Animal Fat, Plant Fat, and the Journey From Your Plate to Your Arteries

Every food you will ever eat comes from one of two kingdoms, and the two kingdoms package their fat completely differently. Understand that one fact, then follow a single fatty meal through your digestive tract, your lymphatic system, your bloodstream, and your liver, and the entire diet debate snaps into focus. You will also understand why someone can feel amazing on a carnivore or keto diet while their arteries quietly disagree, and why a nutrition professor once proved you can lose 27 pounds and drop your LDL cholesterol 20 percent eating almost nothing but Twinkies.

This post walks through all of it, from basic chemistry up through the largest studies ever run and the conclusions of every major health body on Earth, in plain language, with every claim referenced at the end.

Two kingdoms, two kinds of fat

Do all animal foods contain saturated fat?

Start with an uncomfortable fact for anyone selling animal foods as health foods: there is essentially no such thing as a saturated-fat-free whole animal food. Beef, pork, chicken, fish, eggs, milk, cheese, butter, every one of them contains saturated fat, because saturated fat is a structural part of animal cell membranes and animal energy storage. The amounts range from meaningful to massive (isolated fractions like egg white or skim milk are the rare near-zero exceptions), and every whole animal food also comes packaged with cholesterol, which exists in no plant food in more than trace amounts, and with exactly zero fiber. In the American diet, the leading sources of saturated fat are overwhelmingly animal-based: cheese, butter, and meat-centered mixed dishes like burgers, pizza, tacos, and sandwiches.

Which saturated fats raise cholesterol the most?

It gets more specific than that, because “saturated fat” is not one molecule, it is a family, and the members behave differently. The World Health Organization’s commissioned analysis of 84 controlled trials estimated the effect of each one on LDL cholesterol, the particle that drives artery disease: lauric, myristic, and palmitic acid all raise it, with myristic and palmitic the strongest, while stearic acid showed no significant effect. Now look at where those fatty acids live. In meat, palmitic acid is the predominant saturated fatty acid. Dairy fat is about 27 percent palmitic acid, 12 percent stearic, 9 percent myristic, and 3 percent lauric. So the saturated fat in animal foods is led by palmitic acid, one of the two strongest LDL-raisers in the family, with dairy adding a meaningful share of myristic acid, the other. That is not an accident of one food; it is the signature of the animal kingdom.

Are any plant oils high in saturated fat?

The plant kingdom’s signature is the opposite. The fats in nuts, seeds, olives, avocados, soybeans, and grains are overwhelmingly unsaturated. The plant world has exactly three famous saturated exceptions, and their different fatty acid profiles prove the science is following molecules, not ideology. Coconut oil is loaded with lauric and myristic acid, and in a meta-analysis of clinical trials it raised LDL cholesterol significantly compared with other plant oils, so health authorities treat it like an animal fat. Because coconut oil gets defended so often, it has its own deep dive, Coconut Oil vs. Animal Fat, which ranks it head to head against butter, tallow, and lard. Palm oil is rich in palmitic acid and gets the same treatment. Cocoa butter is rich in stearic acid, the one that showed no significant effect on LDL, which is why dark chocolate gets a partial pass. Same rules for every food, plant or animal: the fatty acids decide.

Where do essential fatty acids actually come from?

One more asymmetry between the kingdoms, and it might be the most under-appreciated fact in all of nutrition. The two fats human beings actually require, the essential fatty acids linoleic acid (omega-6) and alpha-linolenic acid (omega-3), are made by plants. No animal, including us, can build them from scratch. Strictly speaking these two are the only truly essential fatty acids, because the body can make everything else in the omega-3 and omega-6 families from them. The long-chain omega-3s in fish, EPA and DHA, are originally manufactured by algae; fish are just the middlemen who ate them (whether you need to supplement them at all is its own question, worked through in Do Vegans Need DHA and EPA?). So when it comes to fat, animal foods deliver large doses of the fatty acids most strongly linked to disease, while the fats we cannot live without originate in the plant kingdom. Animals never make essential fats. They only pass along what plants made first.

Follow the fat: from your fork to your bloodstream

What happens to fat after you eat it?

Now eat a fatty meal and watch what happens, because the route fat takes through your body is stranger than most people realize.

Fat is not water-soluble, so your body has to run a whole logistics operation to move it. In your small intestine, your liver and gallbladder release bile, a detergent-like fluid that breaks the fat into microscopic droplets, the way dish soap breaks up grease. Enzymes from your pancreas then snip the fat molecules apart so they can be absorbed into your intestinal cells. Inside those cells, the pieces are reassembled and packed into transport particles called chylomicrons, which are essentially cargo ships for fat.

Why does dietary fat bypass the liver at first?

Here is the strange part. Almost everything else you absorb from food goes straight to the liver for inspection. Fat does not. Chylomicrons are too big for the intestine’s blood capillaries, so they are loaded into the lymphatic system instead, drifting up through lymph vessels to a duct near your collarbone, where they pour directly into your bloodstream. For the next several hours after a fatty meal, your blood literally carries a fatty haze, which is why a blood sample drawn after a greasy meal can look milky. As these particles circulate, tissues strip out fat for fuel and storage, and the leftovers, the chylomicron remnants, are finally delivered to the liver.

The liver is the traffic control center for all of it. It repackages fat and cholesterol into particles called VLDL, ships them out, and as those particles unload their cargo in the body they shrink down into LDL, the famous “bad cholesterol” particle. The liver also runs the cleanup side: its surface is studded with LDL receptors, molecular hands that grab LDL out of the blood and recycle it. Michael Brown and Joseph Goldstein won the 1985 Nobel Prize for working out this receptor system.

How does saturated fat raise LDL cholesterol?

