From anxiety to cancer, the evidence against ultra-processed food piles up | Hacker News Reader

From anxiety to cancer, the evidence against ultra-processed food piles up

text.npr.org

75 pointsrntn66 comments

I often find the definition of ultra processed difficult. I put pea protein powder in my smoothies. That is certainly ultra processed however it has good macro nutrients and is not empty calories.

I understand the concept of “empty calories”, but ultra processed seems like a blend of that and how the food was made.

> certainly ultra processed

Pardon, but where's the ultra processing? Isn't pea protein just dried peas in a blender?

I figured ultra processed was reserved for things like ascorbic acid, pectin, and xanthan gum.

Most people buy pea protein isolate. This is a more complex product where the protein has actually been separated from the remainder of the peas.

(Not sure if it would qualify as ultra processed though.)

You're right on the isolate.. Just found this video showing how they're separating the starch and fiber from pea protein.

https://youtu.be/wbX_w0ZIunM

You’re literally pulverizing the peas. That’s pretty extreme processing compared to, say, steaming some veggies

But your mouth basically does the same thing?

Smashing, not grinding. Not nearly as fine. Also, chemistry of food is very complicated... even if we stipulate that it's getting basically powder-ized in either case, if there's head involved in the process at any point it will behave very differently if it's pulverized before or after. (Think grinding flour vs chewing bread). I'll admit I'm now curious if plain, normal flour meets the definition of 'ultra processed'.

Yes, but then you spend more time chewing. And maybe that causes you eat less.

> Pardon, but where's the ultra processing? Isn't pea protein just dried peas in a blender?

Obviously not. Protein meal contains 20 to 25% protein, while pea protein concentrates/isolates have more than 80% protein. For more information on the process required to reach such high levels of protein, see https://en.wikipedia.org/wiki/Pea_protein

> I figured ultra processed was reserved for things like ascorbic acid, pectin, and xanthan gum.

You could have chosen better examples. All those three things are naturally occurring in non-processed foods, two of them are not even digestible by humans (both pectin and xanthan gum are technically "soluble fiber"), and the other one is a water-soluble vitamin (i.e., good luck overdosing or suffering from chronic exposure from that).

Yes, pretty much. They're made by a process called fractionation which is basically a mill (a blender if you like) and then clever air classification and centrifuge so that the constituent parts get separated by weight.

Yup, for pea protein concentrates obtained by dry fractionation (arguably a slighly more complicated processing process than "just dried peas in a blender").

>Ultra-processed foods (…) are not modified foods but formulations made mostly or entirely from substances derived from foods and additives, with little if any intact Group 1 food.

https://world.openfoodfacts.org/nova

https://www.cambridge.org/core/journals/public-health-nutrit...

There's obviously nothing inherently evil about "processing". If you put some fruit in a blender you've made no significant changes to its nutritional value.

The terminology is so annoying because it's hopelessly vague to the layperson. The actual research I've seen generally finds that the problem with "ultra processed" commercial foods are the usual familiar villains: salt, sugar, saturated fat.

Having put fruit in a blender you are changing how it’ll get absorbed, because the blender did the breaking down versus your body. You probably changed how nature of the fiber in the fruit.

Also blending means no more chewing, aka less saliva to help breaking down things. Most blenders warm up the mix because yeah high speed, which means less vitamin as well. And there may be more aspects, I'm no nutritionist...

also means you eat a lot faster and therefore probably more

This is true. Also, when was the last time you sat down and ate five oranges? Processing changes what you eat. Oranges are pretty good for you, orange juice often is not.

I read something saying blending in industrial food production affects absorption into the body which is thought to cause some of tut negative effects. So yes, the industrial production methods are also implicated.

It's a great example of correlation vs causation.

"Processed" or "ultra-processed" foods are often unhealthy, but they're not inherently unhealthy. What matters more than anything is the specific ingredients in them, like how much salt/sugar has been used to alter their taste or keep them preserved.

Consider the goal of these things. It's not to be scientific. It's to lead people make healthier choices and to lead decision makes to guide for healthier outcomes. Most people don't read labels. They don't even read articles. They see headlines.

"Ultra processed" is headline friendly and can steer people away from things.

This is health marketing for the masses.

If 'healthier choices' isn't based on science, what is it based on?

Why should someone believe you if it isn't based on science?

Many people (myself included) read health headlines, read that everything is causing cancer, that everything is simultaneously bad/good for you, and just tune this out altogether.

I just rely on normal intuition 'maintain a healthy calorie budget and keep greasy foods to a reasonable level' and treat all other nutrition advice as superstition, hearsay, and noise.

This meme, that the average American is stupid and needs to be told what to do about everything, is patronizing and needs to die.

The average 50th percentile IQ person has a finely tuned bullshit detector. When they sense they're being lead by an author or speaker toward making a certain decision, or feeling a certain way about the set of facts surrounding a current event, it makes them lose trust in the speaker, writer, and the institution they represent.

People have everyday experience with this, for example the moment they realize the person in front of them is shamelessly trying to sell them a car, timeshare, etc. and likely does not have their best interests in mind. You immediately stop believing a word this person is saying. Contrast this with, say, the guy behind the counter at your favorite hardware store who you trust is trying to get you the right solution for your needs.

The problem is when the speaker is an SME in a public service role. Trust in media, government, scientists, the academy, is low for good reason and it's causing huge problems in American society.

