The new science on what ultra-processed food does to the brain
wsj.com
wsj.com
> we introduce GroceryDB, a database with over 50,000 food items sold by Walmart, Target, and Wholefoods, unveiling the degree of processing characterizing each food. GroceryDB indicates that 73% of the US food supply is ultra-processed, and on average ultra-processed foods are 52% cheaper than minimally-processed alternatives. We find that the nutritional choices of the consumers, translated as the degree of food processing, strongly depend on the food categories and grocery stores. We show that there is no single nutrient or ingredient "bio-marker" for ultra-processed food, allowing us to quantify the individual contribution of over 1,000 ingredients to ultra-processing. GroceryDB and the associated https://www.TrueFood.tech website make this information available, aiming to simultaneously empower consumers to make healthy food choices, and aid policy makers to reform the food supply.
Machine Learning Prediction of Food Processing (2021), https://www.medrxiv.org/content/10.1101/2021.05.22.21257615
GroceryDB: Prevalence of Processed Food in Grocery Stores (2022), https://www.medrxiv.org/content/10.1101/2022.04.23.22274217
I looked up "plain hummus" and it has a score of 70? https://www.truefood.tech/product/wf_cedars-mediterranean-fo...
But the ingredient list shows almost zero additives. The same goes for a lot of other hummuses, but then some of them have a score of 1 despite having ingredient lists that are not discernably different. https://www.truefood.tech/product/wf_just-hummus-original-hu...
Idk. Maybe I'm missing something here.
The website goes on to say:
> In this example the Organic Plain Yogurt is at the 4th percentile among all yogurts, meaning if we have 100 different yogurts on a shelf then 95 of them are more processed than the Organic Plain Yogurt.
So, the 70 just means only 30% of all hummus are more processed than this one? But, does that really tell me anything about how processed the hummus is objectively? What if all hummus is extremely highly processed, how would this site indicate that? A 1 could still be highly processed in a category that is generally ultra-processed. :/
Almost zero != zero. Your cited hummus with a score of 70 includes the additive "guar gum". Other hummus vendors with a score of 1 do not include that additive, so the simple numerical score succeeds in flagging the product for manual human review.
Washington Post (2022) article lists guar gum as an ultra-processed additive, https://archive.is/LG1V1
Many ultra-processed foods are made in industrial machines that subject grains, corn and other raw ingredients to extremely high pressures and temperatures. This can destroy micronutrients and create new compounds that can be harmful, including carcinogens, said Carlos A. Monteiro, an expert on ultra-processed foods and a professor of nutrition and public health at the School of Public Health at the University of Sao Paulo in Brazil.
“These foods contain many chemical compounds that are not nutrients,” he added.
Ultra-processed foods often contain an array of additives whose effects on our health we don’t yet fully understand, said Mozaffarian. “It’s not just the salt and sugar, which are the obvious ones, but the artificial sweeteners, artificial colors, emulsifiers, stabilizers, guar gum and xanthan gum,” he said. “We don’t know that they’re innocuous.”Maybe it's fine! Your personal level of concern may vary, but if it's anything but zero, then surely it's more worthy of concern than hummus without guar gum.
Cooking routinely adds things for no nutritional benefit but because it changes the texture or taste to something better. Adding some cinnamon to oatmeal or a red pepper flakes to a pasta sauce isn't going to change the nutritional benefits by any large margin, but I do it because it helps me enjoy the food more. You don't necessarily add egg to cake mix for the nutritional benefit the egg provides you do it so its springier and has a better texture. One might add a little guar gum to hummus to make it thicker and have a slightly different mouth feel which some might prefer.
What exactly about guar gum is bad?
https://pubmed.ncbi.nlm.nih.gov/28322183/
Also, I never meant to imply there were conclusive negative effects of guar gum. I don't know! I do know it's an entirely unnecessary ingredient in hummus, so why would you want something that unneeded? Glass beads are probably more inert, but I still wouldn't eat them if I had a choice.
