow up healthier than every other child. The childhood research is still young and mostly observational, and pretending otherwise would cost this guide the credibility that makes the rest of it worth reading. Best-reasoned, not yet fully proven, and, as you will see, that is still firmer ground than the default diet it replaces, which has never been “proven” for children either and produces the very outcomes we are trying to spare them.

What does a well-planned vegan diet actually mean?

Almost every scare story about vegan children is really a story about a deficient child, usually one on an extreme, unsupplemented, or improvised diet. That is not what we are talking about, and the distinction matters more than any single study. A well-planned vegan diet, the kind this guide is about, means three things: reliable vitamin B12 every single day, from a supplement or fortified foods; real variety across whole plant foods, legumes, whole grains, vegetables, fruits, nuts, seeds, and soy foods; and enough calories for a growing body.

When I say whole plant foods, I mean real food across the full range, legumes, greens, grains, fruits, vegetables, mushrooms, nuts, seeds, seaweeds, herbs and spices, eaten with variety. I do not mean the shrink-wrapped novelty aisle with a vegan label on the front. Some of those products are fine as occasional stepping stones, but they are not the diet, and a company selling a child food that quietly undermines their health is not doing veganism any favors; humans are animals too. The plate that is best for your child turns out to be the same plate that is least processed and least harmful to the animals and the planet they will inherit, which is not a coincidence. I follow that thread all the way down, from a forkful to a cell to the whole living world, in The Anti-Inflammatory Journey.

Is vitamin B12 really non-negotiable on a vegan diet?

Yes, and it is the one rule you never bend. B12 is not really an “animal” nutrient at all; it is made by bacteria, which is why the farmed animals most people eat are themselves routinely given B12 or cobalt in their feed. Vegans simply take the direct route and skip the animal. If you read one companion piece to this guide, make it my deep dive on vitamin B12. The broader picture of how plants cover every nutrient a child needs is in my nutrition overview.

Does a vegan child need more planning than a child who eats meat?

Every diet for a child requires planning. Every pregnant woman, vegan or not, is handed folic acid, vitamin D, an iron target, an iodine concern, and a list of fish to avoid for mercury. The standard Western diet is never called “well-planned omnivorous eating,” but it needs planning too, and by most measures it is planned badly, which is why chronic disease is where it is. The vegan list differs from the omnivore’s by essentially one item moving from optional to mandatory: B12. Get B12 right, add variety and enough calories, and you have cleared the bar that the fear-mongering depends on you tripping over.

Does a vegan diet reduce a child’s risk of chronic disease later in life?

The diseases that shorten and worsen most lives in wealthy countries, heart disease, type 2 diabetes, high blood pressure, several cancers, obesity, are the ones a well-planned plant-based diet is most consistently linked to preventing. This is not a fringe claim. It is the conclusion the largest dietetic body in the United States has reached and re-reached across editions spanning more than three decades. Its current position paper, published in 2025, formally scopes its statement to adults, and you will see anti-vegan commentators spin that as the Academy quietly withdrawing support for feeding children this way. It is not. The 2025 update was a deliberate choice to narrow the scope of that particular paper, not a reversal, and the Academy’s earlier and still widely cited 2016 position stated plainly that well-planned vegetarian and vegan diets are appropriate for all stages of life, including pregnancy, infancy, childhood, and adolescence. The support is echoed by dietetic associations in the UK, Canada, and the Nordic countries. Even the bodies that were historically most cautious have moved. The German Nutrition Society, long cited by critics as the authority that refused to endorse vegan diets for children and pregnancy, updated its position in 2024 and now says it “cannot make a clear recommendation either in favour of or against” a vegan diet for these groups, given limited data, provided B12 is supplemented and the diet is well planned with qualified guidance. That is a softening from an outright non-recommendation, and it moved in the direction the evidence has been moving. If you want the why underneath all of this, the actual cellular reason whole plant foods calm inflammation while animal foods stoke it, told as a journey a non-scientist can follow, that is the whole point of The Anti-Inflammatory Journey.

The numbers behind it are genuinely striking. In the Adventist Health Study-2, of more than 73,000 people, vegetarians had about a 12 percent lower risk of dying from any cause over the study period than non-vegetarians (hazard ratio 0.88), with the strongest benefits in men. In that same population, vegans had the lowest average body weight of any diet group and the lowest rates of type 2 diabetes, a clean stepwise gradient, leanest in vegans and heaviest in omnivores. In EPIC-Oxford, of 48,000 people, vegetarians and vegans had about 22 percent lower risk of ischaemic heart disease. And when Stanford researchers ran the cleanest test possible, a randomized trial in identical twins where one twin went vegan and the other stayed omnivorous, the vegan twins’ LDL (“bad”) cholesterol dropped by about 14 mg/dL in eight weeks, along with fasting insulin and body weight. Same genes, different plates, measurably different hearts. I unpack what that remarkable Adventist population eats in The Adventist Health Studies, and how lower calorie density drives the weight difference in the best diet to lose weight.

Why does what a child eats matter for heart disease sixty years later?

Because the runway is longer than people think. Autopsy studies of children and young adults, the Bogalusa Heart Study and PDAY, found the earliest streaks of artery disease already forming in childhood and adolescence, tracking with cholesterol levels. The habits and the cholesterol of childhood are not waiting for adulthood to start mattering; they are already writing the first lines of the story. For why cholesterol is the thread running through all of this, see Cholesterol, From the Ground Up.

What are the honest limits of the chronic-disease evidence?

Most of this evidence is observational. People who go vegan also tend to smoke less, drink less, and exercise more, and no study fully removes that “healthy-user” advantage. It is not all one-directional either: EPIC-Oxford also found vegetarians had a slightly higher stroke risk. Naming these things yourself is what separates a persuasive parent from a preachy one. The direction and consistency of the benefit is real; the certainty is “strong and reasonable,” not “proven beyond all doubt.” That is still a better foundation than the diet it is being compared against, which has never cleared that bar either. If a critic reaches for the “but longevity hotspots eat meat” line, I have fact-checked it in Blue Zones: debunked or not.

Is a vegan diet safe for children?

The evidence in children is younger and thinner than the adult evidence. There are no long-term randomized trials raising children vegan; there never will be, because you cannot randomize someone’s childhood. What we have is a handful of cross-sectional studies and cohorts, and in 2025 the European paediatric gastroenterology society (ESPGHAN) reviewed them systematically and concluded that a nutritionally complete vegan diet supporting normal growth in children is “potentially achievable but cannot be confirmed” with current evidence, and requires supplementation and monitoring. That is a finding of insufficient evidence and a need for care, not a finding of harm. Applied consistently, that same standard would forbid nearly all specific childhood dietary advice, since almost none of it rests on lifetime trials.

North American paediatricians land in a similar place. The American Academy of Pediatrics, in its Pediatric Nutrition handbook, treats appropriately planned vegetarian and vegan diets as able to meet the nutritional needs of infants, children, and adolescents, while stressing that vegan children in particular need reliable supplementation (B12 above all, along with attention to vitamin D, iron, and calcium) and professional monitoring. (I am describing the handbook’s general position rather than a single edition’s exact wording, because the phrasing has shifted between editions and earlier editions were more cautious; treat this as the AAP’s broad stance, not a verbatim endorsement, and check the current edition if you want to quote it precisely.)

Do the risks come from the plants, or from skipped supplements?

Within that limited evidence, one pattern stands out clearly, and it is the most important thing in this entire guide: outcomes track supplementation, not the label “vegan.” Sort the studies by how well-supplemented the children were. In the German VeChi Youth study, where about 88 percent of vegan children took B12, researchers found no specific nutritional risks in vegan children and adolescents versus omnivores. In a German adult cohort where 97 percent supplemented, vegan blood markers largely matched omnivores’. In the Polish Desmond study, where only about 44 percent of vegan children took B12, the vegan children had three times the rate of B12 deficiency and more iron-deficiency anaemia. The deficiencies are not coming from plants. They are coming from skipped supplements. This is enormously good news, because it means the risks are almost entirely under your control. A supplement is not a sign the diet is broken; it is the diet working exactly as designed, the same way the Institute of Medicine already tells everyone over fifty to get their B12 from supplements or fortified foods, because aging guts stop absorbing it well.

Do vegan children grow up shorter?

This is the objection you will hear most, and the honest answer is reassuring, but it takes two studies to tell properly, so let me give you both rather than the flattering one.

The best meta-analytic answer is no clear height difference. When Sutter and Bender pooled the vegan-child growth studies in 2021, vegans came out shorter only because one included study had a large age imbalance, its vegan children were mostly 10 to 14 years old while the comparison group was mostly 15 to 19. Remove that single mismatched study and the height gap disappears. The lacto-ovo evidence points the same way: Sabate’s studies found vegetarian children reached normal, even slightly taller, attained height, in some studies up to about an inch, roughly 2.5 centimeters, taller. And the frightening old stories of stunting come from macrobiotic diets, the severely restricted, unsupplemented regimens in the Dutch cohort (Dagnelie), not from well-planned modern veganism.

Where does that leave the Polish Desmond study, which did find vegan children about 3 cm shorter with lower bone mineral content? It is real, and I will not bury it. The reconciliation is that height in vegan children tracks how well the diet is planned and fed. Desmond’s own authors attributed the difference to lower intakes of protein, calcium, and calories, and that was a cohort where most children were not even supplementing B12. Where protein, calories, and calcium are adequate, growth matches omnivores, which is exactly what the German VeChi children showed. So the takeaway is not “vegan children are short.” It is that a well-planned, well-fed vegan child grows normally, and the way you guarantee that is on the plate.

Could slower, steadier growth actually be a good thing?

There is a real mechanism worth putting on the table here, carefully. Dairy and animal protein raise IGF-1, a growth-signaling hormone, and higher IGF-1 in adulthood is associated with higher risk of several cancers. A slower, steadier childhood growth trajectory, rather than the accelerated growth that dairy in particular tends to drive, is therefore biologically plausible as a feature rather than a flaw. I want to be precise about the strength of this: it is a well-grounded hypothesis, not a proven outcome. No study has followed vegan-raised children all the way to cancer outcomes to confirm it. So treat a modest, well-nourished difference in growth pace as reassuring rather than alarming, without claiming veganism prevents cancer by lowering IGF-1. The honest version is the more interesting one anyway: faster is not automatically better when it comes to a growing body.

What about bone health in vegan children?

This is the one area I treat as genuinely needing care, and I will say so plainly. In adults, vegans in the EPIC-Oxford cohort had more fractures, about 20 extra total fractures and 15 extra hip fractures per 1,000 people over 10 years. That cohort was recruited in the 1990s, when vegan calcium intakes were low and fortified plant milks were rare, and in a separate Adventist cohort, supplementing calcium and vitamin D roughly halved the excess hip-fracture risk. In children, Desmond found lower bone mineral content even after accounting for body size, and up to about 90 percent of the bone mass you will ever have is laid down by around age 18 to 20. So the paediatric bone plan is not optional: calcium from calcium-set tofu, fortified plant milks, and low-oxalate greens (kale, bok choy, broccoli); vitamin D; adequate protein; a healthy, not-too-low body weight; and weight-bearing play, running, jumping, and climbing, which has the strongest evidence of anything for building young bones. I will be straight that this plan is mechanistically sound and guideline-consistent rather than proven in a trial to close the gap, which is all the more reason to take it seriously.

Is a vegan diet safe in the first year of life?

The first year carries the highest stakes, because a baby’s needs are enormous relative to a tiny stomach, and because an unsupplemented, B12-deficient diet in a breastfed infant can cause real and sometimes permanent harm. It is worth being precise about what that harm looks like, because precision is more reassuring than vagueness here. Caught early, it is largely reversible: with treatment, even the brain shrinkage that severe deficiency can cause tends to resolve on later scans. Caught late, it is a different story. When treatment is delayed, the imaging may recover while the child does not fully, and long term deficits in cognition, movement, and language can persist. The lesson is not that plants are dangerous. It is that this one nutrient is time sensitive, which is exactly why you never skip it and never delay if something seems wrong. This is exactly why the early years are the one stage where “wing it” is not an option, and where a paediatric dietitian is worth their weight in gold. Handled properly, it is completely manageable. Handled carelessly, it is the one place veganism can genuinely go wrong. I would rather you hear that from someone who supports you than from a critic. And raw diets, specifically, are not appropriate for infants or young children, whose small stomachs cannot pull adequate energy from that much low-density bulk, which I cover in Raw and Cooked.

Are vegetarian and vegan children smarter?

You will run into this claim, and it is worth handling honestly, because the honest version is more disarming than the boast. Vegetarian children have tested higher on IQ, in one classic study about 16 points higher than average, with mental ages running ahead of their chronological ages. But before anyone reaches for “plants make kids smarter,” look at what the researchers themselves concluded. The association is almost certainly confounded: more highly educated parents are more likely to raise vegetarian children, and they pass down both their books and their beans. And a large UK cohort that followed roughly 8,000 children settled the direction of causation the other way: it measured IQ at age 10 and found that the smarter children were more likely to become vegetarian as adults. In other words, brighter kids chose the diet; the diet did not manufacture the brightness. Presented that way it is a genuinely interesting finding. Presented as “veganism raises IQ” it would be the mirror image of the carnivore overreach this whole guide refuses. The researchers behind the UK cohort could not resist quoting Benjamin Franklin, who credited a vegetable diet with “greater clearness of head and quicker comprehension.” A nice line to keep in your back pocket, offered in the same spirit: with a smile, not as a boast.

How do you feed a vegan child at every age?

Here is the practical heart of it, stage by stage: what is happening, the foods that carry the load, and what to supplement.

What should a vegan mother eat during pregnancy?

Your body is doing something extraordinary, and a well-planned vegan diet is fully capable of supporting it. A plant-rich pregnancy may even be protective against pre-eclampsia and excess pregnancy weight gain, and the birth weights of babies born to well-planned vegan mothers do not differ meaningfully from anyone else’s. Eat generously and do not under-fuel, because pregnancy raises your calorie needs and plant foods are less calorie-dense, so lean on energy-rich additions: nut and seed butters, tahini, avocado, olive oil, dried fruit, full-fat soy. Build meals around legumes and whole grains for protein; pair iron-rich foods (lentils, tofu, fortified cereals, pumpkin seeds) with vitamin C (peppers, citrus, tomatoes) to boost absorption, and keep tea and coffee away from meals. Get calcium from calcium-set tofu, fortified plant milks, and low-oxalate greens. Supplement folic acid, B12 daily, vitamin D, iodine, iron as advised, and algal DHA. Consider choline too; it is concentrated in eggs, so vegans should plan for it, and honestly only about 8 percent of pregnant women meet the recommended intake on any diet. It is worth knowing, when someone frames vegan pregnancy as uniquely risky, that B12 insufficiency is common in pregnancy regardless of diet: one study found roughly 21 percent of pregnant women insufficient in the first trimester, rising to about 29 percent by the third, in a population that was not even vegetarian. The difference is that a well-planned vegan mother is already supplementing and watching for it, rather than assuming her diet has it covered. And do not fear healthy plant fats, or the plant “toxins” myth.

