"Former Vegan Cardiologist" Video, Fact-Checked: What Ken Berry and Jami Dulaney Got Wrong

A note on sourcing: quotations come from the video’s automatically generated transcript. Section headings paraphrase each claim rather than quote it verbatim. Nothing in the argument below depends on exact wording.

Health information, not medical advice. If anything here touches your own situation, talk to a physician or registered dietitian who understands plant-based nutrition.

The numbers she gives, and what they show

What did her labs actually say?

Dr. Ken Berry has published an interview with Dr. Jami Dulaney, a board-certified cardiologist who spent years building a practice around plant-based nutrition and now eats red meat and eggs and says she was wrong. Most of the coverage stops at the story. The interesting part is that she gives her actual lab values late in the conversation, and they do not support the case being built on them.

Fasting insulin: 4.9. That is a normal value. It is not hyperinsulinemia. Berry spends the interview calling insulin “the Achilles heel of the modern plant-based recommendations,” and the patient at the center of the story has normal fasting insulin.

A1C: 5.9. Elevated, and worth taking seriously. But look at when it was taken. Her own account is that the first step in her transition was adding fish back into her diet, and only after that did she run the labs that produced this number. In a separate video she says she had avoided testing her A1C for a long time because she did not want to know what it was. So there is no A1C on record from the plant-based years. The number being attributed to plants was measured after she had already started eating animals, and it is the only elevated value in the entire story.

Worth knowing too that endurance athletes run elevated A1C and fasting glucose as a matter of course. Studies in marathoners and endurance athletes find A1C values in exactly her range producing false positives for insulin resistance. She was running 50 miles a week.

Credit where it is due on that sequencing point: Mic the Vegan caught it first, in his video breakdown of the same interview, which also surfaced her apoB figure and the endurance-athlete A1C literature. Worth watching alongside this.

After going low-carbohydrate: total cholesterol from 180 to 313, LDL from 80 to 210, apoB 217. Her words: she was ready for 240 and was not ready for 313. She states the apoB figure herself and acknowledges in the same breath that under 90 is the target. She is at more than double it. She and Berry then spend several minutes explaining why apoB is not the thing that matters.

So the intervention fixed nothing that was measurably broken on a plant-based diet, because nothing was measured on a plant-based diet, and it roughly tripled the marker with the most thoroughly established causal link to heart disease in her own specialty.

What was she actually eating?

This is the part that decides the whole question, and it is all in her own words.

She struggled to hit 50 grams of protein a day while running 50 miles a week. For an endurance athlete at that volume, that is roughly a third of what she needed.
She was eating 75 to 80 percent carbohydrate, and she lists the sources: white rice, sourdough bread twice a week, and mangoes off her own trees. Elsewhere she mentions weekends on Oreos and Coca-Cola.
Her only regular supplements were B12 and vitamin D, three times a week each. No algal omega-3, despite identifying low DHA as the first thing that shook her confidence.
She names beans and potatoes as her carbohydrate problem rather than her protein solution.

That is not a well-planned vegan diet. It is a high-refined-carbohydrate, protein-deficient diet that happens to contain no animals, in a person running 50 miles a week with grandparents and both parents diabetic. Every problem she describes is a predictable consequence of that specific implementation, and every one of them has a fix that is not beef.

Why this video was predictable

What is the difference between a partial relapse and a total one?

Before the claims, the shape. In my book Veganism Rewired I spent a long section on relapse, and this follows it closely. More about the book, along with free advocacy resources, is at veganlinked.com/advocacy.

Behavior change is not a line. The psychologists who built the stage-of-change model found that people climb toward a new behavior, fall part of the way back, and climb again, usually more than once, before the change holds. Change is a spiral, and a relapse is a loop in the climb, not a fall to the bottom of it. But relapses come in two kinds. In a partial relapse the behavior breaks and the values stay intact: the person still believes what they believed and tends to come back. In a total relapse the person abandons the behavior and then turns on the reasons they once held, because rejecting the reasons is the fastest way to quiet the dissonance of having fallen away. That second kind lands a person in what I call recycled precontemplation, and it produces the most volatile critic a movement ever faces: someone who knows every argument from the inside and now aims that insider precision at the position they used to hold.

Why is relapse most likely at the beginning of a movement?

Because relapse pressure is environmental. It is highest now, when anyone adopting the change does so as a small minority in a world built for the old default: isolation, constant social friction, thin infrastructure, few others walking the same road, and an opposition with more money and reach that actively recruits anyone who wavers. Early movements everywhere endure high relapse rates, not because their members care less, but because the world offers them so little to lean on. Every abolitionist, suffragist, and anti-smoking campaign had its prominent defectors, and each defection was marketed by the other side as proof the cause was wrong.

