I was at Costco last Saturday, working the sample tables like everybody else.
The guy at the beef stick cart held one out to the woman ahead of me. She waved it off and said she’d gone plant-based. He grinned and told her, “Good for you. Staying healthy and saving the planet.”
She smiled, thanked him, and got on with her Saturday. That’s fine. Her dinner is none of my business.
But that sentence rode home with me, “…Staying healthy and saving the planet.”
Because that guy didn’t reason his way to it. He absorbed it, like it was some kind of accepted trueism. Something as true as 1 + 1 \= 2. But it’s not true, and institutions have spent an enormous amount of money making sure a man handing out beef samples would say it, for free, without being asked.
There are two claims stuffed in there. Healthy, and saves the planet. I want to take the first one seriously, because it’s the one people actually type into Google.
The short answer: Is meatless meat healthy?
A little, yes. Pooled across seven randomized trials and 369 adults, swapping meat for plant-based versions lowered LDL by 0.25 mmol/L, about 12 percent from baseline. Real effect. Modest one.
What’s more interesting is everything that didn’t move. That same analysis01428-X/abstract) found no significant change in HDL, triglycerides, blood pressure, or fasting glucose. Stanford’s SWAP-MEAT trial, one of the studies in that pool, reported the same shape of result. LDL and weight came down. HDL, triglycerides, insulin, glucose, and blood pressure all sat still.
So one marker moved. That’s the whole case.
And the reason it moved isn’t a very complex story either. The plant products simply carried less saturated fat and more fiber00890-5/fulltext), both of which were already known to lower LDL. You’d get the same direction from beans.
Now the cost side. Most of these products meet the NOVA criteria for ultra-processed food, and the NIH ran the cleanest test we have30248-7) of what that category does to people. Twenty adults lived in a metabolic ward eating ultra-processed and unprocessed diets for 14 days each, matched for calories, sugar, fat, sodium, fiber, and macronutrients. While eating the processed version, they ate about 500 more calories a day, and their weight followed.
Sodium is the other tab. Both major meatless patties carry more than four times the sodium of a beef burger. And while I think most people overreact to sodium in general, it’s easy to overdo it with processed foods.
And the LDL result isn’t even universal. An eight-week trial of 89 adults00396-4/fulltext) comparing plant-based meats against their animal equivalents found no significant effect on LDL or any other cardiometabolic marker.
Some plant-based meat is better than the industrial meat it replaces. Almost none of it is better than the whole food it replaces. Which package you’re holding matters more than the words on the front of it.
That’s the part the label can’t help you with, because “plant-based” describes where an ingredient came from, not what it does once you eat it. You know what else is plant-based?
- Crisco. Soybean oil, fully hydrogenated palm oil, palm oil. The company’s own site calls it plant-based.
- High fructose corn syrup
- Betty Crocker vanilla frosting
- Sour Patch Kids. The US version contains no gelatin.
- Fritos Original. Corn, corn oil, salt.
“Plant-based” is a category. Categories tell you almost nothing. Method tells you everything.
Quality in, quality out.
So rather than argue about categories, let me walk you through how these things get made. Once you see the process, you won’t need me to hand you a verdict.
How plant-based meat is actually made
Start with a yellow pea. Or a soybean. Or wheat.
Strip out everything that isn’t protein. A dry split pea is about a quarter protein and a quarter fiber by weight. This isolation process only keeps the first quarter.
Push the isolate through an extruder under heat, pressure, and moisture. The proteins are then denatured and reassembled into layered fibers that read, roughly, as muscle. This is the whole trick of the category, and as far as tricks go, it’s a pretty good one, considering we’re taking essential peasant food and using alchemy to turn it into something that is kind of like a muscle. By the way, the industry calls it “high-moisture extrusion,” not cooking, which tells you something.
Then, they add fat back in, because protein isolate on its own is dry and joyless. Hopefully, coconut oil, but more often it’s sunflower oil, or canola oil. The good news is that Beyond switched to avocado oil in 2024, which brought the saturated fat in its burger down to 2 grams per serving.