And this is exactly where saturated fat does its damage. A diet heavy in saturated fat turns down the liver’s LDL receptor activity, so fewer hands are grabbing LDL out of your blood, and there is evidence the liver ships out more cholesterol-carrying particles as well. This is not just theory; it has been measured in people, where cutting saturated fat was associated with higher LDL receptor levels on blood cells, a measure that reflects the liver’s own receptor status. Fewer receptors clearing plus steady or greater output equals LDL rising and lingering, which is why the relationship between saturated fat intake and LDL is so consistent that researchers wrote it down as predictive equations back in 1965, equations that still work today.

LDL that lingers is LDL that ends up in artery walls, where it gets trapped, chemically modified, and swallowed by immune cells that turn into the foam cells of atherosclerotic plaque. A consensus statement from dozens of the world’s leading lipid scientists concluded that LDL is not merely associated with heart disease, it is the causal driver of it. So the chain from plate to plaque is complete, and every link is measured: animal-dominant fatty acids in, LDL receptors down, LDL up, plaque built. The trapping step itself, how an LDL particle gets caught in the artery wall, oxidizes, and hardens into plaque, has its own companion piece: Cholesterol, From the Ground Up.

What do unsaturated fats do instead?

The mirror image, and this is the part that usually goes unsaid. If palmitic and myristic acid turn the receptors down, unsaturated fats do the opposite: they increase the number of LDL receptors on the liver and speed up how fast LDL is cleared from circulation. More hands, working faster. This holds for both monounsaturated and polyunsaturated fats, and it is why the swap works rather than merely avoiding harm. Removing saturated fat stops the suppression; putting unsaturated fat in its place actively restores the clearance machinery.

Olive oil is the cleanest illustration. Its signature fat is oleic acid, the monounsaturated fat of olives, avocados, and most nuts, about as far from the palmitic-heavy profile of butter and tallow as a common cooking fat gets. Olive oil also carries polyphenols, chiefly hydroxytyrosol and its relatives, and those are not a marketing flourish. The European Food Safety Authority reviewed the trial evidence and approved a formal health claim that olive oil polyphenols contribute to protecting blood lipids from oxidative stress, restricted to oils containing at least 5 milligrams of hydroxytyrosol and its derivatives per 20 grams. That is a regulator concluding, on the evidence, that these compounds help keep circulating lipoproteins from being oxidized into their more dangerous form. It is also why extra virgin matters, since refining strips most of the polyphenols out.

One honest ordering, because this post is not in the business of ranking one plant fat above another for its own sake. On LDL specifically, polyunsaturated fat outperforms monounsaturated fat. In controlled feeding trials, swapping saturated fat for polyunsaturated fat lowers LDL more than swapping it for monounsaturated fat does, and in the cohort data the coronary risk reduction is larger as well. Olive oil is very good. The seed oils that get demonized are, on this one measure, slightly better. Both sit in a different universe from butter and tallow, which is the comparison that actually matters. The full case for the seed oils specifically, walked through every tier of evidence, is in Saturated Fat, Seed Oils, and Heart Disease.

One boundary worth drawing clearly, since it is the question many whole-food plant-based eaters actually care about. Everything above compares plant fats against animal fats, and on that comparison the answer is settled. It does not tell you whether adding extracted oil to an already whole-food plant-based diet beats getting the same fats from whole olives, nuts, seeds, and avocados. A 2024 randomized crossover trial came closest, putting forty higher-risk adults on a whole-food plant-based diet with either four tablespoons of olive oil daily or almost none: both versions lowered LDL, ApoB, glucose, and C-reactive protein significantly, with a nonsignificant trend favoring the low-oil version. Four weeks and forty people is not a hard-endpoint answer, and the trial nobody has run, metabolically healthy people eating whole plant foods, randomized to oil or no oil, followed for years, is the one that would settle it. Whole foods first is the defensible default; whether the oil on top helps or hurts a person already eating that way is genuinely unresolved.

The other victim: your blood sugar

Does saturated fat cause insulin resistance?

Heart disease is not the only thing on the line. Saturated fat also interferes with how your body handles blood sugar, and here it is worth being precise, because the popular image of fat “clogging” insulin receptors is not quite how it works. The reality is worse, better documented, and has a surprisingly old paper trail. In the early 1930s, the physician Harold Himsworth showed something that ran against every intuition of his era: feeding a diet rich in fat worsened tolerance to sugar and blunted the action of insulin on blood sugar, while diets richer in carbohydrate improved it. He demonstrated it first in healthy men and then confirmed it in animal experiments. Carbohydrate tolerance was being degraded by fat, not by carbohydrate, and it took another sixty years of biochemistry to explain exactly how.

Insulin is the hormone that tells your muscle and liver cells to pull sugar out of the blood, and it does this by ordering a transporter called GLUT4 to move to the cell surface, where it acts as the doorway sugar walks through. When excess fat, particularly saturated fat, accumulates inside those cells, it does not sit there inertly. It gets converted into a specific set of bioactive intermediates, chiefly diacylglycerol, ceramide, and long-chain fatty acyl-CoA, and those intermediates activate a family of enzymes called protein kinase C. PKC then attaches the wrong kind of phosphate tag, a serine phosphate where insulin’s own signal calls for a tyrosine phosphate, onto IRS-1, a key relay switch in the insulin signaling chain. The relay jams, GLUT4 never gets the order to move to the surface, and the doorway stays shut. Insulin is knocking; the lock has been sabotaged from the inside by fat. That is insulin resistance, the engine of type 2 diabetes. Scientists can induce it on demand: infuse fat into healthy volunteers and their insulin sensitivity drops within hours, and the intracellular mechanics of how those fat intermediates jam the signal have been worked out in detail.

Does it matter where your body stores fat?