Does relapse pressure ever go away?

It eases, and that is the hopeful part. As adoption rises and the movement’s capacity grows, with more vegans, better and cheaper alternatives, denser community, and infrastructure filling in, the world begins to carry the weight the individual anchor once carried alone. The friction of staying falls just as the friction of starting does, not through anyone’s change of heart but through a change of world. Past a certain point the movement generates a gravity of its own, and the defection that makes headlines today becomes the rare event it should have been all along. Every advocate who lowers that friction for the next person is shrinking the pool this video was drawn from.

What is the Reverse Transtheoretical Model?

Relapse is not the only backward motion. There is a second one that is chosen rather than suffered, and it runs the entire stage-of-change model in reverse. I call it the Reverse Transtheoretical Model. It begins with reverse contemplation, a settled vegan questioning the commitment as doubt builds. Then reverse preparation, where the person seeks out the influencers and narratives that would justify leaving. Then reverse action, the deliberate reintroduction of animal products, and reverse maintenance, where the new habits are locked in with routines, relationships, and justifications. Past that lies carnist reintegration, a full return to the dominant norm with veganism reframed as naive, and for a few, carnivore radicalization, where the motion overshoots into an oppositional identity.

The tell that separates a quiet relapse from that arc is simple: the person does not keep their head down. They produce the case. In this interview she describes herself as all-in by temperament, black and white, consuming a hundred books on one side and then a hundred on the other. That is the same disposition running in reverse, which is why the exit looks as total as the entry did.

Berry sits at the other pole and always did. He was never a vegan who slipped. He is what the book calls an active carnist: someone aware of the case against animal use who has rejected it and now spends content and influence producing the case for the other side. The messages this pipeline produces fall into three kinds: lines that justify animal use, lines that discourage action, and lines that attach a social cost to changing at all. That third category, vegans are extreme, vegans are unwell, is what this video is built to deliver. A former plant-based cardiologist saying plants made her sick is the most valuable possible artifact for that message, which is why the video did not need a doctor with data. It needed a doctor with a story.

One last point for anyone who watched and felt the floor tilt. Organized opposition on this scale is a response. Nobody funds a case against a practice that threatens nothing, and the effort now spent recruiting doctors to argue against plants is a measure of how much ground plants have taken. The demoralization the video produces in advocates is not a side effect; it is one of the pipeline’s products, aimed at the people best positioned to change things. Read it that way and it stops working on you.

The claims, and what the evidence says

What does the video actually argue?

Ten claims do the work. Each is stated in one line, then answered.

A paper on “bonking” shows the problem is blood glucose, so you can run on very little carbohydrate. That trial found carbohydrate improved performance, including in the low-carbohydrate-adapted athletes. It is about race fueling, not about what diet to eat.
Lean plant-based patients showed elevated insulin and A1C. Her own insulin came back at 4.9, which is normal, and her A1C was measured only after she had reintroduced fish. Every cohort and trial that has measured this properly finds plant-based diets improve insulin sensitivity and carry the lowest diabetes rates.
DHA levels were low. Probably true and probably irrelevant. Whether a lower index matters in a healthy vegan is unsettled, seven randomized trials found no cognitive benefit from raising it, and she never took the algal supplement that addresses it.
A dog developed pancreatitis, and dogs do not make amylase well. Dogs carry roughly seven times the pancreatic amylase gene copies wolves do. Pancreatitis is triggered by high-fat meals and treated with a low-fat diet.
Plant-based eating was 100 percent unsuccessful in patients over a BMI of 35. A statement about one clinic. Randomized trials in obese adults show some of the largest ad libitum weight losses in the literature.
Cholesterol was never the primary cause of coronary disease, and apoB is at the scene of the crime rather than causing it. Her own apoB is 217 against a target she names as under 90. LDL causality rests on more than 200 studies, over two million people, Mendelian randomization, and multiple unrelated drug classes.
You will “very seldom” completely reverse type 2 diabetes on a plant-based diet. DiRECT achieved 46 percent remission at one year on a diet that was roughly 59 percent carbohydrate.
Being vegan is a first step toward keto or carnivore. No cohort or trial has ever found a point at which adding animal food back to a well-planned plant-based diet improves a hard outcome.
She would now tell a parent not to raise a child vegan. Pediatric and dietetic bodies across four continents say the opposite, provided the diet is planned.
No society has ever lived to a hundred on plants, and the Blue Zones are full of holes. The Adventist cohort is a prospective study of 96,000 people, not a demographic claim, and it is the strongest longevity cohort we have.

What is the structural problem?