What about breastfeeding on a vegan diet?

The breast milk of a well-nourished vegan mother is nutritionally equivalent to anyone else’s; that is the finding of the systematic reviews. It reflects your diet, a little higher in healthy unsaturated fats, which is fine, and as long as you keep taking B12, iodine, and algal DHA, your milk carries them to your baby. One lovely symmetry to remember: every breastfed baby, vegan or not, is recommended a daily vitamin D supplement (about 8.5 to 10 micrograms in the UK, 400 IU in the US), because breast milk is naturally low in it regardless of the mother’s diet. So the one supplement your newborn needs is not a “vegan” supplement at all; it is the same one the whole world’s babies get. If you cannot breastfeed or choose not to, note that a truly vegan infant formula is not yet on general sale in the UK or US as of 2026, and soy formula is dairy-free but usually not strictly vegan, so this is worth discussing with your health visitor or paediatrician, without shame either way.

How do you start solids on a vegan diet (6 to 12 months)?

Around six months, alongside continued breast milk or formula, the adventure of solids begins. The nutrients that need the most attention now are iron, zinc, and B12, because these are the hardest to get without animal foods at this age. Good first foods that pull their weight: iron-fortified baby cereal; well-cooked, mashed lentils and beans; smooth tofu; finely ground seeds. Because tiny stomachs fill up fast on bulky plant food, this is the stage to deliberately add calories and ease off the fibre a little: thin nut and seed butters into porridge, mash in avocado, add a little oil to vegetable purees, and do not be afraid of some refined grains. Whole nuts are a choking hazard until age 5, so use smooth butters instead. Fortified soya can go into cooking from six months but should not be a main drink until twelve months, and rice milk is not suitable under five because of arsenic. Continue vitamin D, add B12 for the baby as they take more solids and less milk, and offer iron-rich foods from six months, with a supplement only if advised. This is the stage where a dietitian’s eye is most valuable.

Which foods are choking hazards, and how do you serve them safely?

This deserves its own moment, because plant-based weaning leans on exactly the foods that need care: nuts, chickpeas, firm raw vegetables. Whole nuts are a choking hazard until around age 4 or 5, but nut butters thinned into food, and nuts or seeds ground to a powder and stirred through porridge, are both safe and are also how you introduce these allergens early, which is now what allergy guidance recommends. The general rule with round foods is to make them not round: quarter grapes and cherry tomatoes lengthwise rather than slicing them into coins. Skip popcorn, whole nuts, hard raw carrot, and thick sticky spoonfuls of nut butter until your child chews well. And it matters as much how a child eats as what: sitting upright, unhurried, supervised, and not distracted or laughing. One reassurance worth having, because it frightens every new parent: noisy gagging is normal and protective, and it is how babies learn to manage food. Real choking is silent. Learning infant first aid is the single most useful hour a new parent can spend, on any diet.

How do you feed a vegan toddler (1 to 5 years)?

Boundless energy, tiny stomachs, and strong opinions. Keep meals small, frequent, and energy-dense. Every meal wants a protein anchor (beans, lentils, tofu, chickpeas, soy yoghurt), a calcium source (calcium-set tofu, fortified plant milk; check the label, because many organic brands are not fortified), and something calorie-rich. An unsweetened, calcium-fortified soy or pea drink can become a main milk drink from twelve months. This is also the golden age of food exposure, which brings us to the single most useful feeding advice in this whole guide, further down.

How do you feed a vegan child at school age (6 to 12 years)?

Now the world gets bigger: packed lunches, other kids’ parties, opinions about what is “normal.” Nutritionally this is one of the easier stretches: a varied diet of legumes, whole grains, fortified foods, fruit, vegetables, nuts, and seeds does the job, with calcium and iron as the priorities to keep an eye on. Encouragingly, vegan children in the research consistently eat more fibre, fruit, vegetables, and legumes and less saturated fat than their omnivore peers, and population studies link vegetarian diets to a lower prevalence of obesity in children as well as adults; they are often eating closer to what every child is supposed to eat. This is the stage to make packed lunches a point of pride, not apology, and to start handing over real cooking skills.

How do you feed a vegan teenager (12 to 18 years)?

The growth spurt raises the stakes on iron (especially for girls once periods start), calcium, and protein and calories for the building body, and B12 as always. Because plant protein is slightly less digestible, aim a little higher, roughly 15 to 20 percent more than the standard recommendation, which happens naturally with generous beans, tofu, tempeh, seitan, soy milk, and nuts. The reassuring headline: the VeChi Youth study, which was mostly adolescents, found no specific nutritional risks in vegan teens versus omnivores, with high supplement use doing the heavy lifting. A vegan teenager can be a thriving athlete; they just need enough total food and the same protein-timing basics as anyone. Two things belong here honestly. Calcium intake tends to run low in vegan teens, and, tellingly, in omnivore teens too, so make fortified foods and calcium-set tofu daily habits. And adolescence is when eating disorders can emerge; more on the honest version of that concern below.

How do you prepare a vegan young adult to eat well on their own?

Your goal for eighteen years has been to make yourself unnecessary here. A young adult who can batch-cook a pot of chili, knows a five-item pantry (beans, grains, frozen vegetables, a fortified plant milk, and a jar of peanut butter will keep anyone alive and well), and can read a menu for the vegan option will carry this effortlessly. Make sure they leave home still taking their B12. That is the one habit that has to travel with them.

What supplements does a vegan child need, by age?

Keep this simple. B12 is the star; the rest are supporting cast you adjust with your health professional.

• Pregnancy. Always: B12 daily, vitamin D, folic acid, iodine, algal DHA. Consider, as advised: iron and choline.
• Breastfeeding mother. Always: B12 daily, vitamin D, iodine, algal DHA. Consider: iron.
• Breastfed baby. Always: vitamin D (all babies, every diet). Consider: B12 as milk intake decreases, and iron from 6 months.
• Toddler and preschooler. Always: B12, vitamin D. Consider: iodine, algal DHA, and iron or zinc if intake is low.
• Child 6 to 12. Always: B12, vitamin D. Consider: iodine, iron, and calcium if intake is low.
• Teenager 12 to 18. Always: B12, vitamin D. Consider: iron (especially girls), calcium, iodine, and algal DHA.

For monitoring, a simple periodic check, B12 (ideally active B12 or MMA and homocysteine), ferritin with a blood count, and vitamin D, is the sensible, empowering move, done the same way you would get any child’s health checked. No fixed schedule is universally mandated; annual is a common, practical rhythm. Think of it as confirmation you are doing it right, not as a search for what is wrong.

On B12 specifically, it helps to have real numbers rather than just “take a supplement,” so here is the schedule most vegan paediatric dietitians work from, published by Baroni and colleagues. Every figure below is cyanocobalamin, which matters, and I will explain why in a moment.

• Infants and toddlers, from about 6 months to 3 years. 5 micrograms once a day, or 1 microgram twice a day.
• Children 4 to 10. 25 micrograms a day.
• Age 11 and up. 50 micrograms a day, or 2,000 micrograms once a week.
• Pregnancy and breastfeeding. 50 micrograms a day, ideally split into two half doses, which improves how much you actually absorb.

Two details worth knowing. The split-dose trick is not fussiness: B12 is absorbed by an efficient route that saturates after a microgram or two and takes hours to reset, plus a passive route that only captures about 1 percent of whatever is left. Two small doses several hours apart use the efficient route twice, which is why 1 microgram twice a day does the same job as 5 micrograms once. And during pregnancy and breastfeeding, dose daily rather than weekly. Weekly dosing is fine for older children and adults, but these are the stages to keep it steady.

Which form of B12 should you buy?

Buy cyanocobalamin. This is the near-unanimous recommendation of the dietitians who specialise in vegan nutrition, including Jack Norris at veganhealth.org, the Vegan Society, and the vegetarian nutrition group of the Academy of Nutrition and Dietetics, and it is the form the dosing schedule above assumes. It is the most stable, the cheapest, and by far the best studied. Methylcobalamin is marketed as more “natural” or “active,” but it degrades faster with light, heat, and other vitamins in the same tablet, and it has a much thinner evidence base. The important practical point: the doses above are not transferable microgram for microgram to methylcobalamin. If cyanocobalamin is genuinely all you cannot find, use a single-nutrient methylcobalamin product, store it somewhere cool and dark, and use a substantially higher dose, at least 25 micrograms a day for a maintenance supplement. There are a few specific situations where a doctor may prefer another form, including kidney disease and certain rare genetic conditions, which is a conversation to have with them rather than a reason to switch by default.

Is the cyanide in cyanocobalamin dangerous?

No, and this one is worth addressing directly because it circulates constantly and frightens parents. Cyanocobalamin does contain a cyanide molecule, and the amount is trivial. A 1,000 microgram tablet, far larger than anything in the schedule above, delivers roughly 20 micrograms of cyanide. The average person already takes in somewhere around 113 to 164 micrograms of cyanide a day from ordinary food and air. So even a large B12 tablet is a rounding error against your everyday background exposure, and it sits far below any safety threshold. Cyanocobalamin is the form used in infant formula, which is about as strong a vote of confidence as regulatory science offers. The real danger to a vegan child is not the supplement. It is skipping it.

One last note on amounts. There is no established upper limit for B12 and no meaningful toxicity, because it is water soluble and the body clears the excess. That does not mean more is better; there is no benefit to routine mega-dosing a child. Aim for the schedule, and treat larger corrective doses as something to use with your health professional if a gap is actually documented. Treat all of this as a starting point to confirm with them, not a prescription.

Beyond B12: iron, iodine, and the rest

B12 is the one rule that never bends, but a few other nutrients reward a little understanding. None of these require anxiety. They require knowing one or two practical things each.

How do you make sure a vegan child gets enough iron and zinc?

Plant iron and zinc are real and plentiful; they are just absorbed less efficiently, because compounds in whole grains and legumes bind to them. The official response to this is simply to aim higher: dietary guidelines suggest vegetarians need roughly 80 percent more iron than the standard target, and somewhat more zinc. That sounds daunting and is not, because the same handful of habits fixes both. Pair iron-rich foods, which means lentils, beans, tofu, fortified cereals, pumpkin seeds, and dark greens, with a vitamin C source at the same meal, such as peppers, tomatoes, citrus, or strawberries, which can multiply absorption several times over. Keep tea and coffee away from meals, since they work strongly against it. And lean on foods that have been soaked, sprouted, fermented, or leavened, because all of those processes break down the binding compounds: yeast-risen wholegrain bread, tempeh, miso, and soaked or well-cooked beans all give up their minerals more readily than the raw equivalents. Cooking acidic foods in a cast-iron pan genuinely adds iron. Iron deficiency is the most common nutritional deficiency in young children on every diet, so this is not a vegan problem, it is a childhood one, and it deserves the same attention any parent would give it.

How much iodine does a child need, and is seaweed safe?

Iodine is the nutrient where vegan families most often get it wrong in both directions, which is why it is worth a moment. Children need roughly 90 micrograms a day in the toddler years, rising to about 150 micrograms for adults and around 200 micrograms in pregnancy and breastfeeding. The reliable sources are iodised salt and a supplement.

The trap is seaweed. Kelp and other brown seaweeds, including kombu, arame, and hijiki, can carry staggering amounts of iodine, sometimes thousands of micrograms in a single gram, and a single serving can blow past the safe upper limit many times over. Too much iodine harms the thyroid just as deficiency does, and infants and unborn babies are the most vulnerable of all. So do not use kelp as your child’s iodine source, do not give kelp supplements to children, and avoid hijiki entirely, which also carries inorganic arsenic. A small amount of nori is fine; a sensible ceiling for a baby is no more than about half a sheet, and not every day. Get the iodine from iodised salt or a measured supplement instead, where you know exactly what the dose is. Thyroid specialists recommend a supplement providing 150 micrograms of iodine, as potassium iodide rather than kelp, before and during pregnancy and while breastfeeding.

How much vitamin D does a child need?

This one is refreshingly simple, and it is not a vegan question at all. Paediatric guidance is 400 IU, which is 10 micrograms, every day for all infants under one year, starting in the first days of life, because breast milk is naturally low in vitamin D whatever the mother eats. From age one onward the target is 600 IU, or 15 micrograms, a day. Vitamin D2 is always vegan, and vegan D3 made from lichen is now widely available; either works.

Does a vegan child need a DHA supplement?

Probably worth it, and I will be honest that the evidence for benefit is not settled. Land plants contain no EPA or DHA, and the body converts the plant omega-3 in flax, chia, and walnuts into DHA at a low rate, so vegan children do run lower. What has not been demonstrated is that this causes any measurable harm; no deficits in vision or development have been shown in vegan children. Given that an algae-based supplement is cheap, safe, and removes the question entirely, most vegan dietitians suggest one. Useful targets are about 100 milligrams of DHA a day for babies and toddlers, and around 250 milligrams of combined EPA and DHA a day for older children, with pregnancy and breastfeeding a little higher. Serve the plant sources daily regardless, since they do plenty of other good. More on this in my deep dive on DHA and omega-3s.

Are vegan children getting enough protein?

Almost always yes on total protein, which is the question everyone asks. The more interesting question is quality. Plant protein is digested slightly less efficiently, so the practical target for a vegan child runs roughly 15 to 20 percent above the standard recommendation, and the amino acid most likely to run short is lysine. That sounds technical and resolves into one sentence: serve legumes every day. Beans, lentils, chickpeas, peas, peanuts, and especially soy foods like tofu, tempeh, and soy milk are the lysine-rich foods that close the gap, and soy protein in particular is comparable in quality to animal protein. A child eating legumes or soy foods at most meals, plus grains, nuts, and seeds, is not going to have a protein problem. A child living on fruit, plant milk, and crackers might, which is another way of saying that variety and adequate calories do most of the work here.

Do children naturally care about animals?