After adding fat, you add methylcellulose, a cellulose gum that binds, so the patty holds together in the pan like ground beef. You know, the stuff that termites eat.
Add salt, because nothing I’ve mentioned so far tastes like anything people want to eat.
Then put back the iron, B12, and zinc you stripped out at step two. If the manufacturer bothers. An Australian audit of 132 of these products found only 12 percent were fortified with all three.
And when they do bother, the iron goes back in as a non-heme salt like ferrous fumarate, which absorbs less readily than the heme iron in meat. The label matches. The absorption doesn’t.
That’s it. That’s the food. Excuse me… “food.”
The problem isn’t that it’s made of plants, far from it!
You can make a plant protein with three ingredients and almost no processing. People have been doing it for a thousand years. One example is called tempeh, which is made from soybeans, a culture (i.e., bacteria), and time. And it isn’t only the old stuff: in that same meta-analysis01428-X/abstract), the mycoprotein products came out ahead of the category average, dropping LDL by 0.37 mmol/L against 0.25 for everything pooled. Mycoprotein is grown as a whole organism rather than as an isolated, re-textured Franken-food.
So it isn’t that nobody has figured out how to do this well. It’s that the version that won shelf space is the one engineered to be produced cheaply, shipped shelf-stable to forty thousand stores, and cooked like beef by someone who has never cooked a bean. Every decision on that list falls out of those requirements, and none of them fall out of nutrition.
So what would a legitimate meat alternative look like?
It would look like beans. But… Now we need to talk about beans, which vegans love to say is full of protein. But beans are not beef.
Protein quality is scored with DIAAS, the Food and Agriculture Organization’s (FAO) system, which measures how much of each essential amino acid you actually absorb rather than how much is printed on the panel. Meat products score between 99 and 130. Cooked pulses (beans, peas, lentils, etc.) land between 60 and 88 for adults, and the limiting amino acid in nearly every legume is methionine. Under 75, the FAO says you can’t make a protein quality claim at all. The Stein lab at Illinois ran the head-to-head and found that beef and pork burgers scored higher than plant-based burgers.
Any literate vegan will tell you (and I agree and think it’s really cool that our Earth was designed this way) that you can close the amino acid gap by pairing beans with rice. But you can’t close the rest of the gap, as if amino acids were the only virtuous aspect of eating red meat. For example, there is no B12 in a bean, or nearly any plant, for that matter. The iron is non-heme and absorbs poorly.
So this isn’t a pitch that beans replace steak. They don’t. They emphatically don’t.
The point is narrower than that. If you want beans, eat beans. Cheap, filling, high fiber, nothing on the label to decode. If you want the most nutrient-dense protein available to you, eat red meat. Both of those are food.
The dubious proposition isn’t the plants, it’s the process. Meatless meat tends to be an ultra-processed product engineered to resemble one of them while underperforming both.
When you have to add back the iron, the B12, the zinc, the fiber, and the flavor, one at a time, just to get a panel that reads like the food you’re imitating, you haven’t built an alternative. You’ve built a costume.
What’s actually in the package
Typical values for a four-ounce serving. Products vary a lot, which is the whole point, so read your own label.
| Plant-based patty | 80/20 ground beef | Lentils (1 cup) | |
|---|---|---|---|
| Calories | 230 to 240 | 290 | 230 |
| Protein | 19 to 20 g | 24 g | 18 g |
| Protein quality | Lower | Complete | Lower |
| Sodium | 370 to 400 mg | 75 mg | 4 mg |
| Saturated fat | 5 to 8 g | 6.5 g | 0.1 g |
| Iron | Added (variable), non-heme | 2.4 mg, heme | 6.6 mg, non-heme |
| Vitamin B12 | Added (variable), or none | 2.4 mcg | None |
| Fiber | 2 to 3 g | 0 g | 15 g |
| Cholesterol | 0 mg | 80 mg | 0 mg |
| Ingredients | 18 to 21 | 1 | 1 |
Two rows deserve more than a row.