There is not a single hard outcome in the interview. No events, no imaging, no before-and-after comparison of her patients. What there is: a continuous glucose monitor, one normal insulin value, one elevated A1C, a fatty acid level, one dog, one race, and podcasts.

Meanwhile the video concedes on tape that patients moved from the standard American diet to plant-based eating lost weight, that many had chest pain improve, and that “we had success.” Berry adds his own concession, agreeing that moving someone from an ultra-processed diet to whole plant foods improves many chronic conditions.

Her own practice testimonials point the same direction. One describes a patient who had been type 2 diabetic for 14 years on 2,000 mg of metformin daily with a rising A1C, and who after three months in her care came off metformin entirely, dropped from 6.9 to 5.8, and lost 15 pounds. That is a documented outcome in an actual patient, achieved on the diet now being renounced, and it is a better result than anything reported in this interview.

The diet being abandoned has randomized trials, an identical-twin study, a metabolic-ward feeding study, and cohorts of hundreds of thousands. The diet being adopted has one uncontrolled online survey (Lennerz 2021). Berry, who hosts the conversation, has been debunked repeatedly and continues to make the same claims; he makes several errors in this video alone, listed below.

What are the confounders nobody isolates?

Four, and all four are in the video itself: a family history of diabetes spanning grandparents and both parents, ultra-endurance training at 50 miles a week, weekends on Oreos and Coca-Cola, and a diet that was 75 to 80 percent carbohydrate from white rice, sourdough, and mangoes while delivering under 50 grams of protein. Any one of them can move an A1C. Genetics is the strongest predictor of type 2 diabetes there is. Excessive endurance training impaired glucose tolerance in a controlled study (Flockhart 2021). Nothing in the interview separates these from “plants.”

Claim 1: the endurance paper

What does the research actually say?

The research is real. Tim Noakes has argued for years that falling blood glucose, reflecting liver glycogen rather than muscle glycogen, is what forces fatigue in prolonged exercise (Noakes 2022). In 2025 Prins, Noakes, Koutnik and colleagues published a randomized crossover trial in triathletes adapted to very-low-carbohydrate and high-carbohydrate diets (Prins 2025).

Read the title: carbohydrate ingestion eliminates hypoglycemia and improves endurance performance, in both the low-carbohydrate-adapted and the high-carbohydrate-adapted athletes. Being fat-adapted spared nobody from needing carbohydrate. The finding is about what you consume during the race, and the person it argues against is the one taking in none. A plant-based ultrarunner who swaps bolus sugar for a steady trickle of carbohydrate gets the identical benefit.

Two physiological notes. During prolonged exercise, insulin is suppressed by adrenaline and noradrenaline, so a crash deep into an ultra is liver glycogen running out, not an insulin surge. And post-race soreness is delayed-onset muscle damage from eccentric loading; there is no mechanism by which race-day glucose causes it.

Worth adding that even researchers arguing for lower carbohydrate intakes concede that carbohydrate availability is what limits endurance performance. Plants are where carbohydrate lives.

Claim 2: insulin resistance

What does an insulin of 4.9 mean?

It means normal. That is the number she reports, and it is the load-bearing fact the interview builds an argument on top of without pausing. A single elevated A1C in someone with three generations of diabetes, running 50 miles a week on 75 to 80 percent carbohydrate from white rice and sourdough, is not evidence that plants cause insulin resistance.

And the sequencing matters more than the value. She says the first step in her transition was adding fish back, and the labs came after. Elsewhere she says she had put off testing her A1C because she did not want to know the answer. So the diet being indicted was never tested, and the one measurement that supports the story was taken after the diet had already changed. In any other context a cardiologist would call that an uncontrolled before-and-after with no before.

There is also a well-documented reason an endurance athlete’s numbers run high. Marathoners and endurance athletes show elevated fasting glucose and A1C values in her range that produce false positives for insulin resistance, driven by training load, cortisol response, and red cell turnover rather than by metabolic dysfunction. A CGM reading in the 100s after a morning run is not a diagnosis, and a continuous glucose monitor is not a diagnostic tool for diabetes in the first place.

Fasting insulin is a poor tool for this in any case. Insulin assays are not standardized across laboratories, there is no agreed clinical cutoff, and within-person variability is large (Staten 2010). The cutoffs used in low-carbohydrate circles have never been validated against outcomes.

What does the evidence show when the question is asked properly?

It runs the other way, consistently. In the Adventist cohort, after adjusting for body weight, the odds of developing type 2 diabetes were 0.38 for vegans compared with meat-eaters, falling in near-perfect steps from omnivore to pescatarian to lacto-ovo vegetarian to vegan (Tonstad 2013), with prevalence of 2.9 percent in vegans against 7.6 percent in non-vegetarians (Tonstad 2009). Across three large US cohorts, a healthful plant-based diet was associated with 34 percent lower diabetes incidence (Satija 2016). A federated analysis of nearly two million adults across 20 countries found each modest daily serving of processed meat linked to 15 percent higher type 2 diabetes risk and unprocessed red meat to 10 percent higher (Li 2024).