Here is what the nutrition tables cannot capture, and what may matter most. There is a quiet piece of developmental science that most people never hear. When researchers give children moral dilemmas about humans and animals, children extend far more moral concern to animals than adults do. In one study, most young children valued a dog’s life close to a human’s, and a majority would save ten pigs over one human, a trade almost no adult would make. The tidy conviction that human lives simply matter overwhelmingly more than animal lives turns out to appear late, and to be learned. Other work backs this up: compared to adults, children are less likely to see farm animals as “food,” think farm animals deserve better treatment, and judge eating animals as less acceptable. And when young children who had independently chosen vegetarianism in meat-eating families were asked why, they pointed, universally, to not wanting to hurt animals.

Read that again, because it reframes everything. Raising a child vegan is not imposing an adult ideology on a blank slate. For many children, it is removing the conflict between what they are fed and what they already feel. The natural tenderness a small child shows a dog, a lamb, a chicken, that is not something to grow out of. It is something a plant-based home lets them keep.

One caution that will keep you honest and kind: those same vegetarian children did not condemn others for eating meat. Tolerance, not superiority, is the natural childhood posture, and it is the one to nurture. “This is what our family chooses because we love animals” travels a great deal further, and truer, than “other people are bad.” Avoid “children are naturally vegan” as a slogan; the honest, stronger claim is that this diet lets a child live in line with compassion they already have.

How do you talk to kids about where food comes from?

The research on passing on values is clear: children internalize values that are explained and reasoned, warmly and age-appropriately, far more deeply than values simply enforced. So talk about the why, and let it grow up with them. With ages 2 to 5, keep it simple and warm: “We do not eat animals because we love them and want to keep them safe.” With ages 5 to 8, they can grasp more, and this is when children come to understand that death is permanent, so you can be gently honest about where food comes from without gory detail; let their own questions lead. With ages 9 to 12, bring in the bigger picture, health, animals, the planet, and let them start articulating it in their own words. A kid who can say why is a kid who can hold their ground in the lunchroom. With teens, step back: offer information, respect their autonomy, and let it be genuinely theirs. Values that are chosen stick; values that are policed rebel. And do not hide where meat comes from or distort it; the research suggests children who learn the truth plainly tend, on their own, to want to protect animals. Honesty is on your side.

Does a vegan diet affect a child’s mental health?

The scary headlines linking veganism to depression are riddled with the chicken-and-egg problem, because people already struggling sometimes change their diet, and the actual trials point the other way: plant-based eating has, if anything, shown mood benefits. There is no good evidence that a well-planned vegan diet harms a child’s mental health, and a home built on compassion, family meals, and shared values has a lot going for it. The one thing to watch honestly is disordered eating in adolescence, where vegetarianism or veganism is sometimes used as a socially acceptable cover for restriction. For most vegan teens the diet is a values choice, not a disorder; it is more often a marker or a method in an already-vulnerable young person than a cause. But watch for restriction that keeps escalating beyond veganism, preoccupation with weight or shape, secrecy around food, or withdrawal from family meals, and take it seriously if you see it.

How do you get kids to eat and enjoy their vegetables?

This may be worth more than any supplement.

• Reward the tasting, not the plate. A sticker or genuine praise for bravely trying a new food works, because it rewards the behavior you want. Rewarding how much they eat (“clean your plate for dessert”) backfires: it teaches kids to like the food less and to override their own fullness.
• Never use food as the prize. Making dessert the reward for eating broccoli makes broccoli the loser and dessert the hero. Keep them equal citizens of the plate.
• Repeat, repeat, repeat. A toddler may need to meet a new vegetable 8 to 15 times before accepting it. Early rejection is normal, not a verdict.
• Eat what you want them to eat. Your own plate is the single strongest predictor of theirs. You are the menu.
• Cook and grow together. Kids who help cook, and kids who grow even a pot of tomatoes, eat more vegetables. Hand them a job at the counter.
• Protect family meals. Regular shared meals are linked to better nutrition and real protection against disordered eating in teens. Guard them.
• Divide the responsibility. You decide what, when, and where; let them decide whether and how much. Pressure creates fussier eaters, not better ones.

How do you raise a confident vegan child around people who are not vegan?

Let us be real about the social side. Vegan kids, boys especially, can meet teasing, and there is measurable cultural bias against vegans. You cannot erase that, but you can inoculate against it, and it starts with how the choice lives in your home: as a warm family value, not a judgment of others.

What books and outings help a child understand why?

Picture books do the values work for you, gently: That’s Why We Don’t Eat Animals, Vegan Is Love, and V Is for Vegan (Ruby Roth); Not a Nugget (Stephanie Dreyer); Dave Loves Chickens (Carlos Patino); Lena of Vegitopia (Sybil Severin); and the true story The True Adventures of Esther the Wonder Pig. Timeless classics like Charlotte’s Web and Babe (from The Sheep-Pig) do the same emotional work without ever using the word “vegan.” For you and older teens, Nourish by Reshma Shah MD and Brenda Davis RD is the standout family nutrition guide, and Feeding Your Vegan Child by NHS dietitian Sandra Hood is excellent and practical. And nothing makes the why real like a visit to a farm sanctuary, meeting a rescued pig who runs to greet visitors or a hen who likes to be held; the Global Federation of Animal Sanctuaries keeps a directory of accredited sanctuaries, and many welcome family visits.

How do you handle parties, sleepovers, and eating out?

The dietitian consensus is simple and it works: send a dish to share, so there is always something great your child can eat, and often something the other kids love too; give the host a friendly heads-up in advance; keep a few “safe” vegan snacks in your bag; and arm your child with an easy, non-preachy line: “I am vegan, so I do not eat that, but I brought my own, thank you.” Teach them to scan a menu for the naturally-vegan option (the bean burrito, the veggie pasta, the falafel) so eating out feels like ordering, not negotiating.

When and how should you talk to other parents and schools?

Before a playdate or party, a warm, low-drama message does wonders: “Quick heads-up, my child is vegan. I am happy to send food so it is no trouble at all, and they are very used to it; just wanted you to know so no one worries.” You are not asking anyone to change their kitchen; you are making it easy. That framing turns potential friction into gratitude nearly every time. For school, know your ground. In England, School Food Standards encourage but do not yet require a daily vegan option, so a polite written request for accommodation usually gets results, and ethical veganism is a recognized protected belief under the Equality Act; a well-planned packed lunch is always a solid fallback, so check your school’s packed-lunch policy. In the US, there is no vegan mandate in the National School Lunch Program, but plant-based options are increasingly available, and a confident packed lunch, a wrap, hummus and vegetables, fruit, a flapjack, and a treat, often outshines the cafeteria anyway. And celebrate the reasoning, not just the compliance: when your eight-year-old explains to a friend why they do not eat animals, or your teenager cooks the family dinner, that is the thing to notice out loud. You are not rewarding obedience; you are reinforcing a value they are making their own.

What do critics get right about vegan diets for kids?

Let me hand you the strongest objections, so nothing catches you off guard. The childhood evidence base is genuinely limited: it is young and mostly observational, and paediatric bodies like ESPGHAN rightly ask for better studies. But “not yet proven” is not “shown to be harmful,” and by that same strict standard almost no specific childhood diet is proven. Bone health is the real weak spot, which is why the calcium, protein, calories, and weight-bearing-exercise plan is not optional. Infancy demands supervision, so bring in a paediatric dietitian and never, ever skip B12. And the tragic cases that make headlines almost always involve unsupplemented, raw, or fruitarian regimens, or plant “milks” replacing formula, not well-planned, supplemented veganism; those diets share a word with yours and nothing else. None of these is a reason not to do it. Every one is a reason to do it well, which is exactly what this guide is for. For the meat-and-cancer headline that resurfaces every year or two, here is why one subgroup number did not overturn the link.

The bottom line

You are being asked to extend a little more care than the default parent extends: to read a label, to remember a supplement, to send a dish to a party, to have a few honest conversations. In return: a child whose diet is linked to less of the disease that shortens the lives around them, who eats more of what every child is supposed to eat, whose habits are being set toward a long healthy life, and who gets to grow up without ever having to unlearn their own kindness toward animals. There is one more thing on that list, easy to miss because it is the largest. The same plate that calms the inflammation in a child’s body calms it in the world that child will inherit: fewer forests felled, less water fouled, less of the planet spent to put dinner on the table. A body and a world, quieted by the same meal. I trace that whole parallel, from a single cell to the whole living planet, in The Anti-Inflammatory Journey. That is not a consolation prize for a difficult diet. That is one of the best gifts a parent can give, and the parents who have done it will tell you, again and again, that it was worth every ounce of the effort. Because of course it was. It is their child. It is yours, too.

Meet the families raising their children vegan

The science is the why and the guide above is the how, but nothing lands like watching real children flourish on plants. Over the past year I sat down with a new family or lifelong vegan every week, and the series keeps growing with interviews I recorded in earlier years and continue to add to now. Inside it you will meet a paediatrician, a dietitian, physicians, and an obstetrician talking through the evidence; multigenerational vegan families; parents raising children vegan from conception; and, most persuasive of all, people who were raised vegan and are now thriving, healthy adults who are glad to be vegan, with parents who call it the best gift they ever gave their children.

Prefer to open it on YouTube? Here is the full Raising Children Vegan playlist.

These interviews, and everything else in this guide, now have a home of their own. I have built RaisingChildrenVegan.com as the dedicated companion to this work: a stage-by-stage parenting guide you can search, the full why-vegan evidence, curated books and family resources, kid-friendly vegan places, and a growing community of vegan parents, lifelong vegans, and the doctors and dietitians who support them. If this post spoke to you, that is where to go next.