Hummus itself is entirely unnecessary for humans to consume. Why would I even consider eating it? Maybe because it tastes good and has a texture people enjoy? Maybe guar gum improves the texture to the food? I don't know, I'm not really a big fan of hummus, so I can't speak to whether or not guar gum makes it better or worse. But I imagine the people selling it determined people tend to enjoy the hummus with the guar gum more. They didn't just include it in there because they're an evil megacorporation out to control the world by adulterating hummus with guar gum, they put it in there because in the end they thought they'd sell more hummus.
If the yardstick of what we should eat is based entirely of if its "necessary", most things people generally consider fine wouldn't make the cut. Once again, its not necessary for me to put cinnamon in my oatmeal, should I stop doing that? Its not necessary for me to put red pepper flakes in my pasta sauce, should I stop doing that?
You can make a cake without egg in it, its not "necessary". It won't be the same as a cake with egg in it, but it can still be something sweet, kind of bready, and edible. But I'll continue to put egg in most of my cakes even though its not strictly "necessary" to make something edible.
From your article trying to define ultra-processed:
> The term “ultra-processed” was coined to refer to industrial formulations manufactured from substances derived from foods or synthesized from other organic sources. They typically contain little or no whole foods, are ready-to-consume or heat up, and are fatty, salty or sugary and depleted in dietary fibre, protein, various micronutrients and other bioactive compounds.
Isn't pre-roasted, pre-ground coffee beans made in a factory "industrial formulations manufactured from substances derived from foods"? They're not whole foods anymore, they're ready to consume or heat up, you're brewing it to get the fats out of it, depleted of the original fiber, doesn't have anywhere near the protein content of the original bean, and roasting destroys a lot of the original micronutrients of the original bean. Sure seems to me roasted ground coffee beans meet the definition of ultra-processed laid out by your source.
Instead of debating abstract semantic categories like "processed" and using random items to support or negate specific claims, we can now aggregate data directly against digital identifiers for physical items.
GroceryDB is a research baby step that illustrates what can be done to augment consumer decision making. The processed-food industry can create their own ICantBelieveItsNotProcessedDB to annotate products and ingredients with the benefits of industrial processing. Over time, food consumers can make local purchasing choices after human review of available data on food chains and health consequences.
It's applying an index to things without even really explaining why or what that index really means. What are the actual health impacts of hummus with guar gum versus hummus without it? It's ultimately telling us next to nothing other than highlighting some ingredients the maintainers don't like while ignoring ingredients they do like, seemingly very arbitrarily.
Does the index define roasted and ground coffee as a highly processed food? Why or why not?
From the Washington Post article above, you can add these names:
Kevin Hall, Scientist
National Institute of Diabetes and Digestive and Kidney Diseases
Dariush Mozaffarian, Cardiologist
Dean, Friedman School of Nutrition Science and Policy
Tufts University
Carlos A. Monteiro, Professor of Nutrition and Public health
School of Public Health
University of Sao Paulo, BrazilOn the index, why is garlic powder a 37 while chili powder which contains rice concentrate (what is that?!) is a 1? Is garlic powder, which only lists garlic as an ingredient, 37x worse for me than chili powder? What I information am I actually gaining from this index? How is looking at this index informing me of making a healthy choice at all?
https://www.truefood.tech/product/wf_365-by-whole-foods-mark...
https://www.truefood.tech/product/wf_simply-organic-organic-...
Please do share indexes and professional sources that you find genuinely useful.
The point being if it's extremely variable it /will/ be gamed and will work against you. The circular arrow symbol on plastics is a recent example where poor definition of what it meant has even led to consumers believing a symbol indicating non-recyclable means recyclable. A superior indicator for consumers that would get vastly less wrong is "plastic is not recyclable."
This or any index to provide information to consumers should be aggressively scrutinized. What it is today will not be what it becomes unless it can be defended, which requires thorough and widespread understanding.
If you do later form an opinion on this particular index, it may be more consequential to file a github issue against the index, rather than appending to a dormant HN thread.
Hummus is a dish made from chickpeas, tahini, oil, and garlic. Guar beans aren't a typical ingredient, but they're processed into power and added to give the dish some additional characteristic.
Coffee is a dish made from coffee. Coffee beans are ground up and processed because they're an essential ingredient.
Nobody has a cherished family high-fructose corn syrup cookie recipe, ha!