More than the total does, which is the part most people have backwards. When researchers imaged cross-sections of the thigh, they could separate the fat into distinct compartments: the familiar layer under the skin, a thinner layer beneath the muscle fascia, and fat infiltrating the muscle groups themselves. The compartments behaved completely differently. The subcutaneous layer, by far the largest, did not predict insulin resistance at all. The two small ones, the fat under the fascia and the fat threaded through the muscle, did. Two thighs can hold the same total fat and carry very different metabolic risk depending on where it sits.

That distinction matters here because diet appears to move those compartments differently, and it has been tested directly. In a six-month randomized trial, 74 people with type 2 diabetes ate either a vegetarian diet or a conventional diabetic diet, both restricted to the same calories. Subcutaneous fat fell about equally on both. Subfascial fat fell significantly only on the plant-based diet. Intramuscular fat trended further down on the plant-based diet as well, though that particular difference did not reach statistical significance. The plant-based group also lost more total leg fat overall. Same calories, different compartments emptied.

The reverse experiment is faster than you would expect. When healthy men ate a high-fat diet for a single week, the fat stored inside their muscle cells rose by 54 percent, while their body weight did not change at all. A week of eating differently, no visible change on the scale, and the compartment that actually drives insulin resistance had filled by half again.

Does the type of fat matter for diabetes risk?

The type of fat matters, and this has been tested in randomized trials:

In the multicenter KANWU trial, a saturated-fat diet impaired insulin sensitivity while a monounsaturated-fat diet did not, though the benefit disappeared in people eating more than about 37 percent of calories from fat.
In a striking overfeeding experiment, volunteers ate extra muffins made with either palm oil (saturated) or sunflower oil (polyunsaturated). Same calories, same weight gain, completely different destinations: the saturated fat went disproportionately to liver and visceral fat, the danger zones for metabolic disease, while the polyunsaturated fat went more toward lean tissue. The trial did not measure insulin sensitivity itself, so it shows where the fat went rather than a measured change in insulin resistance.
Vegans have been found to carry significantly less fat inside the soleus muscle than matched omnivores of similar body size, though not in two other leg muscles measured. That study is worth reporting honestly: insulin sensitivity itself did not differ between the groups, which is unsurprising given that the vegans and omnivores were matched on age, sex, and body mass index and both groups were lean and metabolically healthy, leaving little room for a difference to show. What the vegans did show was significantly better beta-cell function, the other half of the type 2 diabetes story, alongside lower triglycerides and fasting glucose.
In a 16-week randomized trial, adults assigned to a low-fat vegan diet drained measurable amounts of fat out of their liver and muscle cells and became significantly more insulin sensitive.

Is red meat linked to type 2 diabetes?

Blood biomarker studies point the same direction: the even-chain saturated fatty acids characteristic of a meat-and-butter dietary pattern predict future diabetes, while markers tied to plant-oriented patterns do not. In the epidemiology, red meat, and especially processed meat, is consistently linked with higher diabetes risk. It is worth being precise here, because the distinction is real and the other side knows it: the large 2010 meta-analysis found that processed meat carried the clear risk, raising diabetes risk on the order of 19 percent per daily serving, while unprocessed red meat came out roughly neutral in that particular analysis. Later cohort work found an association for unprocessed red meat as well, but the strongest and most consistent signal belongs to processed meat. That nuance does not rescue the animal-food case; it just tells you where the sharpest edge is.

Do vegans have lower rates of diabetes?

Plant-predominant eating patterns are associated with roughly a quarter lower risk of developing type 2 diabetes. Among the Adventists, one of the most studied populations in nutrition science, vegans had roughly half the odds of diabetes of their meat-eating neighbors. And in a head-to-head randomized trial in people who already had type 2 diabetes, a low-fat vegan diet outperformed the conventional diabetes diet of its era on blood sugar control and cholesterol.

Pathways that only animal foods can trigger

What is TMAO and does it cause heart disease?

Beyond the fat itself, animal foods carry passengers that plants simply do not.

Carnitine and choline, concentrated in red meat and eggs, are converted by gut bacteria into a compound called TMAO, which promoted atherosclerosis in animal models and predicted heart attacks, strokes, and death in thousands of patients. Whether TMAO is a driver of the disease or mainly a marker that travels with it is still debated, since genetic studies have not cleanly confirmed causation, but the association is consistent, and notably people eating plant-based diets barely produce TMAO even when fed carnitine, because their gut bacteria are different.

Is heme iron bad for you?

Heme iron, the highly absorbable iron form found only in animal tissue, is associated in meta-analyses with higher diabetes risk.

Is processed meat really classified with tobacco?

In 2015, the World Health Organization’s cancer agency, after reviewing more than 800 studies, classified processed meat as a Group 1 carcinogen, the same certainty category as tobacco, and red meat as a probable carcinogen, driven largely by colorectal cancer data. This means the evidence that it causes cancer is strong, not that a hot dog equals a cigarette in magnitude of risk. Dietary cholesterol, found essentially only in animal foods, adds its own modest push on blood cholesterol on top of the saturated fat it travels with.

None of these pathways exist in a bowl of lentils.

The other half of the asymmetry: what plants carry that animals do not

How many more antioxidants do plants have than animal foods?

The two kingdoms differ not just in the harmful passengers animal foods carry, but in the protective ones they lack. When researchers assayed the antioxidant content of more than 3,100 foods for a landmark 2010 analysis, plant foods averaged on the order of 64 times more antioxidants than animal foods. That headline number is a mean, lifted by antioxidant-dense plants like herbs, spices, and berries, but the gap holds even when you compare ordinary foods. Look at the 75th percentile of each group and plant foods come in around 4.1 units per 100 grams against 0.21 for animal foods, still a roughly twentyfold difference with the outliers stripped out, and the median plant food carries nearly nine times the antioxidants of the median animal food. Most animal foods sit near zero. This is a laboratory measure of how much antioxidant capacity a food contains, not a direct measure of what happens inside the body, but it captures a real and enormous difference in what the two kingdoms deliver. Fiber follows the same pattern. Plants are the sole dietary source of it, and animal foods contain none.