Trials measuring insulin sensitivity directly agree. A 16-week randomized trial in overweight adults found a low-fat vegan diet reduced intramyocellular and liver fat and improved insulin sensitivity; note the caveats, since it was run by a plant-based advocacy group and the control arm made no changes, but the direction matches independently funded work (Kahleova 2020). And in the tightest-controlled comparison available, twenty adults on an NIH metabolic ward spontaneously ate 689 fewer calories a day on a plant-based low-fat diet than on an animal-based ketogenic one, with no difference in hunger (Hall 2021).

A larger insulin response to carbohydrate is not insulin resistance; it is what insulin is for. And the mechanism runs against the conclusion drawn here: saturated fat accumulates as intramyocellular lipid and jams the insulin signal, while heme iron, absorbed whether the body needs it or not, is associated with roughly 16 percent higher type 2 diabetes risk per milligram per day (Bao 2012). Red meat and eggs are the richest sources of exactly those two inputs. The mechanisms are traced in dietary fat and the anti-inflammatory diet.

The protein problem she describes and misdiagnoses

Was she eating enough protein?

No, by a wide margin, and she says so plainly. Under 50 grams a day while running 50 miles a week is roughly a third of what an athlete at that volume needs. She then describes the difficulty as intrinsic to plant eating, on the grounds that plant protein comes attached to carbohydrate.

That is the misdiagnosis. Vegans in EPIC-Oxford averaged 0.99 g/kg, about 24 percent above the RDA. Matched for total protein at roughly 1.6 g/kg, habitual vegans and omnivores built muscle and strength identically across 12 weeks of resistance training (Hevia-Larraín 2021). Soy protein isolate scores in the high-quality band on DIAAS. The National Academies RDA is 0.8 g/kg, and athletes need more, which is an argument for eating more legumes and soy, not for eating a cow.

She names beans and potatoes as the carbohydrate problem. They are the lysine solution. A person struggling for protein on plants who treats legumes as the obstacle has identified her own bottleneck and then walked away from the fix. The full treatment is in the protein post.

Claim 3: low DHA

Is low DHA in vegans even a problem?

That is the question the video never asks, and the honest answer is that it is unsettled and trending toward no. Our complete review is in Do vegans need DHA and EPA?

The conversion percentage is the wrong question. A tracer study tells you what fraction of one labeled dose of ALA ended up as DHA. Most ALA is burned for energy or stored, which is what fat is for. Tracer methods likely underestimate synthesis (Domenichiello 2015). And demand is tiny: PET imaging in living humans puts net brain DHA consumption at about 3.8 milligrams a day, with a brain half-life of roughly two and a half years (Umhau 2009), against adipose stores measured in tens of grams.

“Low vegan” is really “low non-fish-eater.” Vegans average an omega-3 index around 3.7 to 4 percent, which in the same study sat right next to omnivorous US soldiers at about 3.5 percent (Sarter 2015). That is a national biomarker, not a plant-based disease.

No blood measure is validated for the brain. The Omega-3 Index was developed as a coronary heart disease risk marker by William Harris, who founded the company that sells the test. It has never been validated as a proxy for brain DHA: in the Adventist brain sub-study, the index correlated with white matter volume but not with hippocampal volume or frontal cortical thickness, and showed no association with any of six cognitive tests (Loong 2023). The much-quoted “4.4 percent threshold for brain aging” is the 25th-percentile cutoff of one cross-sectional Framingham sample, not a validated danger line (Tan 2012).

The trials are null, including the one that closed the loophole. AREDS2 (Chew 2015), VITAL-Cog (Kang 2022), MAPT (Andrieu 2017), OPAL (Dangour 2010), the Alzheimer’s DHA trial (Quinn 2010), and Shinto 2024 all found no cognitive benefit. Then in 2026 the PreventE4 trial gave 2 grams a day of algal DHA to 365 at-risk older adults, nearly half APOE4 carriers, and made cerebrospinal-fluid DHA its primary endpoint. The CSF DHA to arachidonic acid ratio rose significantly in the treatment arm, confirming the DHA reached the central nervous system, the omega-3 index climbed from about 4.9 to about 11 percent, and there was no difference in hippocampal volume, cortical thickness, or cognition over two years (Yassine 2026). The DHA reached the brain and nothing happened. The 2025 Academy of Nutrition and Dietetics position paper says EPA and DHA can be obtained from algal supplements if desired, but that this may not be a requirement for health or disease prevention (Raj 2025). WHO’s dementia guidelines and the Alzheimer’s Association do not recommend supplementation.

