Is red meat bad for you? Limited research robs us of a clear answer.
Over and over, we ask the question: Is Food X good or bad for you? And, over and over, belief in the answer — whether it’s yes or no — is held with conviction totally out of proportion with the strength of the evidence.Today’s illustration: red meat. It has become one of the most-disputed issues in food. It’s so polarizing that some people decide to eat no meat at all, while others decide to eat only meat. It’s poison, or it’s the only true fuel.The latest salvo in the Meat Wars was kicked off by a new report that outlines the optimal diet for both people and planet. The EAT-Lancet Report comes down hard on red meat; its recommended daily intake is a mere 14 grams — that’s half an ounce.Read on, and the news gets worse: “Because intake of red meat is not essential and appears to be linearly related to higher total mortality and risks of other health outcomes in populations that have consumed it for many years, the optimal intake may be zero.”Note that word: “related.” It’s the source of the problem with the report and its recommendation.The EAT-Lancet report, by researchers from 17 countries, bases its recommendation solely on observational data. When you do that, meat comes out looking pretty bad. In study after study, people who report eating a lot of meat have worse health outcomes than people who eat little. Meat-eating correlates with increased risk of heart disease, some cancers and all-cause mortality.But, as always with observational research that attempts to connect the dots between diet and health, the key question is whether the meat itself, or something else associated with a meat-heavy lifestyle, is actually causing the bad outcomes.That’s a hard question to answer, but there are clues that people who eat a lot of meat are very different from people who eat a little.Let’s look at a study, published in JAMA Internal Medicine, cited by the EAT-Lancet report; it has a convenient demographic summary. According to it, people in the top one-fifth of meat eaters are different from people in the bottom fifth in a lot of important ways: They weigh more, they’re more likely to smoke, they’re not as well-educated, they get less exercise, and they report lower intakes of fruit, vegetables and fiber. On the plus side, they report drinking less alcohol. But other than that, we’re looking at a litany of markers for a lifestyle that’s not particularly health-conscious.So, to suss out whether it’s the meat that’s raising disease risk, you have to somehow correct for any of the differences on that list — and most of that information also comes from observational research, so even the confounders are confounded.Then there are the things you can’t correct for. Sleep quality, depression and screen time, for example, all correlate with some of the same diseases meat correlates with, but most studies have no information on those.All this confounding explains one of my all-time favorite findings from observational research. It comes from the same study the demographics came from (analyzed in a 2015 paper). Sure enough, the people who ate the most meat were more likely to die of cancer and heart disease, but they were also more likely to die in accidents. And the biggest difference came from the catchall category “all others,” which invariably includes causes of death that have nothing to do with meat.Basically, there’s a very simple problem with relying on observational research: People who eat a lot of meat are very different from people who eat less of it. The meat definitely isn’t causing the accidental deaths (unless, perhaps, they’re tragic backyard grill mishaps), and it isn’t causing at least some of the “all others” deaths, so we know that heavy and light meat-eaters are different in all kinds of ways.That’s where controlled trials come in.In a perfect world, we could figure this out by keeping a large group of people captive for a lifetime, feeding half of them meat, and seeing what happens. Okay, maybe that’s not a perfect world, but it would be the best solution to this particular problem.Instead, we have trials that are short-term (because of logistics and cost), and necessarily rely on markers for disease, rather than the disease itself. For that to be useful, you need a marker that’s a reliable indicator. For a lot of diseases — including cancer — those are hard to come by. For heart disease, we have a good one: low-density lipoprotein (LDL) cholesterol. So, most of the controlled trials of meat-eating focus on heart disease.If you spend some time reading those trials (and I did, so you don’t have to), you find that most of them show some increase in LDL cholesterol, although it’s generally small.A 2025 analysis of 44 controlled trials on meat found that the only ones showing positive cardiovascular outcomes had links to the meat industry, and even then, only about one in five came out positive. Of the independent studies, about three-quarters showed negative outcomes, and the remaining one-quarter was neutral.This isn’t surprising. Red meat contains saturated fat, and we have countless trials that demonstrate sat fat’s ability to raise LDL. But if the meat you eat is relatively lean, that effect is going to be small.The lesson here is that we don’t have a lot of good evidence on meat and health. The observational evidence is hopelessly confounded, and the evidence from clinical trials is woefully limited. There’s so much we simply don’t know. There may be other ways meat raises risk (leading to over-absorption of heme iron and stimulating the production of TMAO, or trimethylamine N-oxide), but there’s little definitive evidence for them. And, of course, there’s the question of what you eat instead. If you’re eating red meat instead of, say, instant ramen, that may be an improvement. If, instead, you’re cutting back on your lentils, not so much.As always, the single-most important thing to remember about nutrition is that what we know is absolutely dwarfed by what we don’t know. Which means that, if you’re making decisions based on what we do know, you could very well be wrong.So what’s an eater to do? Meat is a nutritious food. In fact, animal foods are the only natural sources of a vitamin we need — B12 — which is an indication that we evolved with meat and dairy as part of our diet. It’s very hard to know whether eating some lean meat leads to better outcomes than eating no meat, but I think some meat is a good hedge against all that uncertainty. (The ethical and environmental concerns are also important, but for today let’s focus on health.)But plant foods are also nutritious. And eating a wide variety of them is also a good hedge against uncertainty. Which means the carnivore diet — all meat, all the time! — is a pretty bad bet.Unfortunately, “uncertainty” is not a word that features prominently in the Meat Wars. Instead, we have an unappetizing combination of nastiness and sanctimony, with each camp convinced that the truth and the light are on their side.Not that this is a metaphor for our times or anything.