Why does the antioxidant gap exist in the first place?

The measurement is one thing. The reason behind it is more interesting, and it comes down to who manufactures these compounds and who merely borrows them.

Plants build their antioxidants themselves. Carotenoids, polyphenols, anthocyanins, and tocopherols are not incidental to a plant, they are its survival kit. A plant sits in full sun running photosynthesis, a process that continuously generates reactive oxygen species, and it cannot walk away from drought, ultraviolet light, or heat. So it synthesizes its own chemical fire suppression and stores it in leaves, skins, and seeds. The color of a tomato, a blueberry, or a carrot is that machinery made visible.

Animals synthesize essentially none of it. No animal can build a carotenoid from scratch, a limitation so consistent that the handful of known exceptions, aphids, whiteflies, and spider mites among them, exist only because they picked up carotenoid-biosynthesis genes by horizontal transfer from fungi or endosymbiotic bacteria millions of years ago. Whatever antioxidants an animal has, it ate. The visible proof is that the color is always borrowed: the pink of salmon is astaxanthin from krill and algae, the yellow of an egg yolk is lutein and zeaxanthin from the hen’s feed. Take those out of the diet and the color goes away.

So why does so little of it survive into meat? Not for lack of raw material, which is the part most people assume. Cattle graze on forage genuinely rich in this chemistry, which is exactly why grass-fed butter is yellower and higher in vitamin E than grain-fed. Feed corn carries ferulic acid along with lutein and zeaxanthin. Soybeans carry isoflavones, saponins, and phenolic acids. Real antioxidant density goes in.

Very little comes out the other end, and several things happen to it along the way. Rumen microbes degrade much of the polyphenol load before it is ever absorbed. What does get absorbed is largely conjugated by the liver and excreted, and reviews of this literature conclude plainly that tissue polyphenol content is not proportional to dietary intake. The fat-soluble compounds that do survive get routed preferentially into liver, body fat, and milk fat rather than into skeletal muscle, which is the tissue you actually eat. And underneath all of it, the animal is alive, running its own metabolism every hour, generating reactive oxygen species and spending antioxidant capacity to neutralize them.

Two points of precision are worth making, because they are the ones a careful critic reaches for. First, cattle are not bad at storing carotenoids in general. They are the best of the common ruminants at it, which is why beef fat is yellower than the near-white fat of sheep and goats. The failure is specific: skeletal muscle is a poor depot, and cattle cannot deposit these pigments into visible muscle the way poultry deposit them into skin and yolk. Second, we are not aware of any study that has directly assayed ferulic acid or isoflavone residues in beef muscle, so the honest claim is trace amounts by inference, not a measured zero. The inference is strong, resting on rumen destruction plus conjugation and excretion plus very low feed-to-tissue transfer rates, but it is inference.

The result on the plate is what matters, and it completes the package-deal argument this post has been building. A whole plant food arrives carrying both halves of the bargain: the fuel, and the compounds that manage the byproducts of burning it. An animal food arrives fuel-dense and redox-poor.

Do antioxidant-rich diets prevent chronic disease?

This matters because the same antioxidant and anti-inflammatory compounds that ride inside plant foods are the ones tied to lower rates of exactly the diseases that dominate modern mortality. Dietary patterns that score as anti-inflammatory are associated with lower cardiovascular disease and mortality, and eating patterns built on whole plant foods are a central recommendation for cancer prevention. The brain benefits too, though here the evidence is mostly observational and worth stating honestly: plant-rich, anti-inflammatory patterns track with slower cognitive decline in population studies, but the one major randomized trial of the brain-focused MIND diet found no significant advantage over a calorie-matched comparison diet at three years, so this is a promising association rather than a settled effect. Meanwhile, patterns heavy in processed and high-heat-cooked animal products score as pro-inflammatory. Some of these plant compounds also do concrete mechanical work on the cardiovascular system, where fat-soluble antioxidants like carotenoids and vitamin E travel inside LDL particles and may help slow the oxidation that turns those particles toxic in the artery wall. That machinery is worked through step by step, along with the important caveat that isolated antioxidant supplements have failed to reproduce the benefit of whole foods, in the companion piece Cholesterol, From the Ground Up.

Do you need to eat fat to absorb antioxidants?

Here is where the two threads of this post meet. Many of the most valuable plant compounds, the carotenoids like beta-carotene, lutein, and lycopene, plus vitamin E, vitamin K, and provitamin A, are fat-soluble and locked inside plant cells, essentially insoluble in water. To get them out and into you, your body uses the exact convoy described earlier: dietary fat triggers the gallbladder to release bile, the fat and these compounds fold together into mixed micelles, the intestine pulls them in using some of the same transporters it uses for cholesterol, and they ride into the body inside chylomicrons, the same cargo ships that carried the fat. No fat in the meal, no ride.

The effect is not subtle. In a controlled crossover trial, salads eaten with fat-free dressing produced almost no measurable carotenoid absorption, while full-fat dressing produced a large rise that kept climbing as more fat was added. Adding avocado to a salad multiplied beta-carotene absorption more than fifteenfold and lutein about fivefold, with whole avocado working as well as extracted avocado oil. How much fat you need depends on which compound you are after: a small amount, on the order of three to five grams, sharply increases absorption of the provitamin-A carotenes, but the full carotenoid and fat-soluble-vitamin mix keeps climbing well beyond that. In a dose-response trial that tested salad dressings containing zero, two, four, eight, and thirty-two grams of soybean oil, maximal absorption from a large salad required more than eight grams and up to thirty-two grams or more. Two guardrails: this applies only to the fat-soluble compounds, not to water-soluble ones like vitamin C or most polyphenols, and individual absorption varies widely.