And the detail that undoes the argument

She never took an algal omega-3 supplement. Her stated regimen was B12 and vitamin D, three times a week each. She identifies low DHA as the first crack in her confidence, declines the one intervention that directly addresses it on the grounds that she will not stand at a sink eating supplements, and then moves to red meat and eggs, which contain essentially none of it. She reports fixing the level by eating fish four times a week, which reversed quickly, and then continues to present the episode as an argument against plants rather than an argument for algae or fish.

Meanwhile the video mocks Michael Greger’s supplement list at length. Her grandchildren, she notes, were given DHA and multivitamin supplements while eating plant-based, and she describes them as extremely healthy, with a six-year-old who has never needed an antibiotic. The supplement she refused for herself was given to the children, and the children did fine.

Claim 4: the dog

Do dogs digest starch?

Yes, and unusually well for a canid. Dogs lack salivary amylase, which is the fragment of truth the claim rests on, but they produce pancreatic amylase and the landmark genomic study of domestication found dogs carry roughly seven times as many copies of the amylase gene AMY2B as wolves, identifying adaptation to a starch-rich diet as one of the genetic signatures of becoming a dog at all (Axelsson 2013).

And the pancreatitis?

The recognized risk factors for canine pancreatitis are high-fat meals and dietary indiscretion, obesity, high blood triglycerides, certain breeds, and some drugs (Cridge 2022). Standard veterinary management is a low-fat diet. Moving a dog with pancreatitis onto raw meat is the opposite of that advice, and raw diets carry documented Salmonella and Listeria risk to pets and to the people handling the food (FDA).

Does the analogy transfer to humans?

It backfires. Humans carry multiple copies of the salivary amylase gene AMY1, with copy number tracking the starch content of traditional diets (Perry 2007). We are more starch-adapted than dogs.

Claim 5: failure above a BMI of 35

Does plant-based eating fail people with severe obesity?

Not in the trials. In the BROAD trial in New Zealand, overweight and obese adults placed on a whole-food plant-based diet with no calorie counting lost about 12 kilograms, a BMI reduction of 4.4 points, at 12 months (Wright 2017). Across 11 randomized trials, vegan diets produced about 4.1 kilograms more weight loss than control diets (Termannsen 2022). The metabolic-ward result above is the mechanism.

The video’s own explanation of who succeeded, people who could calorically restrict and stay strict, describes a restrictive implementation, and restrictive implementations fail across the board. In DIETFITS, the largest head-to-head trial of a healthy low-fat diet against a healthy low-carbohydrate diet, weight loss and dropout were the same at 12 months (Gardner 2018). Why whole plant foods make weight loss unusually forgiving is covered in the best diet to lose weight.

Claim 6: cholesterol and apoB

What does the causal evidence rest on?

Not on hospital snapshots. The European Atherosclerosis Society consensus drew on more than 200 studies covering over two million people, including Mendelian randomization, people with PCSK9 mutations, and randomized trials of unrelated drug classes, and concluded that LDL causes atherosclerotic cardiovascular disease in a dose- and time-dependent way with no threshold (Ference 2017). Reducing saturated fat cut cardiovascular events by 17 percent across 15 randomized trials, with benefit tracking how far LDL fell (Hooper 2020).

The apoB dismissal in this interview, that it is at the scene of the crime rather than driving the train, is the part a cardiologist’s patients should read carefully. ApoB counts the particles that get retained in the artery wall. It is not a bystander; it is the vehicle.

And she gives her own number: 217, against a target she states herself as under 90. That is not a marginal reading. Familial hypercholesterolaemia is the natural experiment here, because those patients have high LDL without high triglycerides or high A1C, and they develop accelerated plaque anyway, sometimes beginning in their twenties and thirties. The same pattern appears in people carrying PCSK9 gain-of-function variants. Low triglycerides and a good HDL ratio do not neutralise a high particle count; that is precisely what the genetic evidence rules out.

The “large fluffy particles are harmless” argument raised in the interview does not rescue it either. When large and small LDL have been compared directly against cardiovascular events, both track with risk. ApoB is preferred as a marker because it counts every atherogenic particle regardless of size, which is the point.

The A1C-versus-LDL comparison offered in the video is also a false choice. Diabetes and LDL are both causal, they multiply rather than compete, and no cardiologist has to pick one. Triglycerides, HDL, and diabetes matter too; cardiology has known that for fifty years. The full chain is in cholesterol, from the ground up.

What happened to her lipids?