References

1. Melina V, Craig W, Levin S. Position of the Academy of Nutrition and Dietetics: Vegetarian Diets. J Acad Nutr Diet. 2016;116(12):1970–1980. PubMed
2. Raj S, Guest NS, Landry MJ, Mangels AR, Pawlak R, Rozga M. Vegetarian Dietary Patterns for Adults: A Position Paper of the Academy of Nutrition and Dietetics. J Acad Nutr Diet. 2025;125(6):831–846.e2. PubMed
3. Verduci E, Koglmeier J, Haiden N, et al. Vegan diet and nutritional status in infants, children and adolescents: A position paper based on a systematic search by the ESPGHAN Nutrition Committee. J Pediatr Gastroenterol Nutr. 2025;81(5):1318–1345. PubMed
4. Orlich MJ, Singh PN, Sabate J, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Intern Med. 2013;173(13):1230–1238. PubMed
5. Tonstad S, Butler T, Yan R, Fraser GE. Type of vegetarian diet, body weight, and prevalence of type 2 diabetes. Diabetes Care. 2009;32(5):791–796. PubMed
6. Tong TYN, Appleby PN, Bradbury KE, et al. Risks of ischaemic heart disease and stroke in meat eaters, fish eaters, and vegetarians over 18 years of follow-up: results from the prospective EPIC-Oxford study. BMJ. 2019;366:l4897. PubMed
7. Landry MJ, Ward CP, Cunanan KM, et al. Cardiometabolic effects of omnivorous vs vegan diets in identical twins: a randomized clinical trial. JAMA Netw Open. 2023;6(11):e2344457. PubMed
8. Berenson GS, Srinivasan SR, Bao W, et al. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults (Bogalusa Heart Study). N Engl J Med. 1998;338(23):1650–1656. PubMed
9. PDAY Research Group. Relationship of atherosclerosis in young men to serum lipoprotein cholesterol concentrations and smoking. JAMA. 1990;264(23):3018–3024. PubMed
10. Alexy U, Fischer M, Weder S, et al. Nutrient intake and status of German children and adolescents consuming vegetarian, vegan or omnivore diets: results of the VeChi Youth Study. Nutrients. 2021;13(5):1707. PubMed
11. Weder S, Hoffmann M, Becker K, Alexy U, Keller M. Energy, macronutrient intake, and anthropometrics of vegetarian, vegan, and omnivorous children (1–3 years) in Germany (VeChi Diet Study). Nutrients. 2019;11(4):832. PubMed
12. Weder S, Keller M, Fischer M, Becker K, Alexy U. Intake of micronutrients and fatty acids of vegetarian, vegan, and omnivorous children (1–3 years) in Germany (VeChi Diet Study). Eur J Nutr. 2022;61(3):1507–1520. PubMed
13. Weikert C, Trefflich I, Menzel J, et al. Vitamin and mineral status in a vegan diet. Dtsch Arztebl Int. 2020;117:575–582. PubMed
14. Desmond MA, Sobiecki JG, Jaworski M, et al. Growth, body composition, and cardiovascular and nutritional risk of 5- to 10-y-old children consuming vegetarian, vegan, or omnivore diets. Am J Clin Nutr. 2021;113(6):1565–1577. PubMed
15. Desmond MA, Sobiecki J, Fewtrell M, Wells JCK. Plant-based diets for children as a means of improving adult cardiometabolic health. Nutr Rev. 2018;76(4):260–273. PubMed
16. Sutter DO, Bender N. Nutrient status and growth in vegan children. Nutr Res. 2021;91:13–25. PubMed
17. Koller A, Rohrmann S, Wakolbinger M, et al. Health aspects of vegan diets among children and adolescents: a systematic review and meta-analyses. Crit Rev Food Sci Nutr. 2024;64(33):13247–13258. PubMed
18. Sabate J, Lindsted KD, Harris RD, Sanchez A. Attained height of lacto-ovo vegetarian children and adolescents. Eur J Clin Nutr. 1991;45(1):51–58. PubMed
19. Dagnelie PC, van Dusseldorp M, van Staveren WA, Hautvast JG. Effects of macrobiotic diets on linear growth in infants and children until 10 years of age. Eur J Clin Nutr. 1994;48 Suppl 1:S103–S112. PubMed
20. Sabate J, Wien M. Vegetarian diets and childhood obesity prevention. Am J Clin Nutr. 2010;91(5):1525S–1529S. PubMed
21. Dwyer JT, Miller LG, Arduino NL, et al. Mental age and I.Q. of predominantly vegetarian children. J Am Diet Assoc. 1980;76(2):142–147. PubMed
22. Gale CR, Deary IJ, Schoon I, Batty GD. IQ in childhood and vegetarianism in adulthood: 1970 British cohort study. BMJ. 2007;334(7587):245. PubMed
23. Weaver CM, Gordon CM, Janz KF, et al. The National Osteoporosis Foundation’s position statement on peak bone mass development and lifestyle factors. Osteoporos Int. 2016;27(4):1281–1386. PubMed
24. Tong TYN, Appleby PN, Armstrong MEG, et al. Vegetarian and vegan diets and risks of total and site-specific fractures: results from the prospective EPIC-Oxford study. BMC Med. 2020;18:353. PubMed
25. Thorpe DL, Beeson WL, Knutsen R, Fraser GE, Knutsen SF. Dietary patterns and hip fracture in the Adventist Health Study 2: combined vitamin D and calcium supplementation mitigate increased hip fracture risk among vegans. Am J Clin Nutr. 2021;114(2):488–495. PubMed
26. Baroni L, Goggi S, Battaglino R, et al. Vegan nutrition for mothers and children: practical tools for healthcare providers. Nutrients. 2019;11(1):5. PubMed
27. Baroni L, Goggi S, Battino M. Planning well-balanced vegetarian diets in infants, children, and adolescents: the VegPlate Junior. J Acad Nutr Diet. 2019;119(7):1067–1074. PubMed
28. Karcz K, Krolak-Olejnik B. Vegan or vegetarian diet and breast milk composition: a systematic review. Crit Rev Food Sci Nutr. 2021;61(7):1081–1098. PubMed
29. Ureta-Velasco N, Keller K, Escuder-Vieco D, et al. Human milk composition and nutritional status of omnivore human milk donors compared with vegetarian/vegan lactating mothers. Nutrients. 2023;15(8):1855. PubMed
30. Koeder C. Toward supplementation guidelines for vegan complementary feeding. Food Sci Nutr. 2024;12(12):10962–10971. PubMed
31. Wallace TC, Fulgoni VL. Usual choline intakes are associated with egg and protein food consumption in the United States. Nutrients. 2017;9(8):839. PubMed
32. Gupta PM, Perrine CG, Mei Z, Scanlon KS. Iron, anemia, and iron deficiency anemia among young children in the United States. Nutrients. 2016;8(6):330. PubMed
33. Weyand AC, Chaitoff A, Freed GL, Sholzberg M, Choi SW, McGann PT. Prevalence of iron deficiency and iron-deficiency anemia in US females aged 12–21 years, 2003–2020. JAMA. 2023;329(24):2191–2193. PubMed
34. Wilks M, Caviola L, Kahane G, Bloom P. Children prioritize humans over animals less than adults do. Psychol Sci. 2021;32(1):27–38. PubMed
35. McGuire L, Palmer SB, Faber NS. The development of speciesism: age-related differences in the moral view of animals. Soc Psychol Personal Sci. 2023;14(2):228–237. PubMed
36. Hussar KM, Harris PL. Children who choose not to eat meat: a study of early moral decision-making. Soc Dev. 2010;19(3):627–641. PubMed
37. MacInnis CC, Hodson G. It ain’t easy eating greens: evidence of bias toward vegetarians and vegans from both source and target. Group Process Intergroup Relat. 2017;20(6):721–744. PubMed
38. Grusec JE, Goodnow JJ. Impact of parental discipline methods on the child’s internalization of values. Dev Psychol. 1994;30(1):4–19. PubMed
39. Speece MW, Brent SB. Children’s understanding of death: a review of three components of a death concept. Child Dev. 1984;55(5):1671–1686. PubMed
40. Hahn ER, Gillogly M, Bradford BE. Children are unsuspecting meat eaters: an opportunity to address climate change. J Environ Psychol. 2021;78:101705. PubMed
41. Robinson-O’Brien R, Perry CL, Wall MM, Story M, Neumark-Sztainer D. Adolescent and young adult vegetarianism: better dietary intake and weight outcomes but increased risk of disordered eating behaviors. J Am Diet Assoc. 2009;109(4):648–655. PubMed
42. Sergentanis TN, Chelmi ME, Liampas A, et al. Vegetarian diets and eating disorders in adolescents and young adults: a systematic review. Children (Basel). 2020;8(1):12. PubMed
43. Cooke LJ, Chambers LC, Anez EV, et al. Eating for pleasure or profit: the effect of incentives on children’s enjoyment of vegetables. Psychol Sci. 2011;22(2):190–196. PubMed
44. Birch LL, Marlin DW, Rotter J. Eating as the “means” activity in a contingency: effects on young children’s food preference. Child Dev. 1984;55(2):431–439. PubMed
45. Wardle J, Cooke LJ, Gibson EL, et al. Increasing children’s acceptance of vegetables: a randomized trial of parent-led exposure. Appetite. 2003;40(2):155–162. PubMed
46. van der Horst K, Ferrage A, Rytz A. Involving children in meal preparation: effects on food intake. Appetite. 2014;79:18–24. PubMed
47. Hammons AJ, Fiese BH. Is frequency of shared family meals related to the nutritional health of children and adolescents? Pediatrics. 2011;127(6):e1565–e1574. PubMed
48. Satter E. The feeding relationship: problems and interventions. J Pediatr. 1990;117(2 Pt 2):S181–S189. PubMed
49. Beezhold BL, Johnston CS. Restriction of meat, fish, and poultry in omnivores improves mood: a pilot randomized controlled trial. Nutr J. 2012;11:9. PubMed
50. World Health Organization. Guideline for complementary feeding of infants and young children 6–23 months of age. Geneva: WHO; 2023. WHO
51. Sabate J, Wien M. Vegetarian diets and childhood obesity prevention. Am J Clin Nutr. 2010;91(5):1525S–1529S. PubMed
52. Wirnitzer KC, Drenowatz C, Cocca A, et al. Health behaviors of Austrian secondary level pupils at a glance: first results of the From Science 2 School study focusing on sports linked to mixed, vegetarian, and vegan diets. Int J Environ Res Public Health. 2021;18(23):12782. PubMed
53. Sukumar N, Rafnsson SB, Kandala NB, et al. Prevalence of vitamin B-12 insufficiency during pregnancy and its effect on offspring birth weight: a systematic review and meta-analysis. Am J Clin Nutr. 2016;103(5):1232–1251. PubMed
54. Kocaoglu C, Akin F, Caksen H, et al. Cerebral atrophy in a vitamin B12-deficient infant of a vegetarian mother. J Health Popul Nutr. 2014;32(2):367–371. PubMed
55. Green R, Allen LH, Bjorke-Monsen AL, et al. Vitamin B12 deficiency. Nat Rev Dis Primers. 2017;3:17040. PubMed
56. Baroni L, Goggi S, Battaglino R, et al. Vegan nutrition for mothers and children: practical tools for healthcare providers. Nutrients. 2019;11(1):5. (Table 2 is the source of the age-banded B12 dosing schedule used here.) PubMed
57. Agnoli C, Baroni L, Bertini I, et al. Position paper on vegetarian diets from the working group of the Italian Society of Human Nutrition. Nutr Metab Cardiovasc Dis. 2017;27(12):1037–1052. PubMed
58. Deutsche Gesellschaft fur Ernahrung (German Nutrition Society). Update of the DGE position on vegan diet. Ernahrungs Umschau. 2024;71(7):60–84. DGE
59. Institute of Medicine. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. Washington, DC: National Academies Press; 2001. (Source of the higher vegetarian iron and zinc targets.) PubMed
60. Committee on Nutrition, American Academy of Pediatrics. Prevention of choking among children. Pediatrics. 2010;125(3):601–607. PubMed
61. American Academy of Pediatrics. Nutritional aspects of vegetarian diets. In: Pediatric Nutrition (handbook). Elk Grove Village, IL: American Academy of Pediatrics. (Position described from the handbook’s vegetarian and vegan chapter; exact wording varies by edition and the current edition should be consulted directly, so this citation is given without a pinpoint edition or page.)

A note on the evidence: this post is built on primary research and the positions of major health bodies, and it aims to be honest about the strength of that evidence in both directions. Where findings are strong, it says so; where the evidence is younger, thinner, or mixed, it says that too. It is for general information and education, not personalized medical advice. For guidance specific to your family, consult a physician, registered dietitian, or nutritionist, ideally one who practices lifestyle medicine, since that field is evidence-based and works with your values rather than against them. Please bring any questions, concerns, or the need for a tailored plan to them, especially during pregnancy and your child’s first two years, when the stakes are highest and good guidance matters most.

Read more...

You’ve probably seen the claim by now: the Blue Zones are fake. It’s all bad paperwork and pension fraud. Dead people cashing checks. Longevity hotspots that dissolve the moment anyone audits the records.

That argument comes from one researcher. His name is Saul Newman. He’s a demographer, he won an Ig Nobel Prize for this work, and he has a book out about it. When someone online tells you the Blue Zones were debunked, Newman is almost always who they’re citing.

So I went and checked it, against the actual registries, the actual studies, and the actual actuarial methodology. And what I found is worth understanding, because it’s not a simple story of “he’s wrong” or “he’s right.” It’s more specific, and more damning, than that.

There are five Blue Zones. To conclude that a given one is fake, Newman needs a specific piece of evidence against it. Every one of those load-bearing pieces falls apart against the primary sources. Meanwhile, the things he gets right turn out to be about something else entirely: American supercentenarian records that lack birth certificates, the Greek pension database, UN estimates for countries with no functioning registration, the supplement and longevity-clinic industry. Real problems, all of them. Just not the four validated zones his headline is about.

And there’s a pattern in how the errors work. The same move, over and over: Newman looks at records nobody ever checked, then draws conclusions about records that were checked, cleaned, and corrected. Dirty data in, conclusions about clean data out, every time in the direction that supports his thesis.

Let me walk through it.

Prefer a quick, point-by-point version? The companion piece Blue Zones Debunked or Not answers the viral thread one claim at a time. This one takes on the source, Newman himself, as a single argument.

Japan: the number came from the wrong book

Did Japan find that 82% of its centenarians were dead?

This is the most-quoted claim in the entire story, and it’s the pillar holding up the case against Okinawa. Newman says Japan discovered that 230,000 of its centenarians were dead, 82% of them, alive only on paper.

Here’s what you need to know: Japan keeps two separate registries, not one.

The first is the koseki, a family register. It’s genealogical, it traces bloodlines back through generations, and it has no purge mechanism. When someone dies, if a relative never files the paperwork, the entry simply stays there. Forever. It was never a count of living people. It’s a family tree.

The second is the resident register, the jūminhyō. It’s tied to your current address, it’s what pays pensions, and it’s what produces Japan’s official count of living centenarians.

The 230,000 figure came from the family tree.

The exact number was 234,354 entries registered before 1910 that had never been crossed out. When demographers actually studied it, those entries amounted to roughly half of one percent of every birth recorded in Japan between 1872 and 1910, and the researchers concluded the effect on Japan’s life expectancy statistics was, in their words, effectively nil.

What did Japan’s real centenarian count actually show?

So what did the register that counts living people show? About 44,000 living centenarians (44,449, all confirmed alive) and a nationwide verification sweep that turned up 584 who were genuinely unaccounted for. That’s a bit over one percent unconfirmed, not eighty-two. And most of that wasn’t fraud at all: Japan’s demographers attributed it to deaths that went unrecorded in the chaos after the war, incomplete linkage between the two registries, and people who emigrated without notifying anyone. Press coverage at the time found registry entries so old they were physically impossible; one listed a man who would have been 186. Nobody was cashing a pension check for a 186-year-old. That’s a typo in a genealogy book.

Substitution number one: Newman found errors in an archive nobody maintains, then reported them as errors in Japan’s population data. Two different books. He cited the one nobody keeps up to date.

Okinawa: the theory Japan already tested

Does WWII firebombing explain Okinawa’s longevity?

Even with a number, Newman still needs a mechanism, a reason Okinawa specifically would be fake. His answer is striking: during World War II, American bombers destroyed whole neighborhoods. When a household is wiped out, who reports the deaths? Nobody. So the dead stay alive on paper. In his preprint he points to a proxy for wartime bombing intensity and reports that it predicts 79% of the variation in centenarian status across Okinawa, a correlation he credits to the demographer Michel Poulain.

It’s a genuinely clever hypothesis. It was also tested, directly, and it failed.

After the pension scandal broke, Japan sent officials out in person, across all 47 prefectures, to physically verify centenarians whose status was in doubt. Nationwide, they found 584 genuinely missing, the “about 600” you’ll sometimes see quoted.

In Okinawa, they found zero.

Consider what that means. Okinawa is the one part of Japan that was ever a Blue Zone, the exact place where, if unreported war deaths were manufacturing phantom hundred-year-olds, you would expect them stacked up. The proof-of-life sweep found none. This is arguably the single most important fact in the whole debate, and it almost never gets mentioned.

There’s a second, structural problem. Okinawa’s status was based on centenarian prevalence, a rate, a share of the population. Phantom survivors would inflate both the numerator and the denominator of that fraction, so the theory doesn’t even do what it needs to do.

Weren’t Okinawa’s records destroyed and rebuilt from memory?

Newman has a stronger version of the war argument worth taking seriously: that when destroyed records were later reconstructed, ages could have been inflated in the rebuilding. It’s his best card here. But three things blunt it. Twenty microfilmed copies of the Okinawan birth records survived off the island, so the reconstruction worked from preserved documents rather than memory. When Willcox and colleagues checked 52 centenarians against their records, 49 (94.2%) matched exactly, with no pattern of exaggeration. And most fundamentally, Japan’s life-expectancy statistics are computed from the census and the resident registry, not from the family register that holds the reconstructed entries. The reconstruction can’t inflate a number that isn’t built from it.

I’ll also concede what Newman’s defenders concede: Okinawa no longer qualifies as a Blue Zone. Its advantage faded in the postwar decades as the traditional diet gave way to imported, westernized food during the long US military presence. But notice what that does to the argument. If the longevity were a records artifact, it would have been fake all along. Instead it was real and then declined, declined exactly as the lifestyle behind it disappeared. That’s not evidence the zone was never there. It’s evidence the behaviors were what mattered.

Substitution number two: Newman theorized about what missing records would do. Japan went and looked. In Okinawa, nothing was missing.

Costa Rica: he accused the man who had already caught the problem

Was the Nicoya Blue Zone just bought and paid for?

Newman’s line on Costa Rica is that the Blue Zone was essentially bought, that a demographer was paid, and produced a Blue Zone the next day.

There’s a grain of truth in the setup, so let’s be precise about it. The Nicoya label did emerge from a 2007 expedition that National Geographic helped fund, when Buettner traveled there with the demographer Michel Poulain to verify the data. But the underlying finding wasn’t produced on that trip and wasn’t Buettner’s. Luis Rosero-Bixby had already established Costa Rica’s exceptional old-age survival independently, in peer-reviewed work, using the country’s own registries, and that body of research is what has to be wrong for Newman’s claim to land. It isn’t.