That is the plant kingdom’s answer to this post’s whole thesis. Plant fat is not merely less harmful than animal fat. It is the delivery vehicle for the antioxidants that only plants supply, which is why a drizzle of tahini, a handful of nuts, or a few slices of avocado alongside your vegetables is doing real work.

Does eating plant fat protect your bones too?

Vitamin K rides that same convoy, and once it arrives it does real work: your body uses it to carboxylate osteocalcin, a protein embedded in bone matrix, one of the steps bone needs to mineralize properly. It is tempting to stop right there and call plant fat a bone protector, and the marketing behind vitamin K2 supplements does exactly that, claiming K2 steers calcium into bone and away from arteries. The trials do not back that up. The two largest randomized trials of K2 for arterial calcification, run in Denmark and the Netherlands, both failed on their primary endpoint, and the K2-for-bone trials are small and mixed, with the one positive result funded by a K2 manufacturer. Vitamin K1 from leafy greens is easy to get and adequate on its own; K2 is optional, not required, and no honest reading of the evidence lets it license a bigger calcium supplement.

The stronger, better-supported bone connection running through this post’s own logic is a different plant compound entirely, and it is an amino acid rather than a fat-soluble vitamin: lysine, which along with leucine is one of the two indispensable amino acids vegans are most likely to fall short on once digestibility is accounted for, even though most vegans in that analysis met their overall protein requirement. Legumes are where lysine concentrates, and in the Adventist Health Study-2, people eating legumes once daily or more had a 64 percent lower rate of hip fracture than infrequent eaters, with daily eaters of meat analogues, which are also legume-based, showing a 49 percent lower rate. That does not prove lysine is the mechanism, but the chain, low legume intake to low lysine to worse bone outcomes, is coherent enough that eating legumes daily rather than occasionally looks like more than a general protein habit. And on the question hip-fracture headlines love to raise, whether animal protein is simply better for bone than plant protein, a meta-analysis of seven randomized trials found the bone density difference between soy protein and animal protein was close to zero, a result funded in part by the beef, egg, and dairy industries, none of whom had a reason to publish it if it were not true. The fuller picture, including why vegan cohorts still show somewhat higher hip fracture rates overall and exactly what fixes that, is in do vegans have weaker bones.

Which basket calms a body?

Put the two kingdoms in two baskets and the question mostly answers itself. One basket holds every whole plant food: all of the dietary fiber in existence, the overwhelming share of the antioxidants, the origin point of both essential fatty acids, and none of the specifically inflammatory passengers. The other holds every animal food: no fiber at all in meat, fish, eggs, or dairy, antioxidant values clustered near zero, and a cargo of heme iron, cholesterol, the fatty acids that suppress LDL receptors, the carnitine and choline that feed TMAO production, and whatever forms when any of it meets a hot pan. A single favorable item, oily fish being the usual nominee, cannot lift a basket that also contains bacon, cheese, and beef.

Two honest qualifications keep this precise. This is a claim about the two groups, not a rule that every plant beats every animal food on every measure, and this post has already treated coconut oil as an animal fat for exactly that reason, because the science follows molecules rather than slogans. And the standard assay behind those antioxidant numbers measures ferric-reducing capacity, which does not capture sulfur-based antioxidants like glutathione. That limitation is worth naming, and it matters less than it first appears: glutathione is not an essential nutrient, your cells synthesize it on demand from amino acids, and dietary glutathione is largely broken apart in the gut before it is ever absorbed (although a small fraction may be absorbed intact). The compounds the assay does measure, the carotenoids, tocopherols, and polyphenols, are precisely the ones no animal can manufacture and every animal must eat. So the test undercounts a category we make for ourselves while measuring the category we can only obtain from plants. The gap is real, large, and consistent across every way of slicing the data.

Climbing the hierarchy: what happens to actual people

Mechanisms are the bottom of the evidence pyramid. The question that matters is what happens to human beings over years and decades, so let’s climb.

What do decades-long studies say about eating meat?

In over half a million Americans followed by the NIH-AARP study, higher red and processed meat intake meant higher total, cardiovascular, and cancer mortality. In the Harvard cohorts, simply substituting plant protein for animal protein was associated with lower death rates, and people who increased their red meat intake over an eight-year span had about a 10 percent higher death rate in the years that followed compared with people who did not. Plant-based diet scoring across more than 200,000 people found that diets built on whole plant foods were associated with roughly 25 percent lower coronary disease, while, crucially, “plant-based” diets built on refined grains, potatoes, and sugar were associated with higher risk. Hold that finding; it comes back later. Among vegetarians and vegans specifically, the Adventist Health Study-2 found lower mortality than in comparable omnivores, the EPIC-Oxford cohort found 22 percent lower ischemic heart disease (alongside a modestly higher stroke rate in that one cohort, a finding worth watching), and a meta-analysis across cohorts found vegetarians had about 25 percent lower heart disease incidence and death.

Is butter bad for you?

On the specific animal fat that keeps getting rehabilitated, the largest and longest study yet, 221,000 adults followed for up to three decades, found that the highest butter eaters had about 15 percent higher total mortality, while swapping a small daily amount of butter for plant oils was associated with roughly 17 percent lower total and cancer mortality.

Can a plant-based diet reverse heart disease?