Total cholesterol 180 to 313. LDL 80 to 210. Her own numbers, reported in the interview, after a few months of carbohydrate restriction. Berry volunteers that his own went to a total of 350 and an LDL of 250.

For context: pooling 30 randomized trials, vegetarian and vegan diets lowered LDL and apolipoprotein B (Koch 2023). In the identical-twin trial, the vegan twin dropped LDL about 14 mg/dL more than the omnivore twin in eight weeks (Landry 2023). An umbrella review of 68 randomized trials found only four conclusions about ketogenic diets supported by high-quality evidence, and one is that they raise LDL (Patikorn 2023). And the study the low-carbohydrate world leaned on to argue that high LDL on keto does not build plaque, the KETO-CTA “lean mass hyper-responder” paper, was retracted by its journal in May 2026 after critics documented selective reporting, questionable statistics, and an undisclosed financial conflict (JACC retraction notice; Retraction Watch 2026).

In EPIC-Oxford, vegetarians and vegans had 22 percent less ischemic heart disease over 18 years (Tong 2019). That is the diet being left behind.

Claims 7 and 8: Berry’s contributions

Can a plant-based diet reverse type 2 diabetes?

More often than “very seldom.” And it is worth asking what the diet being adopted does over time: in pooled cohort analyses, low-carbohydrate patterns built on animal foods are associated with higher all-cause mortality, while low-carbohydrate patterns built on plant foods are associated with lower mortality. The macronutrient label is not what moves the outcome; the source is. Remission itself is driven by energy deficit and weight loss, particularly loss of liver and pancreatic fat, not by removing carbohydrate. The DiRECT trial produced remission in 46 percent of participants at one year and 36 percent at two, using a formula diet that was roughly 59 percent carbohydrate (Lean 2018). Barnard’s randomized trial found a low-fat vegan diet beat the standard ADA diet on A1C, a 1.23-point drop against 0.38 among people who did not change medications (Barnard 2006). Low-carbohydrate diets do show a remission advantage at six months, which largely disappears by twelve months in the same meta-analysis (Goldenberg 2021). The carnivore diet has never been tested for diabetes remission in a randomized trial at all.

Is veganism a stepping stone to carnivore?

The data run the other way. When cohorts score diets by quality rather than label, the healthful plant-based pattern is associated with lower mortality and less heart disease across cohort after cohort, and no study has ever identified a point at which adding animal food back to a well-planned plant-based diet improves a hard outcome. Our 48-study review of the whole gradient is in the nutritional case for plants over animals.

Claim 9: children

Is a well-planned plant-based diet safe for children?

The institutions say yes. The Academy of Nutrition and Dietetics’ 2025 position paper holds that appropriately planned vegetarian and vegan patterns are nutritionally adequate (Raj 2025). The American Academy of Pediatrics holds that appropriately planned vegetarian and vegan diets can be nutritionally adequate during infancy, childhood, and adolescence with attention to B12, vitamin D, iron, and calcium. The Australian Dietary Guidelines say the same, explicitly for all stages of the lifecycle.

The interview’s own evidence points the same way. Her grandchildren were raised plant-based with DHA and multivitamin supplements, and she describes them as extremely healthy, with a six-year-old who has never had an antibiotic. The reasons she now gives for advising against it are social: travel, sports teams, other children, the child asking questions about eating animals. Those are real challenges. They are not nutritional findings, and they should not be presented as if they were. Our full guide is raising children vegan.

Claim 10: nobody has lived to a hundred on plants

What does the longevity evidence actually show?

The Adventist Health Study-2 is not a demographic claim about a village; it is a prospective cohort of roughly 96,000 people with dietary data and hard endpoints. In it, vegans had a type 2 diabetes prevalence of 2.9 percent against 7.6 percent in non-vegetarians, markedly lower hypertension, 16 percent lower overall cancer incidence and 34 percent lower female-specific cancers, and all vegetarians combined had lower all-cause mortality. The vegan-specific mortality estimate trended favorably without reaching significance on its own, which is a sample-size problem rather than a contrary signal, and this site says so plainly in the Adventist Health Studies.

The Blue Zones demographic critiques are real and we have addressed them at length in the Saul Newman fact-check and are the Blue Zones debunked?. Loma Linda is the one that rests on cohort data rather than birth registries, which is exactly why it survives the critique. Dismissing the cohort by invoking the demography is a category error.

And the same standard applied evenly is fatal to the other side: there is no long-lived carnivore population, no cohort, and no randomized trial. The entire published outcome literature is one online survey (Lennerz 2021).