Here’s what’s actually true: Costa Rican old-age survival is one of the better-documented findings in demography. In 2008, in the journal Demography, Luis Rosero-Bixby found that mortality at age 90 in Costa Rica was at least 14% lower than the average of thirteen high-income countries. Costa Rican men who reach 90 live another 4.4 years on average, half a year longer than in any other country on Earth. And he stated plainly that these estimates do not use self-reported ages; they use birth dates drawn from Costa Rica’s official birth-registration ledgers. A 2016 paper in PNAS found American men dying at rates 18% higher than Costa Rican men.

Nobody checks for age exaggeration, right?

But here’s the part that ends this one. Rosero-Bixby caught the age exaggeration himself. He compared census-reported ages against the national ID card in a sample of 7,400 seniors and found that roughly 30% of people over 90 had overstated their age, versus about 10% of people in their sixties. He published it. That is precisely the problem Newman says the field can’t detect, and the Costa Rican researcher detected it, measured it, printed it, and then built his estimates on registry birth dates specifically to route around it.

Then he did one more thing: he published a paper showing the Nicoya longevity advantage is shrinking, down to about a quarter of its original size. A consultant delivering a paid marketing result does not publish the collapse of his own headline finding.

Substitution number three: Newman treated Costa Rican age data as uncorrected. It had already been corrected, by the very researcher he’s accusing, in print, years earlier.

Greece: the wrong database entirely

Were 72% of Greek centenarians collecting pensions from the grave?

Newman says Greece had more than 9,000 centenarians on the books, and that after the financial crisis forced an audit, at least 72% were found to be collecting pensions from the graveyard.

Those 9,000 were names on a pension payment roll, an accounting document, a list of bank accounts receiving money. It was never a demographic count, and it was never the research.

Greece’s Blue Zone is the island of Ikaria, and Ikaria was validated in 2008 and 2009, before the bailout audit, using entirely separate sources: national death statistics going back to 1995, the municipal registry, birth records complete from 1913, and in-person interviews with every single resident aged 90 and over, cross-checked against datable historical events. That work found 124 people aged 90 or older living on the island: 1.49% of the population, nearly five times the 0.33% on the Greek mainland.

Different database. Different purpose. Years apart. No overlap.

Substitution number four: Newman audited the payments system and reported the result as though it invalidated the research.

Sardinia: measuring the wrong map

Isn’t Sardinia just a poor region with bad records?

Now count with me. Okinawa, Ikaria, Nicoya: that’s three of the four scientifically validated Blue Zones, and every argument against them has collapsed.

So what does Newman have against Sardinia, the fourth?

Not a fraud case, a statistical sleight of hand. He notes that Italy’s oldest-old cluster in regions with low income, low literacy, and short average life expectancy, and that Sardinia ranked only 51st of 128 European regions for old-age life expectancy when Eurostat began keeping records. It sounds damning. It relies on a switch of subject.

The Sardinian Blue Zone was never the whole island. It’s a cluster of about a dozen mountain villages in the Ogliastra region, roughly twelve thousand people, defined by a validated rate of people reaching 90 and beyond. Whole-island averages, whole-island crime rates, whole-island literacy tell you nothing about those villages, any more than a national statistic tells you about one town. Newman is measuring Sardinia to make a claim about a zone that is a small, specific part of Sardinia. And poverty isn’t fraud regardless: being poor doesn’t make your parish records fake.

How is a Sardinian centenarian’s age actually verified?

And here’s what Sardinia’s verification actually looked like: every person aged 90 or older was cross-checked against three independent sources, government civil registers back to 1866, church archives running from the 1600s, and a complete genealogical reconstruction of every village. That process caught a fake. A woman recorded as dying at 110 turned out to be her own younger sister; the first child had died, and the parents gave the next baby the same name three years later. The researchers found her, published the case, and deleted her from the database.

A system that finds and publicly reports its own false positives is the opposite of a system being fooled.

The insurance argument runs backwards

Does bad longevity data make your insurance cost more?

This is the part engineered to make you angry, and it’s why the whole thing went viral. Newman says this bad data feeds the models that set your pension and your life insurance, and that if the models assume you’ll live longer than you actually will, you pay more, every single day.

It’s wrong on the mechanism and backwards on the direction.

The mechanism first: life insurance tables are not built from supercentenarian records. Social Security’s own published methodology says so directly. At extreme old age, the actuaries don’t use the raw data, they extrapolate, precisely because the data up there is sparse and of questionable quality. They already treat those records as unreliable; it’s in the manual. What actually sets your rate is experience data covering millions of lives, concentrated between ages 65 and 95. For scale, the international database of validated supercentenarians launched with 672 people worldwide, a number far too small to move an aggregate table.

Now the direction. If a model assumes you’ll live longer, your life insurance gets cheaper, not more expensive. Think about it for a moment: life insurance pays out when you die. If the insurer expects you to die later, it’s less likely to pay during the policy term, so it charges you less. Longer assumed lifespans do raise the cost of annuities and defined-benefit pensions (that part is real) but those are priced off retiree experience data, not off a list of 110-year-olds.

So the emotional payoff of the entire argument, this is why your bill is high, is the piece that falls apart the moment you check how insurance actually works.

Loma Linda: his best shot, and why it backfires

Is Loma Linda actually a longevity hotspot?

Now the fifth zone: Loma Linda, California. This is Newman’s one genuine hit, and it’s worth giving him fully.

He points out that the founder only added Loma Linda because his editor asked for an American Blue Zone, that he spotted it near a freeway exit driving to Las Vegas. And on city-level data, Newman is right: Loma Linda doesn’t stand out. Plenty of American communities post higher life expectancy. The Blue Zones marketing made this worse, too. Their own materials state that the average American lives to 78 but in Loma Linda men average 89 and women 91, and those are not city numbers. They’re figures that trace back to the Adventist cohort study, and Adventists are only a large minority of the town, roughly 9,000 of some 23,000 residents, around a third to two-fifths. That’s a cohort statistic dressed up as a geography statistic, and it shouldn’t be presented that way.

Why the Adventist cohort is the strongest evidence, not the weakest

So Newman lands a hit. And then something interesting happens, because when you look at what’s actually underneath Loma Linda, it isn’t the weak link. It’s the strongest evidence in the entire Blue Zones project.

Here’s the research. Fraser and Shavlik followed 34,192 California Seventh-day Adventists. Adventist men had life expectancy at age 30 running 7.28 years higher than other white Californians; women, 4.42 years higher. The authors described it as possibly the highest life expectancy of any formally described population.

Now watch what makes that different from every other Blue Zone. Every geographic Blue Zone has a confounding problem baked in. Sardinia, Okinawa, Ikaria, Nicoya: all of them entangle behavior with everything else about a place, isolated genetics, climate, altitude, local healthcare, historical food supply. You cannot cleanly separate “they eat this way” from “they’ve lived on an isolated island for a thousand years.”

Loma Linda has none of that. The Adventists live in the same country as their neighbors, shop at the same supermarkets, breathe the same air, use the same medical system, and sit in the same mediocre county, San Bernardino, which is not a healthy place. The one thing that systematically differs is what they do.

And there’s a dose-response signature on top of it. Adventist men live 7.3 years longer than other Californians; Adventist men who are vegetarian live 9.5 years longer. Same religion, same town, same everything: more of the behavior, more of the benefit. That’s exactly the pattern you’d expect if the behaviors are doing the work, and it’s the kind of gradient a place-based confounder can’t easily produce.

Here’s the irony, and it’s a big one. Newman attacked Loma Linda as the fake one, the soft target, the freeway-exit Blue Zone. But what he attacked was the packaging, the geographic label. Underneath that label is the cleanest natural experiment in the whole field. The zone with the weakest demographic credentials has the strongest causal evidence for the behaviors themselves. The Blue Zones takeaway (eat mostly plants, keep moving, don’t smoke, stay socially connected, have a sense of purpose) is better supported by Loma Linda than by any of the places with the impressive centenarian counts.

He hit the label. He never touched the finding.

This is also where the argument stops being only about demography and starts having a practical payoff. The Adventist cohort is worked in depth in two companion pieces: What the Longest-Living People Actually Eat traces the plant-forward dose-response gradient the cohort reveals, and What the Adventist Health Studies Actually Tell Us is an honest accounting of what that mortality data can and can’t establish.

Adding it up

So are the Blue Zones real or not?

Let’s total the scoreboard.

Okinawa: the 82% came from a genealogy book, Japan searched Okinawa in person and found zero missing, and the zone’s later fading tracks the loss of the traditional diet, not a records error. Ikaria: that was a pension payment roll, and Ikaria was validated years earlier on entirely different records. Nicoya: the Nicoya label came from a National Geographic trip, but the demography under it is independent, and the researcher Newman accused had already published the age exaggeration himself, and later published his own zone’s decline. Sardinia: a whole-island statistic aimed at a twelve-thousand-person zone it doesn’t describe, backed by a validation system that publicly deletes its own errors. Loma Linda: a fair hit on the marketing, and underneath it, the strongest behavioral evidence in the entire field. And the insurance argument that carried all of it into the mainstream runs the opposite direction from how insurance actually works.

Every load-bearing claim. Every single one.

What Saul Newman gets right, and why it isn’t about the Blue Zones

So is Saul Newman right about some things? Yes, almost certainly. And that’s exactly the point. Notice where he’s right: UN estimates for countries with no functioning birth registration, supplement clinics, biological-age hustles in Las Vegas, Silicon Valley blood transfusions. All of it defensible. None of it about the Blue Zones. He is a credible critic of the longevity industry, and he spent that credibility attacking the longevity research, where he doesn’t land a punch.

That is how half-truth works. It isn’t fabrication; fabrication is easy to catch. Half-truth is taking something real (extreme-age records genuinely are unreliable in much of the world) and stretching it across cases where it doesn’t apply. The true part carries the false part. The credibility earned on the accurate claims gets spent on the inaccurate ones.

And whether Newman means to do this doesn’t ultimately matter. You can’t measure what someone intended; you can measure what’s circulating. Right now, people believe Japan found 82% of its centenarians dead; it found about one percent. They believe firebombing explains Okinawa; Japan searched Okinawa and found zero missing. They believe Costa Rica’s data was never corrected; the researcher published the correction himself. They believe this is why their insurance costs more; it works the other way. People are repeating those numbers today believing they’re true. They aren’t. That’s the fact, regardless of intent.

The science was always narrower than the branding. That’s the real story here. It’s just less fun to post.

One last thing worth adding, because it’s the part with an actual takeaway. Once you accept the validated zones are real, the useful question is what the people in them ate, and there the answer is unusually consistent. An analysis of 154 dietary surveys across the five zones found that roughly 95% of what the longest-lived residents ate came from plants: beans above all, plus whole grains, vegetables, greens, tubers, and fruit, with meat a rarity. The demographic fight is about whether the zones exist; the more useful conversation is about the pattern they share. Both companion pieces below pick that up, including why the Loma Linda cohort, the one zone built on modern vital records, turns out to be the strongest evidence of all.

Related reading

Blue Zones Debunked or Not answers the viral thread claim by claim. What the Longest-Living People Actually Eat and What the Adventist Health Studies Actually Tell Us work the Loma Linda cohort in depth.

Sources: Austad & Pes, “The validity of Blue Zones demography: a response to critiques,” The Gerontologist (Dec 2025); Saito, Yong & Robine, “The mystery of Japan’s ‘missing centenarians’ explained,” Demographic Research (2012); Rosero-Bixby, “The Exceptionally High Life Expectancy of Costa Rican Nonagenarians,” Demography (2008); Rosero-Bixby, “The vanishing advantage of longevity in Nicoya,” Demographic Research (2023); Rosero-Bixby & Dow, PNAS (2016); Willcox et al., “They Really Are That Old,” Journals of Gerontology Series A (2008); Fraser & Shavlik, “Ten Years of Life: Is It a Matter of Choice?,” Archives of Internal Medicine (2001); Newman, bioRxiv preprint 704080 (2019, rev. 2024); U.S. Social Security Administration, Office of the Chief Actuary, Actuarial Study No. 120. Note: the “72%” Greek figure and the “186-year-old” registry entry come from Newman’s own estimate and contemporary press coverage respectively, not peer-reviewed sources.

Read more...

Coconut oil and animal fats (butter, lard, beef tallow) are both dominated by saturated fatty acids, but they are dominated by different saturated fatty acids, and those molecules behave differently in the body. Coconut oil’s signature fat, lauric acid (12 carbons), raises HDL cholesterol more powerfully than any other saturated fat, and in head-to-head trials coconut oil produces a better lipid profile than butter. That difference is real and reproducible. But beating butter, the single worst common fat for cholesterol, is a low bar, and clearing it is where coconut oil’s good news ends. Compared with liquid plant oils like olive, canola, soybean, or sunflower oil, coconut oil substantially raises LDL cholesterol, the single best-established causal driver of atherosclerosis, and the HDL it raises has never been shown to be functional or protective. Every attempt in medical history to buy cardiovascular safety with a higher HDL number has failed.

When you walk the entire hierarchy of evidence from cell biology to genetics to outcome trials, the ranking that emerges is remarkably consistent: liquid plant oils at the top, and coconut oil somewhere in the middle of the saturated pack (better than butter, roughly comparable to or slightly worse than tallow and lard on LDL, depending on the comparison). No version of “swap one solid fat for another” substitutes for the swap that actually reduces heart attacks, which is replacing saturated fat with unsaturated fat, the kind that comes overwhelmingly from plants. That is the destination. The rest of this article is the journey, rung by rung, up the evidence pyramid, including the honest complications, the studies skeptics cite, and where the legitimate uncertainty still lives.

What coconut oil and animal fats are actually made of

Lipid scientists stopped treating saturated fat as one substance decades ago, because saturated fatty acids of different carbon chain lengths interact differently with liver receptors, digestion pathways, and blood lipids. The four that matter most in food are lauric acid (C12), myristic acid (C14), palmitic acid (C16), and stearic acid (C18), and they are not evenly distributed across fats.

What is coconut oil made of?