Neither me nor the article are suggesting that all processed foods are inherently bad and should be avoided at all cost. Guar gum is probably safer than coffee. What we are saying is that diets higher in processed foods tend to correlate with worse health, and the WSJ article goes further to suggest causes. Hummus is a terrible example... better might be McDonalds burgers vs those from a farm stand, or lattes vs black coffee. But hummus made from raw ingredients is also probably better for you than one bought from the store filled with preservatives.
Isn't chili powder highly processed in the same way guar gum powder is?
I'm not arguing guar gum is good in any way. For all I know it is massively worse for me than chili powder. Maybe it causes indigestion or irritates my gut. But why can't we use a label like that instead of just an ultra nebulous term of "ultra processed" where some people get to seemingly arbitrarily define what is and is not highly processed?
Where in any of those definitions listed does it suggest any idea of intent for if something is "highly processed"? If intent is what matters, is adding pineapple to pizza, something that should never be done, be considered the thing that makes it highly processed, or should pizza just always be considered highly processed from the start? After all, it's not whole foods, it's being subjected to high temperatures, contains highly processed foods like cheese, etc. Or is cheese somehow not highly processed either?
My guess is hummus would suffer similarly, but I'll wait for a connoisseur to confirm.
Guar gum is definitely an ultra-processed ingredient from what I can tell. It goes through multiple stages of industrial processing, and is used in everything from food products to cosmetics to hydraulic fracking.
It is also associated with esophageal and gastro-intestinal swelling, including to the point of causing obstruction. It is high in galactose and mannose, and subtly alters composition of the microbiome. People with GI disorders are very often advised to avoid it (note that although partially hydrolyzed guar gum was previously recommended for IBS to increase motility, in more recent years experts are increasingly saying to avoid it). It is categorized as "generally recognize safe" food additive in the US and EU.
I understand that ultra-processed foods often lead to being overweight or obese, and I can see how a whole host of problems can flow from that, from that condition of the body.
My question is: as a normal weight person, would these downsides still apply?
It does seem, skimming these articles, that the answer is 'yes' - ultra processed foods being associated with cancers and with a 22% higher chance of death from any cause, for example.
I'd be interested in more on those points, but this is helpful to know.
Edit: to be less accusational
Claiming a poster has a secret intent, is not constructive.
If processed foods(, or alcohol, or meat, or carbs, or whatever) have 25% increased negative outcomes for overweight people, but only 2% increased negative outcomes for non-overweight people, that significantly changes the calculus.
I don't understand how you don't understand why you'd want to know that. It's practically equivalent to avoiding peanuts because other people have peanut allergies.
Effectively they're predigested food molecules that are reconstituted. It would seem the very act of digestion has an important role to play in the regulation of hunger, circadian rhythm, and metabolism.
Normal weight people can still get disease, cancers, poor liver function, obesity, insulin resistance, hypertension.
Furthermore, ultra processed foods can contribute to mental illness and generally feeling like crap
Is it ketchup? Is it protein powder? Is it infant formula? Is it plant-based meat? Is it banana bread?
Is it based on the number of ingredients? Is it based on a preparation method?
This is the core problem with just about every article or study on UPFs
Right now all these studies seem to use broad strokes on the definition of ultra processed that make me think that it would be impossible to accurately remove confounding variables from the study.
https://world.openfoodfacts.org/nova https://en.wikipedia.org/wiki/Nova_classification
There doesn't need to be one formal globally recognised definition if you understand the intent of the meaning of the word.
Ketchup, protein powder, infant formula, plant-based meat and banana bread are all considered highly processed. Banana Bread less so if you decided to make it at home using raw ingredients.
The blog pointed to the very high effect size the experiments revealed (mostly mouse/rat model), and the impossibility to track it down to specific (groups of) substances. Unfortunately i cannot find the blog post.
To me it smells like microplastics might play a role.
The podcast mentions of a food classification system in the US called the NOVA system, which classifies foods in supermarkets based on the degree to which they have been processed based on the above scale.
Edit: a bit of googling turns up a number of results. E.g https://www.reddit.com/r/HubermanLab/comments/18qrjzl/lustig...
With that being said, I know he's a (purportedly) prolific Stanford researcher of all healthy living things...nutrition, sleep, supplements and so on. He knows what he is talking about from what I have listened to.