The biochemistry errors in passing

Four, all from the host, all checkable. He states that we convert ALA to DHA and then EPA; the pathway runs ALA to EPA to DPA to DHA. He calls the plant omega-3 “alpha linoleic acid,” which blends alpha-linolenic acid, the omega-3 in flax and walnuts, with linoleic acid, the omega-6 in sunflower and corn oil. He describes cattle as a toxin filter, which inverts how fat-soluble contaminants bioaccumulate up a food chain. And the gorilla argument offered in the interview, that gorillas are effectively ketogenic because they ferment fiber into short-chain fatty acids, is a point about fiber being valuable, not about meat.

The bottom line

If this video shook you, what should you actually do?

Eat enough protein. Legumes daily, plus soy and whole grains. If you train hard, aim higher than the RDA and spread it across meals. See the protein post.
Verify your B12. Serum B12 with methylmalonic acid, and supplement reliably. See The Great B12 Gamble.
Eat your ALA daily. A tablespoon of ground flax or chia, plus walnuts and hemp. Grind the flax.
Decide about algal DHA knowing it may not be necessary. No trial has shown a healthy vegan benefits, and the trials that raised brain DHA found nothing. Roughly 250 mg a day is cheap insurance if you want it; skipping it is equally defensible. Full review in Do vegans need DHA and EPA?
Build meals on whole foods, not refined starch. Beans, lentils, whole grains, vegetables and fruit, rather than white rice and bread with fruit for calories.
Measure what predicts events. ApoB or LDL, blood pressure, A1C, and a calcium score if you have risk factors.

So what is left of the video?

A cardiologist avoids testing her A1C for years, adds fish back to her diet, then runs the labs. Fasting insulin comes back normal at 4.9. A1C comes back at 5.9, in a 64-year-old running 50 miles a week with three generations of diabetes, eating under 50 grams of protein and 75 to 80 percent refined carbohydrate, taking no omega-3 supplement. She attributes the result to plants, a diet she never measured. She then adopts a diet that takes her LDL from 80 to 210 and her apoB to 217, and explains on camera why the particle count does not matter.

Her patients lost weight. Their chest pain improved. She says so herself. That is the outcome data in this video, and it points the other way.

For the rest of the evidence, see our guides to the nutritional case for plants, cholesterol, dietary fat, seed oils, coconut oil versus animal fat, the Adventist Health Studies, what the longest-living people eat, losing weight, plant toxins and antinutrients, and raising children vegan.

References and further reading

Links open in a new window.

Endurance exercise and glucose

Prins PJ, Noakes TD, Buga A, et al. Carbohydrate ingestion eliminates hypoglycemia and improves endurance exercise performance in triathletes adapted to very low- and high-carbohydrate isocaloric diets. Am J Physiol Cell Physiol 2025. Full text
Noakes TD, Prins PJ, Buga A, D’Agostino DP, Volek JS, Koutnik AP. Carbohydrate ingestion on exercise metabolism and physical performance. Endocrine Reviews 2026;47(2):191–243. DOI
Noakes TD. What is the evidence that dietary macronutrient composition influences exercise performance? Nutrients 2022;14:862. PMC
Flockhart M, et al. Excessive exercise training causes mitochondrial functional impairment and decreases glucose tolerance in healthy volunteers. Cell Metab 2021;33(5):957–970. DOI

Protein

Hevia-Larraín V, et al. High-protein plant-based diet versus a protein-matched omnivorous diet to support resistance training adaptations. Sports Med 2021;51(6):1317–1330. DOI
Raj S, et al. Vegetarian dietary patterns for adults: a position of the Academy of Nutrition and Dietetics. J Acad Nutr Diet 2025;125(6):831–846. PubMed
Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. National Academies Press; 2005.

Insulin, diabetes, and body weight

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
Tonstad S, et al. Vegetarian diets and incidence of diabetes in the Adventist Health Study-2. Nutr Metab Cardiovasc Dis 2013;23(4):292–299. DOI
Satija A, et al. Plant-based dietary patterns and incidence of type 2 diabetes in US men and women. PLOS Med 2016;13(6):e1002039. DOI
Li C, et al. Meat consumption and incident type 2 diabetes: a federated meta-analysis of 1.97 million adults. Lancet Diabetes Endocrinol 2024;12(9):619–630. DOI
Kahleova H, et al. Effect of a low-fat vegan diet on body weight, insulin sensitivity, postprandial metabolism, and intramyocellular and hepatocellular lipid levels in overweight adults. JAMA Netw Open 2020;3(11):e2025454. DOI
Hall KD, 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. DOI
Bao W, Rong Y, Rong S, Liu L. Dietary iron intake, body iron stores, and the risk of type 2 diabetes. BMC Med 2012;10:119. PMC
Staten MA, et al. Insulin assay standardization. Diabetes Care 2010;33(1):205–206. DOI
Wright N, et al. The BROAD study: a randomised controlled trial using a whole food plant-based diet. Nutr Diabetes 2017;7:e256. DOI
Termannsen AD, et al. Effects of vegan diets on cardiometabolic health: a systematic review and meta-analysis of RCTs. Obes Rev 2022;23(9):e13462. DOI
Gardner CD, et al. Effect of low-fat vs low-carbohydrate diet on 12-month weight loss: the DIETFITS trial. JAMA 2018;319(7):667–679. DOI
Lean MEJ, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT). Lancet 2018;391(10120):541–551. DOI
Barnard ND, et al. A low-fat vegan diet improves glycemic control and cardiovascular risk factors in type 2 diabetes. Diabetes Care 2006;29(8):1777–1783. PubMed
Goldenberg JZ, et al. Efficacy and safety of low and very low carbohydrate diets for type 2 diabetes remission. BMJ 2021;372:m4743. DOI