Approximate fatty-acid profiles, as a percentage of total fat, look like this (figures vary by source and processing):

• Coconut oil: about 82–92% saturated. Lauric acid 45–48%, myristic 16–18%, palmitic 8–9%, stearic about 3%, the true medium-chain fats caprylic and capric (C8–C10) about 13%, monounsaturated fat about 6%, polyunsaturated about 2%.
• Butter: about 63–68% saturated. Palmitic 26–31%, myristic 10–12%, stearic 10–12%, lauric about 3%, monounsaturated about 24%, plus roughly 3% naturally occurring trans fat.
• Beef tallow: about 50–55% saturated. Palmitic 24–26%, stearic 17–19%, and about 40–43% monounsaturated oleic acid.
• Lard: about 39–42% saturated. Palmitic 24–26%, stearic 13–14%, and about 44–47% monounsaturated oleic acid.

Values are compiled from USDA FoodData Central and the fatty-acid tables in the American Heart Association Presidential Advisory [1].

Is coconut oil a medium-chain triglyceride (MCT) oil?

No, and this is the most common myth about it. True MCTs, the caprylic (C8) and capric (C10) acids in commercial MCT oil, are absorbed directly into the portal vein and burned rapidly by the liver, but they make up only about 13% of coconut oil. Lauric acid sits on a biochemical borderline: it is technically classified as medium-chain, yet during human digestion roughly 70–75% of it is packaged into chylomicrons and enters the bloodstream exactly like a long-chain animal fat, with only a minority fraction taking the rapid portal route [2, 28, 33]. Studies of purified C8/C10 MCT oil therefore cannot be extrapolated to coconut oil, which is itself about a quarter myristic-plus-palmitic acid, the same long-chain saturated fats found in butter and beef. The full journey these particles take, from bile to chylomicron to lymph to liver, and why the body routes dietary fat so differently from almost everything else you eat, is the subject of a companion piece, Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries.

Is all animal fat the same? Butter vs. beef tallow vs. lard

No. “Animal fat” is not one profile. Butter, a dairy fat, is the myristic-and-palmitic heavyweight, plus roughly 3% naturally occurring trans fats [1]. Beef tallow and lard, by contrast, carry relatively little myristic acid; their saturated fraction is mostly palmitic plus a large share of stearic acid, and stearic acid is the one saturated fat that does not raise LDL cholesterol, a fact established in a classic metabolic study by Bonanome and Grundy in 1988 [24] and confirmed in every pooled analysis since [3, 4]. Tallow and lard are also 40% or more monounsaturated oleic acid, the same fatty acid that dominates olive oil. This is why, gram for gram, butter is generally the worst common fat for LDL, with tallow and lard less bad, though still worse than any liquid plant oil.

Does coconut oil contain any heart-healthy fat?

Barely. Coconut oil has essentially no polyunsaturated fat and almost no monounsaturated fat. Whatever else is true of it, it displaces the fats with the strongest evidence for cardiovascular benefit, the unsaturated fats concentrated in vegetable oils, nuts, seeds, olives, and avocados.

Climbing the evidence pyramid

Good nutrition science is not one kind of study; it is the concordance of many kinds, each with different blind spots. Here is the climb, from cell biology at the bottom to genetics and outcome trials at the top.

How does saturated fat raise LDL cholesterol?

Your liver clears LDL particles from the blood using LDL receptors, molecular vacuum ports on the surface of liver cells. Landmark animal experiments by Spady, Woollett, and Dietschy in the 1980s and 1990s showed that long-chain saturated fatty acids (palmitic and myristic in particular) suppress LDL-receptor activity, so fewer LDL particles get pulled out of circulation, while unsaturated fats and cholesterol depletion up-regulate the receptors [25, 26]. This mechanism was later observed in humans: when men and women reduced saturated fat intake in a controlled feeding study, LDL-receptor abundance on their cells measurably increased, in proportion to the fall in their LDL cholesterol [27]. Receptor down-regulation, not some vague “clogging,” is the core molecular reason saturated fat raises LDL.

Where does lauric acid fit? Its partial medium-chain metabolism and its outsized effect on HDL make it mechanistically distinct, and it is often described as suppressing LDL receptors less aggressively than palmitic acid. That is plausible, but a candid reading of the literature is that the receptor biology of lauric acid specifically is far less studied in humans than that of palmitic acid, and the pooled human-feeding data complicate the tidy story, because per calorie, lauric acid raises LDL at least as much as palmitic acid does. Mechanism proposes; feeding studies dispose.

Does lauric acid raise cholesterol less than other saturated fats?

This is where the coconut debate really lives. Starting with the famous Keys and Hegsted equations of 1965 [17, 18], researchers fed people precisely controlled diets and measured lipid responses. Dozens of these trials were pooled in Mensink and Katan’s meta-analysis of 60 controlled trials (2003) [3] and updated in Mensink’s regression analysis commissioned by the World Health Organization (2016) [4]. Expressed as the effect of replacing 1% of dietary energy from carbohydrate, the per-fatty-acid results run like this:

• Lauric acid (C12): the largest LDL rise of any saturated fatty acid, but also the largest HDL rise of any fatty acid, so it is the only saturated fat that improves the total-to-HDL ratio (by about 0.033–0.037).
• Myristic acid (C14): raises LDL, raises HDL, roughly no change to the ratio.
• Palmitic acid (C16): raises LDL, small HDL rise, roughly no change to the ratio.
• Stearic acid (C18): roughly neutral on LDL, roughly neutral to slightly favorable on the ratio.
• Cis-unsaturated fats (monounsaturated and polyunsaturated): lower LDL and improve the ratio.

Read that carefully, because it contains the two facts that fuel the entire coconut-oil debate, and they point in opposite directions. First: in the pooled regressions, lauric acid was the most potent total- and LDL-cholesterol-raising saturated fatty acid, per calorie [3, 4]. That surprises people, and it directly contradicts a claim you will find in many pro-coconut articles: that lauric acid raises LDL less than palmitic acid gram for gram. The individual head-to-head trials are genuinely mixed. Denke and Grundy’s 1992 comparison found lauric slightly gentler on LDL than palmitic [19], Zock’s trial found myristic acid worse than palmitic [20], and Temme’s lauric-plus-myristic diet landed close to palmitic on LDL [21]. But the best pooled estimate does not support the idea that lauric acid is a weak LDL-raiser. It is not.

Does coconut oil improve your cholesterol ratio?

Here is the second fact: lauric acid raises HDL so much more than it raises LDL that it is the only saturated fatty acid that significantly improves the total-to-HDL cholesterol ratio relative to carbohydrate, by about 0.033 to 0.037 per 1% of energy [3, 4]. Myristic and palmitic acids barely move the ratio. This ratio effect is, as the AHA’s own science commentary acknowledged, the entire basis for the optimistic health claims made about coconut oil [1]. Mensink’s WHO analysis itself, however, cautioned against over-reading it: a favorable shift in the cholesterol ratio produced by a particular fatty acid should not be assumed to translate into lower cardiovascular risk unless outcome studies confirm it, and for lauric acid specifically, they never have [4]. So which fact wins, the LDL rise or the ratio improvement? That question cannot be answered at this rung. It gets answered at the top, by genetics and outcome trials.

One more feeding-study result is worth knowing. In 1985, a beef-industry-funded trial directly compared beef tallow, coconut oil, and safflower oil in the same subjects, and found plasma cholesterol was higher on coconut oil than on beef fat [23]. A decade later, Cox’s New Zealand trial compared coconut oil, butter, and safflower oil and found butter produced the highest LDL, coconut oil intermediate, safflower lowest [22]. Together these two small crossover studies preview the modern ranking: butter worst, coconut oil and tallow trading places in the middle depending on the outcome, unsaturated oil best.

Coconut oil vs. butter vs. olive oil: what the trials show

The single most-discussed modern trial is the BBC-recruited “COB” study led by Kay-Tee Khaw at Cambridge (2018): 96 healthy adults aged 50 to 75 randomized to 50 grams daily of extra-virgin coconut oil, butter, or extra-virgin olive oil for four weeks [5]. Butter raised LDL significantly compared with coconut oil (+0.42 mmol/L, about 16 mg/dL) and compared with olive oil (+0.38), while coconut oil and olive oil did not differ significantly on LDL (a difference of just -0.04). Coconut oil raised HDL versus both butter (+0.18) and olive oil (+0.16), and butter worsened the total-to-HDL ratio versus coconut oil while coconut oil matched olive oil on the ratio [5].

This trial is the strongest card in the pro-coconut hand, and it deserves to be reported honestly: in this one four-week study, coconut oil looked clearly better than butter and, on LDL, statistically indistinguishable from olive oil. It also deserves its caveats. It was a single short trial in healthy older adults, its coconut-versus-olive LDL result is an outlier against the pooled literature, a 2024 secondary analysis of its plasma fatty acids confirmed the expected metabolic signatures of each fat [5], and its own authors wrote that the findings do not change the recommendation to reduce saturated fat overall [5].

Other modern trials fill in the picture. Maki’s 2018 randomized crossover found corn oil significantly lowered LDL and total cholesterol compared with coconut oil [29]. In Kerala, India, coconut country, Vijayakumar randomized about 200 patients with stable coronary disease to coconut oil or sunflower oil as their cooking fat for two years and found no significant differences in lipids or clinical events, a genuinely reassuring null, but from a small study using modest cooking-oil quantities within a mixed diet [30]. Small virgin-coconut-oil (VCO) trials, including Cardoso’s study in Brazilian cardiac patients (higher HDL, smaller waists) [31], Assunção’s in Brazilian women with abdominal obesity, and Harris’s against high-oleic safflower, are frequently cited by coconut marketers; they are short, small, often open-label. A systematic review and meta-analysis of 14 VCO trials totaling 1,049 participants, published in late 2025, concluded the clinical evidence for VCO’s proposed metabolic benefits remains mixed, with interventions lasting only 2 to 24 weeks [32].

What did the big coconut oil meta-analysis find?

The definitive pooling is Neelakantan, Seah, and van Dam’s systematic review and meta-analysis of 16 trials, published in Circulation in 2020 [2]. Compared with nontropical vegetable oils, coconut oil consumption raised total cholesterol by 14.69 mg/dL, LDL cholesterol by 10.47 mg/dL (about 8.6%), and HDL cholesterol by 4.00 mg/dL (about 7.8%), with no significant change in triglycerides. Coconut oil also had no significant effect on body weight, body fat, waist circumference, fasting glucose, or C-reactive protein compared with vegetable oils, quietly demolishing the belly-fat, blood-sugar, and anti-inflammatory claims in one table [2]. Strikingly, coconut oil raised LDL even when compared with palm oil, the other tropical saturated oil [2, 56]. The accompanying editorial by Frank Sacks concluded that coconut oil may be viewed as one of the most deleterious cooking oils for cardiovascular risk, while acknowledging it behaves more favorably than butter [56]. So the biomarker verdict is precise. Against butter, coconut wins. Against liquid plant oils, coconut clearly loses on LDL and “wins” on HDL, and whether that HDL win is worth anything is a question biomarker trials cannot answer.

Does cutting saturated fat actually prevent heart attacks?

Yes. No randomized trial of coconut oil, tallow, or butter specifically has ever been run with heart attacks and strokes as endpoints. What we have are the decades-long saturated-fat-reduction trials, pooled in the 2020 Cochrane review by Hooper and colleagues: roughly 56,000 to 59,000 participants, showing that reducing saturated fat intake cut combined cardiovascular events by 17% (risk ratio 0.83, 95% confidence interval 0.70 to 0.98), with a dose-response, so the more saturated fat and serum cholesterol came down, the fewer the events, although effects on total mortality were small or absent over the trial durations [6]. The number needed to treat was about 56 people reducing saturated fat for four years to prevent one cardiovascular event [6]. The AHA’s separate “core trials” analysis estimated that replacing saturated fat with polyunsaturated vegetable oil reduced cardiovascular disease by roughly 30%, a statin-sized effect [1]. This is the rung that converts “LDL went up 10 mg/dL” from an abstraction into an expectation of harm.

Did coconut-eating populations avoid heart disease?

This is the strongest card the other side holds, so it deserves a fair hearing. Prospective cohort studies broadly agree with the trials, with instructive wrinkles: replacing saturated fat with polyunsaturated fat is associated with substantially lower coronary risk, while replacing it with refined carbohydrate buys nothing [1, 34]. The source of saturated fat matters too, with dairy-derived saturated fat associating more neutrally than meat-derived saturated fat in cohorts like MESA [36], and modeling in the Nurses’ Health Study and Health Professionals cohorts found that swapping butter or margarine for olive oil predicted meaningfully lower cardiovascular risk [38].

Then there are the coconut peoples. Ian Prior’s celebrated “natural experiment” (1981) compared two Polynesian atoll populations: Pukapukans obtained about 34% of their energy from coconut and Tokelauans about 63%, probably the highest saturated-fat intake ever documented in a free-living population [15]. Tokelauans’ serum cholesterol ran 35 to 40 mg/dL higher than Pukapukans’, exactly what the Keys, Hegsted, and Mensink arithmetic predicts, yet the investigators reported that vascular disease appeared uncommon in both groups [15]. Staffan Lindeberg’s studies of Kitava, Papua New Guinea, where coconut is a staple alongside tubers, fruit, and fish, likewise found no apparent stroke or ischemic heart disease [16].

What do these prove? Less than either side wants. These populations ate whole coconut flesh and cream (with its fiber and food matrix, not refined oil), almost no processed food or sugar, abundant fish, and lived intensely active, largely non-obese lives, and event ascertainment on remote atolls in that era was limited. What they legitimately show is that a high-saturated-fat intake from whole coconut does not produce an inevitable epidemic of heart disease within a traditional dietary pattern. Tellingly, when Tokelauans migrated to New Zealand and westernized their diets, their risk profiles deteriorated [15]. They do not show that adding refined coconut oil to a modern Western diet is safe, which is the actual question most readers are asking.

Is LDL cholesterol really the cause of heart disease?

Yes, and this is the tiebreaker for the LDL-up, HDL-up puzzle. The 2017 European Atherosclerosis Society consensus assembled evidence from more than 200 prospective studies, Mendelian randomization studies, and randomized trials, covering over 2 million participants, 20 million person-years, and 150,000 cardiovascular events, demonstrating a consistent, dose-dependent, log-linear relationship between LDL exposure and atherosclerotic disease, in which any mechanism that lowers LDL particle concentration lowers risk proportionally [7]. People born with gene variants that lower LDL have proportionally less heart disease; people born with variants that raise it have more. LDL, more precisely the total count of ApoB-carrying particles, is not a “marker.” It is the payload. Exactly how one of those ApoB particles crosses into the artery wall, binds, oxidizes, and hardens into plaque, the molecular crime scene beneath this entire ranking, is walked through step by step in Cholesterol, From the Ground Up.