Dean Ornish’s Lifestyle Heart Trial randomized heart patients to a very-low-fat plant-based diet plus lifestyle change and, using angiograms, showed their artery blockages regressing while the control group’s worsened, with the gap widening over five years. David Jenkins’ Portfolio diet trial showed a plant-based eating pattern lowering LDL within a hair of a starting statin dose, head to head, in the same study. Caldwell Esselstyn’s long-running series of heart patients on whole-food plant-based diets reported strikingly few cardiac events among the adherent, though as a case series without a control group it sits lower on the hierarchy and should be cited as supportive, not definitive.

What did the vegan identical twin study find?

In the single most elegant genetic control ever run in diet research, Stanford scientists randomized 22 pairs of identical twins, one twin vegan, one omnivore, for eight weeks: the vegan twins ended with significantly lower LDL cholesterol, lower fasting insulin, and more weight loss than their genetically identical siblings. Same genes, different kingdoms, different results.

How do we know LDL actually causes heart disease?

People born with gene variants that raise LDL get heart disease decades early; people born with LDL-lowering variants are protected, and a consensus panel reviewing over 200 studies covering more than 2 million participants concluded that LDL causes cardiovascular disease in a dose-dependent, cumulative way, with no threshold below which lowering it stops helping. Statin trials in 170,000 people, then a completely different drug (ezetimibe), then a third unrelated drug class (PCSK9 inhibitors) all reduced heart attacks in proportion to how much they lowered LDL. And in 2023, bempedoic acid, a fourth mechanism that throttles cholesterol synthesis upstream of where statins act, at ATP-citrate lyase rather than HMG-CoA reductase, cut cardiovascular events in nearly 14,000 statin-intolerant patients in the CLEAR Outcomes trial. Four unrelated drug mechanisms, one common factor. Diet reaches the same particle by a fifth route, and it does not need a pharmacy: pooling 30 randomized trials, moving people from an omnivorous diet to a vegetarian or vegan one lowered LDL cholesterol by about 12 points and apolipoprotein B, the particle count that predicts risk most directly, by roughly 13 mg/dL. This matters for diet because it closes every escape hatch: if LDL causes heart disease no matter what moves it, then a diet that raises your LDL is raising your risk, whether that diet is butter and beef or anything else, and the Cochrane Collaboration’s review of the dietary trials found exactly that dose-response, with bigger cholesterol reductions producing bigger drops in cardiovascular events.

The fad diet illusion: why feeling great proves nothing

Why do people feel better on keto and carnivore diets?

Now we can finally explain the most persuasive trick in the diet world, the one that fuels every Atkins, paleo, keto, and carnivore testimonial: “I feel amazing and my numbers improved.”

What was the Twinkie diet experiment?

In 2010, Mark Haub, a professor of human nutrition at Kansas State University, ran an experiment on himself. For ten weeks he ate a diet composed mostly of Twinkies, Little Debbie snack cakes, Doritos, sugary cereal, and Oreos, one snack every three hours, capped at about 1,800 calories a day. He lost 27 pounds. His LDL cholesterol dropped 20 percent, his HDL rose 20 percent, his triglycerides fell 39 percent, and his total cholesterol went from 214 to 184. By the logic of every low-carb transformation post you have ever seen, Twinkies are a health food.

They are not, and Haub, to his credit, said so himself. What his stunt demonstrated is the single most important confounder in all of popular nutrition: weight loss improves almost every biomarker no matter how you achieve it. Blood pressure, triglycerides, blood sugar, inflammation markers, even LDL often improve during any calorie deficit, because shedding excess fat unburdens the whole metabolic system. Cut out an entire food category, any category, and most people eat less by default, lose weight, and watch their labs improve. That is what keto, carnivore, paleo, and the Twinkie diet all have in common. The short-term glow is the weight loss talking, not the food.

The weight loss itself can mislead too, not just the biomarkers it drags along. In the single most tightly controlled comparison available, twenty adults lived on an NIH metabolic ward and ate two diets in random order, two weeks each, with unlimited food and no calorie counting: a minimally processed plant-based low-fat diet and a minimally processed animal-based ketogenic diet. Left alone to eat as much as they wanted, people spontaneously ate 689 fewer calories a day on the plant-based diet, with no difference in hunger or how much they enjoyed the food. The scale technically favored the ketogenic diet, which produced slightly more total weight loss, but what came off was different: fat-free mass, muscle and water included, dropped significantly on the ketogenic diet, while only the plant-based diet produced a significant loss of body fat specifically. Losing weight and losing fat are not the same measurement, and a diet that wins on the scale can lose on body fat, the number the scale was supposed to be tracking.

Can you feel high cholesterol or artery plaque?

Feeling better is even less informative than the labs, because the things that kill you are precisely the things you cannot feel. Nobody can feel their LDL level. Nobody can feel plaque accumulating in a coronary artery. For a large share of people, the very first symptom of heart disease is the heart attack itself. Framingham-derived data indicate that most people who suffer an acute heart attack had no previously recognized cardiovascular condition, and clinical reviews commonly note that in a majority of patients the first manifestation of coronary artery disease is either a heart attack or sudden death. Energy, mental clarity, and reduced bloating are real experiences, and they are also fully compatible with arteries getting worse, which is why “listen to your body” is useless advice for a disease your body cannot report on.

What did the KETO-CTA study actually find?

So the real question for any eating pattern is not how you feel at week eight. It is what happens to the things weight loss cannot be counted on to fix and you cannot feel. And here the animal-heavy fad diets have a serious problem. Very-low-carb ketogenic diets, in a meta-analysis of randomized trials, delivered weight loss and raised LDL cholesterol at the same time. In lean, athletic people the LDL rise can be extreme, a phenomenon the keto community itself named the “lean mass hyper-responder.” Their community’s answer was to fund a study, the KETO-CTA trial, to prove those sky-high LDL levels were harmless. It did not go well.