DHA, EPA, and the omega-3 index

Sarter B, et al. Blood DHA and EPA in vegans, and effects of an algal-derived omega-3 supplement. Clin Nutr 2015. PubMed
Domenichiello AF, Kitson AP, Bazinet RP. Is DHA synthesis from ALA sufficient to supply the adult brain? Prog Lipid Res 2015. PubMed
Umhau JC, et al. Imaging incorporation of circulating DHA into the human brain using PET. J Lipid Res 2009. Full text
Loong S, et al. Omega-3, cognition, and brain volume in older adults: AHS-2. Brain Sci 2023;13(9):1300. Full text
Tan ZS, Harris WS, Beiser AS, et al. Red blood cell omega-3 levels and markers of accelerated brain aging. Neurology 2012. PubMed
Chew EY, et al. AREDS2 cognitive outcomes. JAMA 2015. PubMed
Kang JH, et al. VITAL cognitive ancillary study. Alzheimers Dement (TRCI) 2022. PubMed
Andrieu S, et al. MAPT trial. Lancet Neurol 2017. PubMed
Dangour AD, et al. OPAL trial. Am J Clin Nutr 2010. PubMed
Quinn JF, et al. DHA supplementation and cognitive decline in Alzheimer disease. JAMA 2010. PubMed
Shinto L, et al. Omega-3 and white matter integrity in older adults. JAMA Netw Open 2024. PubMed
Yassine HN, et al. PreventE4: CNS target engagement of high-dose DHA. eBioMedicine 2026 / SSRN. SSRN

Dogs, amylase, and pancreatitis

Axelsson E, et al. The genomic signature of dog domestication reveals adaptation to a starch-rich diet. Nature 2013;495:360–364. DOI
Perry GH, et al. Diet and the evolution of human amylase gene copy number variation. Nat Genet 2007;39:1256–1260. DOI
Cridge H, Lim SY, Algül H, Steiner JM. New insights into the etiology, risk factors, and pathogenesis of pancreatitis in dogs. J Vet Intern Med 2022;36(3):847–864. DOI
US Food and Drug Administration. Get the facts: raw pet food diets can be dangerous to you and your pet. FDA

Cholesterol, LDL, and heart disease

Ference BA, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease: EAS consensus statement. Eur Heart J 2017;38(32):2459–2472. DOI
Hooper L, et al. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev 2020;CD011737. DOI
Koch CA, Kjeldsen EW, Frikke-Schmidt R. Vegetarian or vegan diets and blood lipids: a meta-analysis of randomized trials. Eur Heart J 2023;44(28):2609–2622. DOI
Landry MJ, et al. Cardiometabolic effects of omnivorous vs vegan diets in identical twins. JAMA Netw Open 2023;6(11):e2344457. DOI
Tong TYN, et al. Risks of ischaemic heart disease and stroke in meat eaters, fish eaters, and vegetarians: EPIC-Oxford. BMJ 2019;366:l4897. DOI

Keto and carnivore

Patikorn C, et al. Effects of ketogenic diet on health outcomes: an umbrella review of meta-analyses of RCTs. BMC Med 2023;21:196. DOI
Lennerz BS, Mey JT, Henn OH, Ludwig DS. Behavioral characteristics and self-reported health status among 2029 adults consuming a carnivore diet. Curr Dev Nutr 2021;5(12):nzab133. DOI
Retraction notice for the KETO-CTA lean mass hyper-responder study. JACC: Advances 2026. JACC
Retraction Watch. Widely criticized keto diet study retracted. May 22, 2026. Retraction Watch

Other coverage of this interview

Mic the Vegan. How a vegan cardiologist ended up going carnivore. YouTube (the source for the timing of her A1C test, her apoB value, and the endurance-athlete A1C point)