Protein quality is not protein quantity. The gram count on the front of the box is honest. What it doesn’t tell you is the amino acid profile, or how much of it you actually absorb. Researchers at the University of Illinois and Colorado State ran both burgers through DIAAS, the FAO’s absorption-based scoring system, and published the numbers in the European Journal of Nutrition. Beef and pork patties scored above 100, excellent for every age group. The Beyond Burger came in at 83, good for ages three and up. Add the bun, which drags every score down, and the gap gets concrete: beef landed at 105 and the Impossible Burger at 86, meaning you need to eat 15 percent more of the Impossible meal to get the same digestible amino acids. And more food means more calories.
Two things I’ll put on the table right now. That study was funded in part by the Beef Checkoff and the National Pork Checkoff. And it found the Impossible Burger’s protein quality comparable to 80 percent lean beef for anyone over three. I’m citing it anyway, because the methodology is the FAO standard and the numbers are what they are, but you should know both of those before you decide how much weight it carries.
Iron from meat and iron from plants are different molecules. Meat gives you heme iron. Plants give you non-heme iron, which absorbs at a fraction of the rate, and phytates and other anti-nutrients block part of what’s left. Fortification doesn’t fully solve it, because the iron gets added back as a non-heme salt like ferrous fumarate, which behaves like plant iron, not meat iron. A 2024 review in Food Research International makes the same point about B12 and flags older adults and people with gastrointestinal disorders as the groups most exposed if these products become the primary protein.
Sodium is the most consistent problem across the category. The Canadian Journal of Cardiology review flags it01882-2/abstract). The George Institute’s product audit flags it. An integrative review of the global product category flags it. Nobody in this argument disputes the sodium.
Where the evidence disagrees with me
Now, the part most people in my corner of the internet skip.
In 2024, a review in the Canadian Journal of Cardiology01882-2/abstract) worked through studies published from 1970 to 2023 and found that swapping conventional meat for these products lowered total and LDL cholesterol in randomized trials. It also found they didn’t raise blood pressure, despite the sodium. Their conclusion was blunt: no currently available evidence suggests the processing and sodium concerns cancel out the cardiovascular benefits.
That’s real. Randomized trials, not a survey. And it points away from where I’ve landed.
So here’s why I still land where I do, and you can judge whether it holds up.
Cholesterol markers moving over eight weeks is not the same claim as being healthy over thirty years. The same review says plainly that long-term evidence on heart attack and stroke risk is missing, and that’s a paper broadly friendly to these products.
Most of those trials also compare the patty against conventional processed meat. Not against lentils. Beating a hot dog is a low bar, and in debate we have a name for picking the weakest available opponent. It’s a strawman.
Here’s another fun fact. There’s well-established data that shows that prolonged fasting (24 hours) will increase cholesterol, something that is universally accepted as healthy these days, actually raises cholesterol in the short run.
Who is answering this question for you
Look at who is producing the answers.
The Good Food Institute publishes a great deal of accessible writing on this topic, and it ranks well. GFI is an advocacy organization for the alternative protein industry. They don’t hide it, it’s their stated mission. They’re just not a neutral referee. Same for Plant Based Health Professionals, where you can guess the position from the domain.
Even careful reporting carries this “meat-bad” narrative. CNN ran a piece in July 2025 on whether these products count as ultra-processed and whether they are a net positive. I’ll save you the click: an “expert” said that eating these ultra-processed foods can be healthy. Who are these “experts?” In this case, it is a registered dietitian, and CNN discloses in the same sentence that she consults for Beyond Meat.
And I’m not exempt. The DIAAS study I leaned on above was funded by the beef and pork checkoffs. The Canadian Journal of Cardiology review I just conceded to was led by a naturopathic doctor whose practice is built around plant-based nutrition. The White Oak Pastures assessment I cited earlier was commissioned by White Oak Pastures itself.