We live in a society...blah blah, make a dollar blah blah. I highly doubt he's going out of his way to get people on things they shouldn't be on or claiming a supp. is something it is not.
But if he is blatantly endorsing garbage/pushing bullocks claims, that'd be beyond highly unethical and astonishing.
Good on him for getting his bag, but the incentives are clear: he made his name before sponsorships with quality content, but now the demand for ad impressions has outpaced his supply of meaningful insights. He needs to keep churning out videos to earn sponsorship dollars, so it's a natural symbiotic choice to invite low quality guests.
They bring the content (and controversy / engagement / followers), he supplies the veneer of legitimacy, studio lighting and somber expression in the thumbnail. The advertisers benefit and all get their payday, as long as his adherents keep overpaying for processed powder with multivitamin.
No it's not evidence.
You wrote so many things but you also could not provide any evidence on what you said, product selling is not proof of anything. Please give proper evidence while accusing others of such and writing long comments.
Didn't he recently announce that he now even has equity in AG1? So, it even goes beyond "affiliate".
That said, if you listen to the actual wordings of the claims he makes about AG1, he very clearly makes it a subjective statement, as in, he himself is taking it, because it sounds to him like a good idea (and walks through his thinking).
He very clearly distinguishes the modality around that from the modality around the science he reviews and that sort of thing. At least, to a trained ear, the distinction comes across very clearly.
Stealing that one, thanks!
Important to note that whilst the podcast brings different opinions on different matters. It doesn't necessarily mean that Andrew or the podcast endorses those opinions.
If you are going to condemn processed foods, show me that it's worse to eat Quest Bars than fruit and nuts with honey.
I hold the problem is sugars (and alcohol).
- https://www.goodreads.com/en/book/show/62586003
The author of the above book has a good talk on the same topic here
- https://www.youtube.com/watch?v=5QOTBreQaIk
The book itself is well referenced with footnotes and an appendix with extra research papers etc.
>Unprocessed or minimally processed foods are whole foods in which the vitamins and nutrients are still intact. The food is in its natural (or nearly natural) state. These foods may be minimally altered by removal of inedible parts, drying, crushing, roasting, boiling, freezing, or pasteurization, to make them suitable to store and safe to consume. Unprocessed or minimally processed foods would include carrots, apples, raw chicken, melon, and raw, unsalted nuts.
>Processing changes a food from its natural state. Processed foods are essentially made by adding salt, oil, sugar, or other substances. Examples include canned fish or canned vegetables, fruits in syrup, and freshly made breads. Most processed foods have two or three ingredients.
>Some foods are highly processed or ultra-processed. They most likely have many added ingredients such as sugar, salt, fat, and artificial colors or preservatives. Ultra-processed foods are made mostly from substances extracted from foods, such as fats, starches, added sugars, and hydrogenated fats. They may also contain additives like artificial colors and flavors or stabilizers. Examples of these foods are frozen meals, soft drinks, hot dogs and cold cuts, fast food, packaged cookies, cakes, and salty snacks.
https://www.health.harvard.edu/blog/what-are-ultra-processed...
> California roll (加州巻き,カリフォルニアロール, kariforunia rōru) or California maki is an uramaki (inside-out makizushi roll) containing imitation crab (or rarely real crab), avocado, and cucumber. Sometimes crab salad is substituted for the crab stick, and often the outer layer of rice is sprinkled with toasted sesame seeds or roe (such as tobiko from flying fish).
> As one of the most popular styles of sushi in Canada and the United States, the California roll has been influential in sushi's global popularity, and in inspiring sushi chefs around the world to create non-traditional fusion cuisine.[2]
That seems misguided. Sushi is just, very loosely, "thing on vinegared rice". If you put unhealthy thing on rice, it's still an unhealthy thing. May as well say "sandwiches are a healthy food" or "salads are healthy food".
Humans have been curing meat forever. The question is whether "cured meat" is an issue or whether "modern curing of meat" is the problem.
When you ordered a ham from my old local meat market, it was a hunk of freshly butchered pig that was brined, cured and smoked. If it had any sugar, I sure couldn't taste it. You took it home, put it in the oven for a couple hours, and sliced it. It tasted wonderful by itself or with just a bit of fresh horseradish.
