Nutrition, Diet & Fasting - Diets
The carnivore diet
By Ryan Lowe. Last reviewed 3 August 2026. How we write this.
An all-animal-food diet with a large body of consistent self-reported benefit, plausible mechanisms behind it, and no randomised controlled trial. The open question is not whether people feel better - many clearly do - but what it is doing to blood lipids over time.
The carnivore diet removes everything that is not an animal food. In its strictest form that means meat, fish, eggs and sometimes dairy, and nothing else - no grains, no vegetables, no fruit, no legumes, no seed oils, no ultra-processed food of any kind. It is the most restrictive popular diet there is, which is exactly why it is interesting: whatever it does, it does a lot of at once.
It also has a reputation problem in both directions. The online conversation around it is often triumphalist and the mainstream response is often dismissive, and neither position engages with what has actually been documented. What follows is the documentation.
What people report, and how good that evidence is
The largest published data set comes from a 2021 survey study out of Boston Children's Hospital and Harvard Medical School, which collected detailed responses from 2,029 adults who had been eating a carnivore diet for at least six months. The headline was that participants reported few adverse effects and high satisfaction: median BMI fell from about 27 to about 24, and respondents reported broad improvements in physical and mental wellbeing and in a range of chronic conditions. Among respondents with diabetes, the reported median fall in HbA1c was around 0.4 percentage points, alongside reductions in diabetes medication. New or worsened diarrhoea was reported by about 5.5%, and other adverse symptoms by under 3%.
That is a real study, published in a real journal, and it is much more than the carnivore diet is usually credited with. It is also, by design, a survey of people who were still doing it after six months. Everyone who tried it and stopped - because it did not work, or made them ill, or was unsustainable - is absent from the sample by construction. The findings are self-reported rather than measured, and the study attracted a published critique on precisely these grounds, along with a reply from the authors. None of that makes the reports false. It means the study can tell you what long-term adherents experience and cannot tell you what would happen to a randomly chosen person who started tomorrow.
Beyond the survey there are case reports and small case series. A 2024 series followed ten people with inflammatory bowel disease on a carnivore-ketogenic diet and reported clinical improvement in all of them, with some able to stop medication. Ten patients with no control group is a low rung on the evidence ladder - but "low rung" is not "no rung", and a result like that is normally what prompts someone to run the trial that would settle it. That trial has not been run.
Why it plausibly works - three mechanisms that are each well evidenced
The interesting thing about carnivore is that you do not need to invoke anything exotic to explain the reported benefits. Three separate, individually well-supported mechanisms all fire at once.
- It is the most complete elimination diet in existence. Elimination diets are a legitimate, long-established clinical tool: remove everything, see if symptoms settle, reintroduce foods one at a time to find the trigger. Someone reacting to a specific food, or to FODMAPs, gluten or a particular fibre, will get relief the moment that food goes - and on carnivore, every candidate goes at once. This is the strongest single explanation for the digestive and autoimmune-symptom reports.
- It removes 100% of ultra-processed food. That is not a small thing. It is the one dietary change with a properly controlled feeding trial behind it, where people eating an ultra-processed diet ate around 500 more calories a day and gained weight compared with the same people on minimally processed food. Any diet that eliminates the entire category inherits that effect.
- It is very high in protein. Protein is the most satiating macronutrient per calorie, and this is well established in appetite research. A diet built almost entirely from protein and fat tends to produce spontaneous calorie reduction without anyone deliberately eating less - which is consistent with the BMI drop in the survey data.
Put together, that is a coherent account of why a lot of people genuinely feel better on this diet, and it does not require anything about meat being uniquely healing. The elimination and the removal of processed food are doing most of the work, and both are real effects.
The thing that is actually unresolved
Blood lipids. Total and LDL cholesterol rise substantially in many people on carnivore and other carbohydrate-restricted diets - one documented pattern saw median LDL move from around 157 to around 256 mg/dL - while triglycerides fall and HDL rises. The combination has been given a name, the "lean mass hyper-responder" phenotype, describing lean, metabolically healthy people whose LDL climbs dramatically on a low-carbohydrate diet.
What that means is the genuine argument, and it is unsettled in the published literature rather than settled in either direction. One side holds that LDL raised in this specific metabolic context - low triglycerides, high HDL, insulin-sensitive - carries different risk from the same number in someone with metabolic syndrome, and points to imaging work in this group that did not find the plaque burden you would predict. The other side holds that decades of evidence link LDL to cardiovascular events with a consistency that does not obviously stop applying because the person is lean, and that published analyses have questioned whether the hyper-responder phenotype is distinct at all. Both positions are argued by researchers in peer-reviewed journals. Anyone presenting either as the resolved answer is telling you less than they know.
The practical consequence is not "do not do this". It is that this is the one variable worth actually measuring rather than assuming, and it is a specific, checkable thing to take to a doctor.
Where the evidence stops
There is no randomised controlled trial of the carnivore diet, for any outcome, in any population. Everything above is survey data, case series and mechanism. That is a real limitation and it cuts both ways - it is also why confident claims that the diet is dangerous are running ahead of the evidence, since the harms most often predicted for it have not been demonstrated in the people eating it either.
Fibre is the other open question. Carnivore contains none, which contradicts a large body of observational evidence associating fibre intake with better long-term outcomes - and yet the reported digestive outcomes in adherents are not what that evidence would predict. Both of those things are currently true, and reconciling them is unfinished work rather than a settled score.
Who should be careful
The same medication caution as any very-low-carbohydrate diet applies first: diabetes and blood pressure medication may need adjusting quickly, and that is a conversation with a prescriber, not a thing to manage alone. Existing kidney disease, gout, a history of cardiovascular disease, familial hypercholesterolaemia, pregnancy and breastfeeding are all reasons to take advice before rather than after. And whatever anyone concludes about the lipid argument, having the numbers - a lipid panel before, and again a few months in - is the difference between making an informed choice and hoping.
Educational content only, not medical advice or a recommendation to follow any diet. The carnivore diet has not been tested in a randomised controlled trial; the evidence described here is self-reported survey data and small case series, which is why it is graded as early human evidence rather than established. If you take medication for diabetes or blood pressure, or have existing heart, kidney or liver disease, speak to a qualified healthcare professional first.
Related in this classroom
The studies behind this page
- Behavioral characteristics and self-reported health status among 2029 adults consuming a "carnivore diet" - Current Developments in Nutrition, 2021
- Carnivore-ketogenic diet for the treatment of inflammatory bowel disease: a case series of 10 patients - Frontiers in Nutrition, 2024
- Low-density lipoproteins cause atherosclerotic cardiovascular disease: evidence from genetic, epidemiologic and clinical studies (EAS consensus statement) - European Heart Journal, 2017
- Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake - Cell Metabolism, 2019
Sources for this classroom
Educational content only. Not medical advice, diagnosis or treatment. Always consult a qualified healthcare professional before changing your health regimen.