Everybody in this fight is funded by somebody. The question isn’t whether a source has a stake. It’s whether they tell you, and whether the method survives you knowing.
Does avoiding meat actually help save the planet?
Now, let’s address the other half of the Costco sample guy’s sentence. The planet-saving argument is weaker than the healthy argument.
In 2019, White Oak Pastures in Bluffton, Georgia released a full life cycle assessment run by Quantis on 2017 farm data. It came back at net negative 3.5 kilograms of CO2 equivalent per kilogram of beef, against 33 for conventional US beef. A peer-reviewed follow-up in 2020 backed the direction and landed on a footprint 66 percent lower than conventional.
I love this study, and frequently reference it, but I hate it when “the other side” cherry picks data, so to be fair, I have to call out four items worth more discussion on this topic.
- Their gross emissions are higher than those of a feedlot. Enteric methane came in at around 29 kg per kg, and manure accounted for another 5 kg per kg. Grass-fed cattle live longer, due to better health, and they take longer to reach a target weight. And grass-fed cattle belch more. The entire net-negative result rests on the soil sequestration side of the ledger, roughly 35 kg, doing very heavy lifting. And how you divide that 35 changes the answer. Quantis credited it to the cattle, but the peer-reviewed version spread it across every animal on the farm, which is part of why its number came in less dramatically. Same soil, same farm, different denominator. If you don’t buy the soil math, you don’t get the headline.
- Soil carbon saturates. Quantis sampled pastures at 0, 1, 3, 5, 8, 13 and 20 years under management. Ground being rebuilt pulls carbon fast, then slows, then levels off. That figure is a snapshot of a farm in the middle of a recovery, not a permanent rate.
- It takes more land. The same peer-reviewed study found the system required 2.5 times the land of conventional production for the same output. That’s the honest cost, and anyone arguing this scales to feed a country has to answer for it. It measured one farm. Not “beef.” Which is the point, since the whole argument is that White Oak and a feedlot are not the same thing, wearing different hats.
- One more, and I enjoy this one. The Impossible Burger’s life cycle assessment was done by Quantis as well. Same firm. So if you want to throw out my number because a consultancy got paid to produce it, you’re throwing out theirs too.
For the record, a peer-reviewed comparison that put both products on the same basis landed Beyond at 3.3 and Impossible at 3.4 kilograms of CO2 equivalent per kilogram. Both positive. So sure, you can offset a friend’s Impossible Burger by eating a pound of regeneratively raised beef.
The bigger thing is that fake meat grows in the same monocropped, tilled, sprayed dirt as everything else in the middle aisles. Beyond’s own assessment attributes 68 percent of its land-use-change emissions to pea protein cultivation and another 28 percent to canola oil. It isn’t an alternative to industrial agriculture. It’s a product of it.
If you’re dead set on meatless meat…
Here’s what the data seems to suggest, if you’re interested in only eating plants… Eat actual plants. Beans, lentils, tofu, tempeh. Whole food plants are cheaper than the meatless patty and cheaper than the beef it’s mimicking, and it’s what the plant-based argument actually supports. If you’re eating less meat for your health, this is the move. It always was.
However, I’m never going to say that’s the best approach. Instead, let’s not kid ourselves, beef from a local source beats beans in almost every meaningful way. After that, good labels from a grocery store is next, and miles down underneath that, at the bottom of the barrel, has got to be actual meatless meat.
One thing nobody mentions: the plant-based option usually costs more per pound than the beef it’s replacing. Not one of the articles ranking for this question says so out loud. Worth knowing before you pay a premium for the responsible choice.
I shop at Costco, and sometimes I buy things I’d rather not. Some weeks I do better than others.
You’re not a bad person for eating a veggie burger, and anybody who wants you to feel like one is usually selling something. Read the ingredient list. Do your best, because that’s all any of us can do.