That ham is quite a lot different in terms of ingredients from the "sliced ham" you will get from your local grocery deli as a "cold cut".
Further attempts to refine the definition also bumps up against pendants like yourself regardless of what definition they use.
But just like porn, we all know what definitely falls into one category or the other. With foods that straddle definitions, just apply some common sense and moderation. Go ahead and eat your ham. If you share it with family and friends you’re probably ok. If you ate the whole thing yourself in a couple days, maybe that’s bad. Same for everything else. Eat them in appropriate quantities and you’ll be fine.
"Machine Learning Prediction of Food Processing" https://www.medrxiv.org/content/10.1101/2021.05.22.21257615v...
"Interestingly, even in presence of detailed ingredient information, the consistency across nutrition specialists in assigning NOVA classes was found to be low [39]"
I don't think bacon consumption has gone down since then.
So now I have stopped eating pastrami, and my reuben sandwiches are ruined, and I'm unhappy about that. They've evolved into an abomination involving chicken tikka and blue cheese, in an attempt to get some kind of taste out.
Nitrites are a bit of an economic factor, due to accelerating the curing process, and also help in mitigating risks around food safety. So they're primarily relevant to entities doing this sort of thing at industrial scale, who are subject to government standards on food safety. Home cooking to the rescue!
...on a related note, I don't really understand why small-scale organic/regional producers of aren't already doing this. At least where I am, in a rural part of Germany, we have farmers markets and other sales outlets that pride themselves on doing pretty much everything in the organic/regional playbook that could possibly set them apart from the mass-market, and who charge very high prices. Surely, they could do me the favour and just skip the nitrites? But, for some reason, they don't.
That's not completely correct if you don't cook the meat quickly afterward.
Apparently the safety issue is mostly botulism. Certain meats (especially sausages) create anaerobic conditions while curing that allows botulism to grow. Nitrites prevent that.
It is possible to cure meats without nitrates (see: Prosciutto di Parma), but it's not a straightforward, blanket thing.
Compare that to millions of cases each year, in the U.S., of foodborne illness, in general.
That's why I'm saying: It's a risk calculation, and government standards on food safety definitely prefer to err on the side of mandating too much processing of food over allowing too much foodborne illness.
Maybe, some of those risk calculations need to be revisited, and there needs to be a societal-level debate about how many cases of tummy-ache we are willing to trade off for how many basis points of cancer risk.
Incidentally, it stands to reason that the same kinds of things we do to kill off pathogens also kill off a lot of the microorganisms that may be useful to the body. To the extent that we're talking about additives, if we ingest them, then it seems likely that they might also kill off (or inhibit the growth of) useful parts of the microbiome already in the body. Of course, all of this I say in a mode of "cocktail party speculation" rather than peer-reviewed science, but there are just tons of unknown unknowns there, when it comes to the microbiome, and it actually offers one possible action mechanism (surely there are more than one), about how exactly it is that relying mostly on highly processed foods would be unhealthy for you.
Precisely because we have engineered our foods and processes for safety.
None of this is new: "When Every Ketchup But One Went Extinct" https://www.atlasobscura.com/articles/history-of-heinz-ketch...
"Their earlier ketchups “were medium bodied with average acidity,” Smith writes. In 1906, they replaced this ketchup with a shelf-stable, preservative-free ketchup that had twice as much salt, sugar, and vinegar as other commercial ketchups"
Is a ketchup with extensively boiled tomatoes (which kills quite a bit of their nutritive value), double the salt and double the sugar better than one with small amounts of benzoate?
And this has issues in more than just "highly processed". Listeria is also an issue.
Finding a deli that makes their own meat is practically impossible nowadays.
Unlike say glycemic index: high glycemic index > releases glucose fast > body can't compensate fast enough > bad (diabetes, poor satiety, etc.) right now it's just processed > ???? > bad (in unspecifiable ways, by unspecifiable means).
"Processing" food is literally what people do in their kitchens. But that's presumably not what they mean.
Certain kinds of processing have to be applied to food to make it safe for consumption. But that's presumably not what they mean either.