We’d all appreciate more definitive guidance. Eating a varied diet is a wise move while we wait.
Over and over, we ask the question: Is Food X good or bad for you? And, over and over, belief in the answer — whether it’s yes or no — is held with conviction totally out of proportion with the strength of the evidence.
Today’s illustration: red meat. It has become one of the most-disputed issues in food. It’s so polarizing that some people decide to eat no meat at all, while others decide to eat only meat. It’s poison, or it’s the only true fuel.
The latest salvo in the Meat Wars was kicked off by a new report that outlines the optimal diet for both people and planet. The EAT-Lancet Report comes down hard on red meat; its recommended daily intake is a mere 14 grams — that’s half an ounce.
Read on, and the news gets worse: “Because intake of red meat is not essential and appears to be linearly related to higher total mortality and risks of other health outcomes in populations that have consumed it for many years, the optimal intake may be zero.”
Note that word: “related.” It’s the source of the problem with the report and its recommendation.
The EAT-Lancet report, by researchers from 17 countries, bases its recommendation solely on observational data. When you do that, meat comes out looking pretty bad. In study after study, people who report eating a lot of meat have worse health outcomes than people who eat little. Meat-eating correlates with increased risk of heart disease, some cancers and all-cause mortality.
But, as always with observational research that attempts to connect the dots between diet and health, the key question is whether the meat itself, or something else associated with a meat-heavy lifestyle, is actually causing the bad outcomes.
That’s a hard question to answer, but there are clues that people who eat a lot of meat are very different from people who eat a little.
Let’s look at a study, published in JAMA Internal Medicine, cited by the EAT-Lancet report; it has a convenient demographic summary. According to it, people in the top one-fifth of meat eaters are different from people in the bottom fifth in a lot of important ways: They weigh more, they’re more likely to smoke, they’re not as well-educated, they get less exercise, and they report lower intakes of fruit, vegetables and fiber. On the plus side, they report drinking less alcohol. But other than that, we’re looking at a litany of markers for a lifestyle that’s not particularly health-conscious.
So, to suss out whether it’s the meat that’s raising disease risk, you have to somehow correct for any of the differences on that list — and most of that information also comes from observational research, so even the confounders are confounded.
Then there are the things you can’t correct for. Sleep quality, depression and screen time, for example, all correlate with some of the same diseases meat correlates with, but most studies have no information on those.
All this confounding explains one of my all-time favorite findings from observational research. It comes from the same study the demographics came from (analyzed in a 2015 paper). Sure enough, the people who ate the most meat were more likely to die of cancer and heart disease, but they were also more likely to die in accidents. And the biggest difference came from the catchall category “all others,” which invariably includes causes of death that have nothing to do with meat.
Basically, there’s a very simple problem with relying on observational research: People who eat a lot of meat are very different from people who eat less of it. The meat definitely isn’t causing the accidental deaths (unless, perhaps, they’re tragic backyard grill mishaps), and it isn’t causing at least some of the “all others” deaths, so we know that heavy and light meat-eaters are different in all kinds of ways.
That’s where controlled trials come in.
In a perfect world, we could figure this out by keeping a large group of people captive for a lifetime, feeding half of them meat, and seeing what happens. Okay, maybe that’s not a perfect world, but it would be the best solution to this particular problem.
Instead, we have trials that are short-term (because of logistics and cost), and necessarily rely on markers for disease, rather than the disease itself. For that to be useful, you need a marker that’s a reliable indicator. For a lot of diseases — including cancer — those are hard to come by. For heart disease, we have a good one: low-density lipoprotein (LDL) cholesterol. So, most of the controlled trials of meat-eating focus on heart disease.
If you spend some time reading those trials (and I did, so you don’t have to), you find that most of them show some increase in LDL cholesterol, although it’s generally small.
A 2025 analysis of 44 controlled trials on meat found that the only ones showing positive cardiovascular outcomes had links to the meat industry, and even then, only about one in five came out positive. Of the independent studies, about three-quarters showed negative outcomes, and the remaining one-quarter was neutral.
This isn’t surprising. Red meat contains saturated fat, and we have countless trials that demonstrate sat fat’s ability to raise LDL. But if the meat you eat is relatively lean, that effect is going to be small.
The lesson here is that we don’t have a lot of good evidence on meat and health. The observational evidence is hopelessly confounded, and the evidence from clinical trials is woefully limited. There’s so much we simply don’t know. There may be other ways meat raises risk (leading to over-absorption of heme iron and stimulating the production of TMAO, or trimethylamine N-oxide), but there’s little definitive evidence for them. And, of course, there’s the question of what you eat instead. If you’re eating red meat instead of, say, instant ramen, that may be an improvement. If, instead, you’re cutting back on your lentils, not so much.
As always, the single-most important thing to remember about nutrition is that what we know is absolutely dwarfed by what we don’t know. Which means that, if you’re making decisions based on what we do know, you could very well be wrong.
So what’s an eater to do? Meat is a nutritious food. In fact, animal foods are the only natural sources of a vitamin we need — B12 — which is an indication that we evolved with meat and dairy as part of our diet. It’s very hard to know whether eating some lean meat leads to better outcomes than eating no meat, but I think some meat is a good hedge against all that uncertainty. (The ethical and environmental concerns are also important, but for today let’s focus on health.)
But plant foods are also nutritious. And eating a wide variety of them is also a good hedge against uncertainty. Which means the carnivore diet — all meat, all the time! — is a pretty bad bet.
Unfortunately, “uncertainty” is not a word that features prominently in the Meat Wars. Instead, we have an unappetizing combination of nastiness and sanctimony, with each camp convinced that the truth and the light are on their side.
Not that this is a metaphor for our times or anything.