Does raising HDL protect your heart?

No, and this is the fact that sinks the coconut argument. In the mirror-image Mendelian randomization study, people genetically endowed with lifelong-higher HDL cholesterol had no reduction in heart attacks [8]. Pharmacology agrees with genetics: niacin raised HDL substantially in the AIM-HIGH and HPS2-THRIVE trials and prevented nothing [10, 11]; the CETP inhibitor torcetrapib raised HDL by about 70% and increased deaths [12]; dalcetrapib and evacetrapib were futile; and anacetrapib’s modest benefit in the 30,000-patient REVEAL trial was fully explained by its LDL and ApoB lowering, not its HDL raising [13, 14]. Every road to cardiovascular protection that ran only through a higher HDL number has been a dead end. This is why the “coconut oil improves the ratio” argument fails at the top of the pyramid even though it is true at the bottom. You cannot manufacture protection by inflating the denominator. The heart attacks track the ApoB particles, and coconut oil raises them.

Does coconut oil’s HDL boost actually do anything?

Since HDL quantity is not causal, the research frontier moved to HDL function.

What is cholesterol efflux capacity?

The leading measure of HDL function is cholesterol efflux capacity (CEC), a laboratory assay of how well a person’s HDL actually pulls cholesterol out of macrophages, the cells that become arterial foam cells. CEC has real credentials: in the Dallas Heart Study, people in the top quartile of efflux capacity had about a 67% lower risk of incident cardiovascular events than the bottom quartile, independent of their HDL cholesterol level [9], a finding replicated in other cohorts [59]. When those results were presented, one of the field’s leading HDL researchers publicly declared the old HDL-cholesterol hypothesis ready for retirement in favor of an HDL-function hypothesis [9].

Does coconut oil improve HDL function?

There is no evidence that it does. No adequately sized human trial has demonstrated that coconut oil improves cholesterol efflux capacity or any other validated measure of HDL function. The claim sometimes made in coconut oil’s favor, that its HDL boost is “functional,” is simply unsupported. The claim sometimes made against it, that the extra HDL is definitely dysfunctional “passive cargo,” is a plausible mechanistic hypothesis that also has not been proven in humans. It is a genuine data gap, and intellectual honesty requires labeling it as one. For contrast, extra-virgin olive oil, the fat coconut oil is usually compared against, has randomized trials showing its polyphenols enhance macrophage cholesterol efflux and improve HDL’s functional quality [57], though even there the literature is mixed, with the recent OLIVAUS trial finding improved HDL lipid composition but no significant change in efflux capacity [58]. If the best-studied oil’s HDL-function evidence is mixed, coconut oil’s is essentially a blank page.

The clinical bottom line does not depend on resolving it. Run the arithmetic from the Circulation meta-analysis: HDL up 4 mg/dL, LDL up 10.5 mg/dL versus vegetable oils [2]. Even in the fantasy scenario where every milligram of that HDL were perfectly functional, no evidence anywhere suggests it could neutralize a 10-point rise in causal, plaque-forming LDL particles, because nothing that merely raises HDL has ever neutralized anything. Meanwhile the LDL side of the ledger requires no speculation at all.

What health organizations say about coconut oil and saturated fat

One of the most under-appreciated facts in this debate is the sheer breadth of independent expert bodies, across countries, specialties, and even ideological orientations toward diet, that have reviewed this evidence and landed in the same place.

• American Heart Association (2017 Presidential Advisory): replace saturated with unsaturated fat (about a 30% event reduction in core trials); because coconut oil raises LDL, a cause of cardiovascular disease, and has no known offsetting effects, the advisory explicitly stated, “we advise against the use of coconut oil.” For people with elevated LDL, it advises 5 to 6% of calories from saturated fat [1].
• AHA 2021 Dietary Guidance and the American College of Cardiology: use liquid plant oils in place of animal fats (butter, lard) and tropical oils (coconut, palm kernel) [46].
• ACC/AHA 2019 Primary Prevention Guideline: replace saturated fat with mono- and polyunsaturated fat within a plant-forward pattern [45].
• National Lipid Association: limit saturated fat to under 7% of calories for adults [45, 46].
• World Health Organization (2023 guideline): reduce saturated fat to 10% of energy or less (a strong recommendation for adults and children), replacing it with polyunsaturated and monounsaturated fats from plant sources [41].
• National Academies (Institute of Medicine), Dietary Reference Intakes: no safe threshold identified; keep saturated fat intake as low as possible within a nutritionally adequate diet [43].
• Dietary Guidelines for Americans, 2025–2030 (released January 7, 2026): retains the long-standing limit of under 10% of calories from saturated fat, even as its new food graphic and lists controversially spotlight butter, whole milk, red meat, and beef tallow, a tension flagged publicly by Harvard nutrition scientists. The scientific arm, the 2025 Dietary Guidelines Advisory Committee, had concluded that substituting butter with plant oils is associated with lower cardiovascular disease risk [44].
• UK Scientific Advisory Committee on Nutrition (2019, 47 systematic reviews): reduce saturated fat to 10% of energy or less; replace with unsaturated fats, for example oils instead of butter [42].
• European Society of Cardiology and European Atherosclerosis Society: LDL is causal; dietary saturated fat should be minimized and replaced with unsaturated fat to lower LDL [7, 47].
• Academy of Nutrition and Dietetics: its 2023 evidence-based practice guideline found moderate-certainty evidence that reducing saturated fat lowers cardiovascular events, prioritizing replacement with polyunsaturated fat, and focusing on total saturated fat rather than any single food source [48, 49].
• American Diabetes Association: saturated fat raises cholesterol and cardiovascular risk; emphasize Mediterranean-style patterns replacing saturated with unsaturated fat [50].
• American Cancer Society (2020 guideline): the healthiest patterns are plant-based and include unsaturated fats; limit red and processed meat [51].
• American Academy of Pediatrics (endorsing NHLBI pediatric guidelines): limit saturated fat to about 8 to 10% of calories from age 2 [52].
• American College of Lifestyle Medicine: the optimal pattern centers on whole, minimally processed plant foods and minimizes red and processed meat, foods high in saturated fat, and ultra-processed foods [53].
• Heart Foundations of Australia, New Zealand, and Canada: reviewed coconut oil specifically and recommend unsaturated cooking oils instead of coconut oil, butter, and other solid fats [42, 46].
• JACC: Advances, “Clinician’s Guide to Trending Cardiovascular Nutritional Controversies” (2026): on the beef-tallow revival, evidence of benefit is lacking and evidence of possible harm is considerable; no data support tallow as healthier than seed oils or tropical oils; each 5% of energy swapped from animal saturated fat to unsaturated fat corresponds to about 10% lower cardiovascular risk [54].
• NutritionFacts.org (Dr. Michael Greger): reviews the same trial literature and concludes coconut oil’s LDL raising is not offset by its HDL bump [55].

Notice what is on this list: the cardiologists, the dietitians, the diabetes and cancer societies, the pediatricians, the lifestyle-medicine and plant-based communities, the U.S. National Academies, the UK’s independent scientific committee, and the WHO, bodies that disagree with each other about plenty, all converging on the same two-part conclusion: limit saturated fat regardless of source, and replace it with unsaturated fat, especially from plants. No major scientific organization anywhere recommends coconut oil, tallow, lard, or butter as a heart-healthy fat. Even the 2025–2030 U.S. Dietary Guidelines, produced amid an administration openly enthusiastic about whole milk and beef tallow, could not bring itself to lift the 10% saturated-fat cap, because the underlying evidence review did not support lifting it [44].

The strongest arguments for saturated fat, examined

A credible article steelmans the other side, so here are the strongest cards in the “saturated fat is fine” deck, and why they change the nuance but not the conclusion.

“The Cochrane review found no mortality benefit”

True. Reducing saturated fat clearly cut cardiovascular events (by 17%) but showed little effect on death rates within trial timeframes [6]. Trials lasting a few years in mixed-risk populations are underpowered for mortality; the event reduction, its dose-response with cholesterol lowering, and the genetics all point the same direction. But it is fair to say the dietary evidence is strongest for preventing heart attacks, not for extending lifespan per se. Some critics press further. Zoë Harcombe, for instance, has argued that in a sensitivity analysis restricted to the trials that actually achieved a measured fall in saturated fat, even the cardiovascular-event benefit loses statistical significance [60]. That is a caveat worth stating plainly, but it does not survive contact with the rest of the pyramid: Cochrane’s pre-specified primary analysis remained significant, the benefit scaled with the degree of cholesterol lowering, and the feeding-trial and genetic evidence converge on the same causal mechanism from entirely independent directions. One underpowered subgroup does not overturn a dose-responsive result echoed at every other rung.

“The food matrix matters, cheese and yogurt aren’t butter”

Also true, and the mainstream has absorbed it. A prominent 2020 JACC review led by Arne Astrup argued that whole-food sources of saturated fat (fermented dairy especially) do not carry the risk their fatty-acid content predicts [40], and cohort data on dairy back this up [36, 37]. The 2023 Academy of Nutrition and Dietetics guideline explicitly weighed this and concluded the evidence still favors reducing total saturated fat rather than exonerating categories [49]. Note that the food-matrix argument helps yogurt and cheese far more than it helps refined coconut oil or rendered tallow, which are stripped fats with no matrix at all.

“PURE and the Polynesian studies show high saturated fat is fine”

The PURE study of 18 countries reported that higher saturated fat intake was associated with lower stroke risk and no increase in heart disease [39], a finding that generated enormous headlines and equally enormous methodological critique, since its low-saturated-fat populations were largely poor, eating white-rice-dominated diets, so “more saturated fat” partly proxied for “less poverty and less refined starch.” The Polynesian data have their own limitations: whole-food coconut, a traditional lifestyle, and limited surveillance on the atolls. These are real observations with real limitations. They demonstrate context-dependence; they do not license adding solid fats to a Western diet.

“The big reviews found no link between saturated fat and heart disease”

Three widely cited meta-analyses of prospective cohorts, Siri-Tarino (2010), Chowdhury (2014), and de Souza (2015), did report neutral associations between saturated fat intake and coronary events [61, 62, 63], and coconut and low-carb advocates cite them constantly. The catch is what these studies did and did not ask: an observational association between “how much saturated fat people ate” and “who had heart attacks” says nothing about what replaced it. When the replacement nutrient is specified, the picture flips. Pooling eleven cohorts, swapping saturated fat for polyunsaturated fat was associated with meaningfully lower coronary risk, while swapping it for refined carbohydrate was not [34]. “Saturated fat is not linked to heart disease in a model that ignores the alternative” and “replacing saturated fat with plant oils lowers risk” are both true, and only the second one is a dietary recommendation.

“Swapping saturated fat for vegetable oil didn’t help in the old trials”

This rests on two recovered-data reanalyses, the Sydney Diet Heart Study and the Minnesota Coronary Experiment, in which replacing saturated fat with linoleic-acid-rich oils lowered cholesterol but did not reduce, and in older Minnesota subjects appeared to worsen, mortality [64, 65]. They are legitimate and sobering data. But both interventions leaned heavily on omega-6 linoleic acid in forms that in that era often carried trans fats, both had high dropout and decades-late data recovery, and both are outweighed by the larger body of substitution trials pooled in Cochrane, where replacing saturated fat with unsaturated fat did cut events [6]. They are a genuine argument against the specific claim that any polyunsaturated swap is automatically protective, not evidence that saturated fat is benign.

“What about insulin, inflammation, and triglycerides?”

Coconut oil’s meta-analyzed effects on glycemia, inflammation, and adiposity versus vegetable oils are null [2], so there is no compensating benefit hiding in those columns. For tallow, the newest clinical review actually tilts the other way, citing evidence of endothelial dysfunction with high-heat tallow cooking, impaired insulin sensitivity, and liver-fat promotion [54].

Where does real uncertainty remain?

Three places, and readers deserve to know them. First, no hard-outcome trial of coconut oil or any single cooking fat exists, so all fat-versus-fat rankings rest on validated surrogates (LDL and ApoB) plus triangulation. Second, the HDL functional effects of coconut oil are unstudied, in either direction. Third, individual responses vary widely, since the same tablespoon moves one person’s LDL far more than another’s, which is an argument for testing rather than guessing.

The verdict, and what to do with it

Stack every rung of the pyramid and the picture is coherent from bottom to top.

Is coconut oil healthier than butter?

On the numbers, yes. Coconut oil beats butter on lower LDL, higher HDL, and a better ratio, in trials and meta-analyses alike [2, 5, 22, 28]. But “healthier than the worst common cooking fat” is not the same as “healthy.” Butter is the benchmark coconut oil clears, and clearing it does not make coconut oil a good choice against the liquid plant oils that outperform both.

Is coconut oil or beef tallow better for you?

The contest is closer and less studied than the coconut-versus-butter one. Tallow and lard have stearic-and-oleic-heavy profiles that make them less LDL-raising than butter, and in the one direct comparison coconut oil raised cholesterol more than beef fat [23]. Calling coconut oil “the healthiest saturated fat” and calling it “as bad as lard” are both defensible readings of different endpoints, which is precisely why the fixation on ranking solid fats against each other misses the point. Neither tallow nor coconut oil has evidence of being a healthy alternative, only a marginally different one [22, 23, 54].

What is the healthiest cooking oil?

Every solid fat in this article, coconut, butter, tallow, and lard, raises LDL and ApoB relative to olive, canola, soybean, sunflower, or corn oil [1, 2, 6, 29, 54], and LDL and ApoB are causally, cumulatively, and dose-dependently linked to atherosclerosis [7]. The swap with proven event reduction is saturated to unsaturated [1, 6, 34]. That is the move; the rest is decoration. If the reflex objection here is that those liquid plant oils are the demonized “seed oils,” that claim gets the same rung-by-rung treatment, and collapses at every level of the hierarchy, in Saturated Fat, Seed Oils, and Heart Disease. In practice: make olive oil (or canola, or another liquid plant oil) the default cooking and dressing fat. Zoom out from the single bottle and the same signal only strengthens, because the eating patterns with the deepest evidence base for heart health, the ones the WHO, the dietitians, and the cancer, cardiology, and lifestyle-medicine bodies converge on, are built predominantly on whole plant foods, with unsaturated plant fats doing the cooking. Coconut oil’s honest place in that picture is an occasional flavor, not a foundation. A tablespoon in a curry within an otherwise unsaturated-fat, whole-food pattern is a rounding error; three tablespoons a day “for health” is a 10 mg/dL LDL experiment on yourself with no demonstrated upside [2].