Certain kinds of processing make the nutritional contents of food more bioavailable. Not what they mean.
Certain kinds of processing are required to preserve food. If humans stopped doing it, their environmental footprint would be orders of magnitude higher than it now is. Presumably, that's not what anyone is suggesting who is opposed to processed foods.
Fermenting is a kind of processing, and fermented foods may be playing a role in a healthy microbiome. ...whoops. Again: Not what they are talking about.
So what exactly are they trying to say? ...surely, picking a word other than "processed" would help in clarifying.
However I would suggest researching about cirrhosis and fatty liver in children. It is a new disease that was unknown up until the 90s and is now common in developed countries. And it has to do with the following: your digestive system has the ability to process nutrients at a certain rate, depending on how hard they are to break down. For example sugar. The cells have a certain rate at which they can metabolize nutrients. The temporary excess is stored in the liver.
Then it becomes not a question of quality rather than quantity. The question then is not "how high is the glycemic index of food X" but "at which rate (i.e. mg of sugar per minute) can this food X be processed by our digestive system, and is this rate too fast for our cells to metabolize, which in turn causes our liver to store the temporary excess?"
Then the boolean variable "is this food X highly processed " becomes a plausible proxy variable, IMO, since refined sugar is something else than whole bread, when we talk about the rate at which nutrients enter the blood stream.
Do I make sense?
To put up a bit of a strawman: By what empirical measure, is an apple different (healthier) than an apple-equivalent of raw glucose/fructose + multi-vitamin + fiber supplement? The latter is obviously more processed, but I literally lack the means[1] to say how, why, and to what degree, one is 'healthier' than the other.
[0]Regardless of name, afaik there is no meaningful way to say e.g. kraft mac and cheese is '80% processed, where as homemade with box pasta, fresh cheddar and bechamel, is 50%, and homemade w/ hand-pulled noodles is 25%.
[1]e.g. a test of 'rate of which simple sugars leach in a ISO standardized simulated stomach'.
(Of course research into better defined categories and the underlying mechanisms of their health impact is also necessary and extremely valuable.)
However, I'm also wondering if we have any processing that makes food less digestible and that decreases it's surface area. If so, it would probably be interesting to compare this kind of processing with the other.
What you call dumbing down is in line with how science works. Rough and inaccurate models first, refined in time or on demand.
Having said that, I think all communication is inescapably slightly manipulative, starting from the mere fact of grabbing the audience's attention. But that doesn't mean we should double down on it.
So, there are tons of people who have rigid schemas, and are actually quite dumb, and watching health food nuts on YouTube is actually a great way to observe them in the wild. If you tell them "don't eat anything that doesn't look the way it looks in nature", they are going to start eating raw soy beans and poison themselves, because "science said so".
I don't think it's doing anyone any good if otherwise reputable news outlets, and actual scientists, throw even more oil into the fire of inaccurate information about nutrition.
* The Conversation (takes a stronger focus on the paper's methodology): https://theconversation.com/ultra-processed-foods-largest-ev...
* Medical News Today (also takes a more technical perspective, though includes more emotionally charged words in the opening): https://www.medicalnewstoday.com/articles/ultra-processed-fo...
* The Guardian (general news coverage) https://www.theguardian.com/society/2024/feb/28/ultra-proces...
Anyone got a more advanced trick on WSJ link?
I think it is well understood by interested people that indulging in cream-filled chocolate snack cakes will lead to bad outcomes.
But what is your beef with pork shoulders?
An interesting thing that I've noticed, is that pork fats seem to vary in their solidifying temperature. If I get pork from the supermarket, the fat resolidifies at room temperature rather quickly. If I get pork from a local farm, it can take all night to resolidify at room temperature. I'm thinking the quality of the pork feed may be affecting the health properties of the fat (higher saturated fat tends to equal higher solidity). Obviously it's still bad if one overeats it, but maybe we could get some health benefits from looking at the quality of the fats through things like animal feeds and living condidtions.
Not sure if we have any processing that makes food less digestible and decreases surface area but it would be interesting to compare this kind of processing with the other.
No, frozen and dried stuff are fine. I don't get the obsession with freshness when talking about healthy food.