The first published paper led with a secondary measure and a headline claiming plaque bears no relationship to ApoB, while burying its own pre-registered endpoint, and outside researchers flagged the switch within weeks. That paper was retracted in 2026, at the authors’ own request, after the automated plaque measurements behind it were judged unreliable. The detail worth knowing: that analysis was produced by an outside imaging company, and one of the paper’s own coauthors served as that company’s chief medical officer.

When the team re-ran the imaging with a different method and re-reported the pre-registered primary endpoint, the soft, rupture-prone noncalcified plaque in these lean, metabolically “healthy” dieters had still increased over a single year, by a median of 5.6 cubic millimeters, a 37 percent relative rise. In fairness to the researchers, they characterized that progression as modest and uneven: about 15 percent of participants showed some regression, and most of the group still sat below the conventional high-risk threshold for total plaque burden. The re-report is also a preprint, not yet peer-reviewed, with no comparison group, so it shows plaque progressing on the diet rather than proving the diet caused it. But hold the framing in mind. This was a study designed and funded by the low-carb community to demonstrate that extreme LDL elevation was harmless in lean, fit people, and the plaque went up, not down. There is still no credible published evidence that diet-induced LDL elevation is safe, and the flagship attempt to produce some did not find it. Tellingly, researchers from that same low-carb circle then published a study using a cholesterol-absorption drug to bring those sky-high numbers back down, a story told in full in Cholesterol, From the Ground Up.

Are low-carb diets safe long term?

The long-range data seals it. When researchers followed low-carb dieters for decades and separated them by food source, a low-carbohydrate pattern built on animal protein and fat was associated with 23 percent higher all-cause mortality, while the identical macronutrient pattern built on plant protein and fat was associated with 20 percent lower all-cause mortality and 23 percent lower cardiovascular mortality, in the same cohorts. A separate analysis of more than 400,000 people found the same split at a similar size, though it asked the question a different way. Rather than scoring how low-carb people were, it modeled what happened when carbohydrate was swapped out: exchanging it for animal fat or protein raised mortality about 18 percent, exchanging it for plant fat or protein lowered it about 18 percent. Two different designs, two different questions, and the kingdom of the replacement decided the answer in both. Same macros, opposite kingdoms, opposite outcomes. The carbohydrate was never the variable that mattered most; the source was. The split shows up in arteries directly, not just in death certificates: in adults followed from young adulthood into middle age, a higher low-carbohydrate diet score tracked with more progression of coronary artery calcium, the calcified plaque visible on a CT scan, and that progression was driven by the animal-based version of the pattern rather than the plant-based one.

Is there any real research on the carnivore diet?

As for carnivore specifically, the published evidence is a handful of small, uncontrolled reports, the largest a self-selected online survey with no measured clinical outcomes in which even the self-reported cholesterol ran high. There is not one outcome trial. Not one prospective cohort. Nothing that could actually detect harm.

Compare that ledger with the plant-based one: angiographic regression trials, a cholesterol trial that came within a hair of a starting statin dose, an identical-twin randomized trial, multiple cohorts running 20 to 30 years, and consistent meta-analyses. One side has testimonials and a retracted paper. The other side has the pyramid.

What “well planned” actually means

Is a vegan diet automatically healthy?

A vegan diet is not automatically healthy; the plant-based scoring studies showed that a diet built on refined grains, potatoes, and sugar is technically plant-based and associated with higher coronary risk. “Well planned” is what does the work, and fortunately it is not complicated. Build meals on vegetables, fruits, whole grains, legumes, nuts, and seeds.

Do vegans need to take B12?

Take vitamin B12, full stop; B12 is made by microbes, not animals, and the dirty secret of the “natural” objection is that farmed animals are themselves routinely supplemented with B12 or cobalt because modern feed does not provide enough. Vegans simply take the supplement directly instead of filtering it through a cow. On omega-3s, the essential requirement is covered by ALA from ground flaxseed, chia, and walnuts. Beyond that, the evidence does not show that a healthy adult on a well-planned vegan diet needs to supplement the long-chain omega-3s EPA and DHA, and it does not show that a small dose hurts, so a roughly 250 mg per day algal supplement is defensible as cheap insurance rather than an established requirement, with the strongest case in pregnancy and breastfeeding and in infancy, and a weaker, more speculative case in older adults with cognitive concerns. Do Vegans Need DHA and EPA? works through the whole question. Paying normal attention to vitamin D and iodine covers the rest.

Do dietitians say vegan diets are safe?

Those few steps are the entire price of admission, and the professional consensus reflects it. The Academy of Nutrition and Dietetics, the world’s largest organization of nutrition professionals, 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. Its previous position went further, judging well-planned vegan diets appropriate for every stage of life including pregnancy, infancy, childhood, and athletics, and the 2025 update focuses on adults, the population covered by its new systematic reviews.

What the expert bodies conclude

Individual studies can mislead. That is why independent expert panels exist, to weigh the totality, and it is worth seeing how uniformly they have landed.

What does the World Health Organization say about saturated fat?

The World Health Organization issued a strong recommendation in 2023 to reduce saturated fat, a conditional recommendation to keep it below 10 percent of calories, and a strong recommendation to replace it preferably with polyunsaturated fats, along with plant-source monounsaturated fats or fiber-rich carbohydrates. The Dietary Reference Intakes from the Institute of Medicine of the National Academies, the foundational nutrient standards for the US and Canada, found a positive linear trend between saturated fat intake and both LDL cholesterol and coronary heart disease risk, set no upper limit because any incremental increase raises that risk, and recommended keeping intake as low as possible while eating a nutritionally adequate diet.

Do the 2025 Dietary Guidelines still limit saturated fat?