Should you get an ApoB test?

If you eat meaningful amounts of any of these fats, do not argue about it, measure it. An ApoB test counts your actual atherogenic particle burden and cuts through every HDL-related ambiguity in this article, and a lipid panel 6 to 8 weeks after a dietary change will tell you what your liver thinks of your cooking fat. A low triglyceride-to-HDL ratio is a reasonable sign of metabolic health, but it is not a hall pass for a high ApoB.

The nuance in lipid science is real. Chain length matters; lauric acid genuinely is a different molecule from palmitic acid; coconut oil genuinely is better than butter; stearic acid genuinely is neutral; traditional coconut-eating populations genuinely did thrive. And after all of that nuance is fully honored, the hierarchy of evidence still funnels to a conclusion of almost boring stability, unchanged from the WHO to the AHA to the dietitians to the lifestyle-medicine physicians: the fats that protect hearts pour, and they come, overwhelmingly, from plants.

Keep reading

For readers who want to go deeper into the primary research, NutritionFacts.org maintains a free, ad-free library of evidence reviews on these exact topics, and every organizational report cited below is publicly available. This post is one stop on a tour through the same evidence pyramid, applied to different corners of the dietary-fat debate:

• Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries follows a single fatty meal from bile to chylomicron to liver, contrasts the animal and plant “fat kingdoms,” and takes apart the fad-diet “I feel great” illusion that powers every keto and carnivore testimonial.
• Saturated Fat, Seed Oils, and Heart Disease confronts the “seed oils are poison” narrative head-on, and shows why the very liquid plant oils that outrank coconut oil in this article lower heart disease at every level of evidence.
• Cholesterol, From the Ground Up is the molecular crime scene beneath this whole ranking: how an ApoB particle crosses into the artery wall, gets trapped, and builds a plaque, and why your arteries can be silently worsening while you feel your best.

This article is for educational purposes and is not medical advice. Lipid responses to diet vary between individuals; decisions about diet, cholesterol testing, and cardiovascular risk should be made with your own clinician.

References

1. Sacks FM, Lichtenstein AH, Wu JHY, et al. Dietary Fats and Cardiovascular Disease: A Presidential Advisory From the American Heart Association. Circulation. 2017;136:e1–e23. doi:10.1161/CIR.0000000000000510
2. Neelakantan N, Seah JYH, van Dam RM. The Effect of Coconut Oil Consumption on Cardiovascular Risk Factors: A Systematic Review and Meta-Analysis of Clinical Trials. Circulation. 2020;141:803–814. doi:10.1161/CIRCULATIONAHA.119.043052
3. Mensink RP, Zock PL, Kester ADM, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr. 2003;77:1146–1155.
4. Mensink RP; World Health Organization. Effects of Saturated Fatty Acids on Serum Lipids and Lipoproteins: A Systematic Review and Regression Analysis. Geneva: WHO; 2016.
5. Khaw KT, Sharp SJ, Finikarides L, et al. Randomised trial of coconut oil, olive oil or butter on blood lipids and other cardiovascular risk factors in healthy men and women. BMJ Open. 2018;8:e020167. (Secondary fatty-acid analysis: Sowah SA, Koulman A, Sharp SJ, Imamura F, Khaw KT, Forouhi NG. J Lipid Res. 2024;65:100681.)
6. Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2020;(8):CD011737.
7. Ference BA, Ginsberg HN, Graham I, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel. Eur Heart J. 2017;38:2459–2472.
8. Voight BF, Peloso GM, Orho-Melander M, et al. Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study. Lancet. 2012;380:572–580.
9. Rohatgi A, Khera A, Berry JD, et al. HDL cholesterol efflux capacity and incident cardiovascular events. N Engl J Med. 2014;371:2383–2393.
10. AIM-HIGH Investigators. Niacin in patients with low HDL cholesterol levels receiving intensive statin therapy. N Engl J Med. 2011;365:2255–2267.
11. HPS2-THRIVE Collaborative Group. Effects of extended-release niacin with laropiprant in high-risk patients. N Engl J Med. 2014;371:203–212.
12. Barter PJ, Caulfield M, Eriksson M, et al. Effects of torcetrapib in patients at high risk for coronary events (ILLUMINATE). N Engl J Med. 2007;357:2109–2122.
13. HPS3/TIMI55–REVEAL Collaborative Group. Effects of anacetrapib in patients with atherosclerotic vascular disease. N Engl J Med. 2017;377:1217–1227.
14. Schmidt AF, Hunt NB, Gordillo-Marañón M, et al. Cholesteryl ester transfer protein (CETP) as a drug target for cardiovascular disease. Nat Commun. 2021;12:5640.
15. Prior IA, Davidson F, Salmond CE, Czochanska Z. Cholesterol, coconuts, and diet on Polynesian atolls: a natural experiment: the Pukapuka and Tokelau island studies. Am J Clin Nutr. 1981;34:1552–1561.
16. Lindeberg S, Nilsson-Ehle P, Terént A, Vessby B, Scherstén B. Cardiovascular risk factors in a Melanesian population apparently free from stroke and ischaemic heart disease: the Kitava study. J Intern Med. 1994;236:331–340.
17. Keys A, Anderson JT, Grande F. Serum cholesterol response to changes in the diet: IV. Particular saturated fatty acids in the diet. Metabolism. 1965;14:776–787.
18. Hegsted DM, McGandy RB, Myers ML, Stare FJ. Quantitative effects of dietary fat on serum cholesterol in man. Am J Clin Nutr. 1965;17:281–295.
19. Denke MA, Grundy SM. Comparison of effects of lauric acid and palmitic acid on plasma lipids and lipoproteins. Am J Clin Nutr. 1992;56:895–898.
20. Zock PL, de Vries JH, Katan MB. Impact of myristic acid versus palmitic acid on serum lipid and lipoprotein levels in healthy women and men. Arterioscler Thromb. 1994;14:567–575.
21. Temme EH, Mensink RP, Hornstra G. Comparison of the effects of diets enriched in lauric, palmitic, or oleic acids on serum lipids and lipoproteins in healthy women and men. Am J Clin Nutr. 1996;63:897–903.
22. Cox C, Mann J, Sutherland W, Chisholm A, Skeaff M. Effects of coconut oil, butter, and safflower oil on lipids and lipoproteins in persons with moderately elevated cholesterol levels. J Lipid Res. 1995;36:1787–1795.
23. Reiser R, Probstfield JL, Silvers A, et al. Plasma lipid and lipoprotein response of humans to beef fat, coconut oil and safflower oil. Am J Clin Nutr. 1985;42:190–197.
24. Bonanome A, Grundy SM. Effect of dietary stearic acid on plasma cholesterol and lipoprotein levels. N Engl J Med. 1988;318:1244–1248.
25. Spady DK, Dietschy JM. Dietary saturated triacylglycerols suppress hepatic low density lipoprotein receptor activity in the hamster. Proc Natl Acad Sci USA. 1985;82:4526–4530.
26. Woollett LA, Spady DK, Dietschy JM. Saturated and unsaturated fatty acids independently regulate low density lipoprotein receptor activity and production rate. J Lipid Res. 1992;33:77–88.
27. Mustad VA, Etherton TD, Cooper AD, et al. Reducing saturated fat intake is associated with increased levels of LDL receptors on mononuclear cells in healthy men and women. J Lipid Res. 1997;38:459–468.
28. Eyres L, Eyres MF, Chisholm A, Brown RC. Coconut oil consumption and cardiovascular risk factors in humans. Nutr Rev. 2016;74:267–280.
29. Maki KC, Hasse W, Dicklin MR, et al. Corn oil lowers plasma cholesterol compared with coconut oil in adults with above-desirable levels of cholesterol. J Nutr. 2018;148:1556–1563.
30. Vijayakumar M, Vasudevan DM, Sundaram KR, et al. A randomized study of coconut oil versus sunflower oil on cardiovascular risk factors in patients with stable coronary heart disease. Indian Heart J. 2016;68:498–506.
31. Cardoso DA, Moreira ASB, de Oliveira GMM, Luiz RR, Rosa G. A coconut extra virgin oil-rich diet increases HDL cholesterol and decreases waist circumference and body mass in coronary artery disease patients. Nutr Hosp. 2015;32:2144–2152.
32. The effect of virgin coconut oil (VCO) on cardiovascular disease risk factors: a systematic review and meta-analysis (14 RCTs, 1,049 participants). Diabetol Metab Syndr. 2025. doi:10.1186/s13098-025-02019-6
33. Panth N, Abbott KA, Dias CB, Wynne K, Garg ML. Differential effects of medium- and long-chain saturated fatty acids on blood lipid profile: a systematic review and meta-analysis. Am J Clin Nutr. 2018;108:675–687.
34. Jakobsen MU, O’Reilly EJ, Heitmann BL, et al. Major types of dietary fat and risk of coronary heart disease: a pooled analysis of 11 cohort studies. Am J Clin Nutr. 2009;89:1425–1432.
35. Zong G, Li Y, Wanders AJ, et al. Intake of individual saturated fatty acids and risk of coronary heart disease in US men and women: two prospective longitudinal cohort studies. BMJ. 2016;355:i5796.
36. de Oliveira Otto MC, Mozaffarian D, Kromhout D, et al. Dietary intake of saturated fat by food source and incident cardiovascular disease: the Multi-Ethnic Study of Atherosclerosis. Am J Clin Nutr. 2012;96:397–404.
37. Pimpin L, Wu JHY, Haskelberg H, Del Gobbo L, Mozaffarian D. Is butter back? A systematic review and meta-analysis of butter consumption and risk of cardiovascular disease, diabetes, and total mortality. PLoS One. 2016;11:e0158118.
38. Guasch-Ferré M, Liu G, Li Y, et al. Olive oil consumption and cardiovascular risk in U.S. adults. J Am Coll Cardiol. 2020;75:1729–1739.
39. Dehghan M, Mente A, Zhang X, et al. Associations of fats and carbohydrate intake with cardiovascular disease and mortality in 18 countries (PURE). Lancet. 2017;390:2050–2062.
40. Astrup A, Magkos F, Bier DM, et al. Saturated fats and health: a reassessment and proposal for food-based recommendations: JACC state-of-the-art review. J Am Coll Cardiol. 2020;76:844–857.
41. World Health Organization. Saturated Fatty Acid and Trans-Fatty Acid Intake for Adults and Children: WHO Guideline. Geneva: WHO; 2023.
42. Scientific Advisory Committee on Nutrition (SACN). Saturated Fats and Health. London: Public Health England; 2019.
43. Institute of Medicine (National Academies). Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, DC: National Academies Press; 2005.
44. U.S. Department of Health and Human Services and U.S. Department of Agriculture. Dietary Guidelines for Americans, 2025–2030 (released January 7, 2026), and Scientific Report of the 2025 Dietary Guidelines Advisory Committee (December 2024). dietaryguidelines.gov
45. Arnett DK, Blumenthal RS, Albert MA, et al. 2019 ACC/AHA Guideline on the Primary Prevention of Cardiovascular Disease. Circulation. 2019;140:e596–e646.
46. Lichtenstein AH, Appel LJ, Vadiveloo M, et al. 2021 Dietary Guidance to Improve Cardiovascular Health: A Scientific Statement From the American Heart Association. Circulation. 2021;144:e472–e487.
47. Mach F, Baigent C, Catapano AL, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias. Eur Heart J. 2020;41:111–188.
48. Vannice G, Rasmussen H. Position of the Academy of Nutrition and Dietetics: dietary fatty acids for healthy adults. J Acad Nutr Diet. 2014;114:136–153.
49. Academy of Nutrition and Dietetics. Saturated fat intake and the prevention and management of cardiovascular disease in adults: an evidence-based nutrition practice guideline. J Acad Nutr Diet. 2023.
50. Evert AB, Dennison M, Gardner CD, et al. Nutrition therapy for adults with diabetes or prediabetes: a consensus report. Diabetes Care. 2019;42:731–754; and American Diabetes Association, Standards of Care in Diabetes (annual).
51. Rock CL, Thomson C, Gansler T, et al. American Cancer Society guideline for diet and physical activity for cancer prevention. CA Cancer J Clin. 2020;70:245–271.
52. Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk Reduction in Children and Adolescents (NHLBI; endorsed by the American Academy of Pediatrics). Pediatrics. 2011;128(Suppl 5):S213–S256.
53. American College of Lifestyle Medicine. Updated dietary position statement for the treatment and prevention of chronic disease. lifestylemedicine.org (2025–2026 update).
54. A Clinician’s Guide for Trending Cardiovascular Nutritional Controversies in 2026 (beef tallow, seed oils, dairy fat, sweeteners). JACC: Advances. 2026.
55. Greger M. Coconut Oil and the Boost in HDL “Good” Cholesterol. NutritionFacts.org (video series and transcripts).
56. Sacks FM. Coconut oil and heart health: fact or fiction? Circulation. 2020;141:815–817.
57. Helal O, Berrougui H, Loued S, Khalil A. Extra-virgin olive oil consumption improves the capacity of HDL to mediate cholesterol efflux. Br J Nutr. 2013;109:1844–1852; Hernáez Á, et al. Arterioscler Thromb Vasc Biol. 2014;34:2115–2119.
58. OLIVAUS trial: Extra virgin olive oil improves HDL lipid fraction but not HDL-mediated cholesterol efflux capacity: a double-blind, randomised, controlled, cross-over study. Br J Nutr. 2023.
59. Saleheen D, Scott R, Javad S, et al. Association of HDL cholesterol efflux capacity with incident coronary heart disease events: a prospective case-control study. Lancet Diabetes Endocrinol. 2015;3:507–513.
60. Harcombe Z. Cochrane saturated fat reviews (analysis of Hooper et al. 2020, Reduction in saturated fat intake for cardiovascular disease). zoeharcombe.com; June 2020.
61. Siri-Tarino PW, Sun Q, Hu FB, Krauss RM. Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease. Am J Clin Nutr. 2010;91:535–546.
62. Chowdhury R, Warnakula S, Kunutsor S, et al. Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-analysis. Ann Intern Med. 2014;160:398–406.
63. de Souza RJ, Mente A, Maroleanu A, et al. Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies. BMJ. 2015;351:h3978.
64. Ramsden CE, Zamora D, Leelarthaepin B, et al. Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis. BMJ. 2013;346:e8707.
65. Ramsden CE, Zamora D, Majchrzak-Hong S, et al. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968–73). BMJ. 2016;353:i1246.

Read more...