The independent scientists of the 2025 Dietary Guidelines Advisory Committee concluded that swapping butter for plant oils is associated with lower cardiovascular disease and death, and recommended elevating beans, peas, and lentils in the protein food rankings. The final Dietary Guidelines for Americans 2025–2030, released in January 2026 by an administration openly fond of beef tallow, still retained the 10 percent saturated fat cap. The 2015–2020 and 2020–2025 editions of the federal guidelines also included a Healthy Vegetarian Dietary Pattern among their officially endorsed eating patterns.

Is beef tallow healthy?

A 2026 clinician’s guide in JACC: Advances, written for cardiologists fielding these exact questions, sorted the popular contenders into three groups: “evidence of harm, limit or avoid,” which is where beef tallow landed because it raises LDL cholesterol; a middle tier lacking evidence of either harm or benefit, which included full-fat dairy, MCT oil, and the nonsugar sweeteners monk fruit and stevia; and “evidence of benefit,” which is where seed oils and seafood landed. The guide concluded flatly that no evidence supports beef tallow as a healthier alternative. Note what that middle tier tells you: this is not a document that condemns everything from an animal, which is exactly why its placement of tallow carries weight.

Which health organizations recommend eating more plants?

The American Heart Association, the American College of Cardiology, the American Diabetes Association, the American Cancer Society, and the pediatric cardiovascular guidelines endorsed by the American Academy of Pediatrics all converge on the same directions: less saturated fat, less red and processed meat, more plants. The American College of Lifestyle Medicine goes furthest, recommending eating patterns based predominantly on minimally processed plant foods for treating and preventing chronic disease, and Kaiser Permanente, one of the largest healthcare systems in America, published guidance urging its physicians to consider recommending plant-based diets to patients.

It is the same abroad. The European Society of Cardiology, the United Kingdom’s Scientific Advisory Committee on Nutrition, the 2023 Nordic Nutrition Recommendations covering five countries, which call for predominantly plant-based diets and cap red meat at 350 grams a week, Canada’s national food guide with its emphasis on plant protein, the World Cancer Research Fund’s global cancer prevention report, and the EAT-Lancet Commission’s planetary health diet all point in one direction, and it is not toward the butcher counter.

Every one of these bodies is independent. They span rival countries, competing medical specialties, and five decades of separate evidence reviews. For the carnivore narrative to be right, all of them would have to be wrong, together, in the same direction, while the actual data sat in plain sight. The far simpler explanation is the true one: the science has been pointing the same way the whole time.

The bottom line

Fat from the animal kingdom arrives led by the saturated fatty acids that dial down your liver’s cholesterol-clearing machinery, packaged with cholesterol, carnitine, and heme iron, containing zero fiber, and carrying only the essential fats the animal took in from plants. Fat from the plant kingdom arrives mostly unsaturated, bundled with fiber, and includes the only fats that are literally essential to human life. Follow the meal from bile to chylomicron to lymph to liver and you can watch the difference play out in your own physiology; follow the people for thirty years and you can watch it play out in their lives.

A fad diet heavy in animal foods can absolutely make you lighter, sharper, and better on a lab printout for a season, and so can a diet of Twinkies, for exactly the same reason. Weight loss flatters everything. The measures weight loss cannot be counted on to flatter, LDL, ApoB, and the plaque in your coronary arteries, are the ones where the animal-heavy diets falter and where whole-food plant eating, up to and including a well-planned vegan pattern with its B12 in hand, holds the strongest record in the entire nutrition literature: regression on angiograms, cholesterol drops approaching a starting statin dose, better cardiometabolic markers in identical twins, and lower disease and death across decades of follow-up.

What is Lp(a) and can diet lower it?

One honest asterisk belongs here, because a good argument names its own limits. There is a cholesterol particle that diet does not meaningfully move: lipoprotein(a), an LDL-like particle whose level is set almost entirely by the genes you inherit. It is a causal driver of heart attacks and aortic valve disease, roughly one in five people worldwide carries a high level, and neither eating well nor statin therapy lowers it, with some analyses finding that statins modestly raise it. Plant-forward eating remains the highest-leverage thing most people can do for the particles diet can move, which is nearly all of the risk for nearly everyone. But someone can do everything right and still carry elevated Lp(a), which is why it is worth measuring once in your life, since the number rarely changes, so you know whether you are the exception who needs help beyond the plate.

You cannot feel your arteries. But you can read the evidence, and the evidence, from the bottom of the pyramid to the top, keeps saying the same thing: eat plants.

For readers who want to explore the primary research further, NutritionFacts.org maintains a free, ad-free library of evidence reviews, and every organizational report cited below is publicly available.

Keep reading

This post is one stop on a tour through the same evidence pyramid. Each companion piece takes one thread further than this post had room to:

Cholesterol, From the Ground Up is the molecular sequel to the digestion journey above: what actually happens once an LDL particle reaches the artery wall, and why your arteries can be getting worse while you feel your best.
Our post on how plaque forms follows what happens inside the artery wall once LDL particles get there.
Saturated Fat, Seed Oils, and Heart Disease takes on the “seed oils are poison” narrative directly, and shows why replacing saturated fat with these very oils lowers heart disease at every level of evidence.
Coconut Oil vs. Animal Fat ranks the solid fats against each other and dismantles the “but it raises good cholesterol” defense of coconut oil.
Losing Weight on a Vegan Diet takes on the mirror image of this post’s warning, that almost any diet which removes weight will improve the markers you can see, which is exactly why the silent ones matter.
Do Vegans Need DHA and EPA? answers the omega-3 question this post raises twice: whether the long-chain fats that originate in algae are something a healthy plant eater actually needs to supplement.

References

Last updated: September 27, 2026, 18:20 UTC.