Most people are, in fact, not stupid. If you educate them on the facts, don't project false confidence about subjective or uncertain things, and basically talk to them like adults, we will have a more educated society with way more trust and tolerance for evolving science.

There is this paper, "Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake", Hall ea 2019 : https://www.sciencedirect.com/science/article/pii/S155041311...

They tried to control for all that, the same amount of sugar, fat, calories etc.

And yet people ate more of the ultra-processed food.

I'm sure "ultra processed" is a term concocted by the industry to cause a muddying of the waters leading to discussions of semantics over specifics.

The definition I've seen before is: If there are ingredients on the label that you cannot buy in a grocery store, it is ultra-processed.

Indeed... this term is about as useful as "chemical" or the general usage of the word "toxin". It can mean a lot of things and is generally ill-defined, but it somehow always manages to sound scary.

That's the weird thing about about this field, the definition for ultra processed they use is pretty arbitrary from (iirc) some Brazilian paper. And still, effects are found.

So it's really important what the mechanism is why e.g. it leads to higher calory intake.

I think it's something to do with mouth texture or chewiness that causes us to eat more of processed foods, but I've only read a small bit.

A helpful framing I once got from a nutritionist: "processed" food exists on a spectrum. Could you make it at home, or at least conceive of how to make it at home? (i.e. a can of soup, a baked cracker); or would it require a lab and advanced chemistry?

Off topic but I just switched to hemp protein. It contains omegas, and has a hugely positive ratio of potassium to sodium, whereas pea protein has no omegas and super high sodium (apparently to isolate pea protein it is washed in a salt brine). The drawback is that the protein per gram of powder is a bit lower.

A concern I've had with past studies has been the dosage of processed foods. It's better to avoid them altogether, but I haven't seen many seeking to find limits on how one can safely have X percent of their diet be from ultra-processed foods and still be healthy.

It seems like the biggest correlation between overall health and processed food intake is that processed foods are calorie dense and lead to weight gain. If excess calories are coming from sugars and unhealthy fats, and one is consuming large amounts of bad additives, that will surely lead to bad outcomes.

What I would like to see is isocaloric studies, where individuals get plenty of fiber, micronutrients and a good macronutritional balance from "healthy" foods, but are also allowed some reasonable percentage of their daily intake to be processed.

Maybe it's the guilt of feeling like I'm a weaker person for being addicted to having desserts or chips on a regular basis, but I also feel like they are pleasures of modern life that are worth enjoying in moderation. I mainly strive to eat healthy for all my core meals, and allow myself ultra-processed snacks, while not gaining weight.

There are guidelines for how much saturated fat, sugar, and other things to have in your diet per day (on average). These are based on many health studies over time.

I try to stay at or below the numbers on average and I do treat myself sometimes. But, I can't very much because it would lead to overeating things that would drive up my cholesterol and cause other bad things. Have to eat lots of fruit and vegetables.

A problem is that most people eating ultra processed foods blow those category numbers away. I was looking at the label on some food this week. One serving size, which is about half of what an American would usually eat, was over half the saturated fat and sugar someone should have per day. For people who eat processed food for most meals, it's easy to go many times these numbers which has a lot of long term consequences.

I have Crohn’s disease and while there’s still a lot that isn’t understood, it’s clear that it’s linked to highly processed food. Crohn’s is more prevalent in western populations and it’s rising in places that are adopting western diets. There are studies showing that patients who don’t see results from medication can still achieve remission through a strict Mediterranean diet.

Obesity is a huge problem and the spotlight on it is warranted, but it’s also a very simple issue to understand and deal with. The effects of ultra processed foods are more complex and far less understood.

I have IBD and definitely believe there is an association. But it still goes back to my question about "is there a safe limit". I think in particular for IBD, IBS, Crohn's etc, my suspicion is that there are two mechanisms at play with ultraprocessed diets:

1. Additives + microplastics in modern diets doing bad things to the microbiome. 2. Fiber intake is being reduced by ultaprocessed diets. I think the lack of pre+pro-biotics is a big deal, and ultraprocessed foods contribute to their reduced intake.

A good way to think of UPF is that it's "pre-chewed" food, due to the highly refined nature of it. The food matrix (structure) is destroyed, which fundamentally changes how your body metabolises it, bearing in mind this is a system evolved over millions of years to break down a mixture of whole foods in stages.

It's not even about the macronutrients - it's completely mismatched to our physiology. The nutrients are quickly absorbed due to the lack of structure so you get big blood sugar spikes and dips, making it hard to manage appetite (invariably leading to over-eating).

UPF is generally low fibre, so the microbiome is starved. It often contains emulsifiers which have the effect of a detergent in the gut lining, further damaging the microbiome.

It shouldn't be called "food". It's industrially produced pseudo-food with increasingly convincing evidence that it's contributing significantly to heart disease, diabetes, mental health issues, etc.

Genuine question: do you have any links you could share regarding the "emulsifiers as detergent" claim? I assume this mainly refers to chemical emulsifiers and not those that are naturally-occuring (like in eggs or garlic).

This is a great example of how bad nutrition data is, and why meta studies are to be treated with caution.

They took 45 studies, but 90% of them were either low or very low quality.

‘Overall, using the GRADE framework, 22 pooled analyses were rated as low quality, with 19 rated as very low quality and four rated as moderate quality.‘

How do these low-quality studies come about? Who is paying for them?

I can highly recommend the book Ultra-processed People by Chris van Tulleken about what UPF does (and doesn't) do with our bodies. And if you don't want to read the book, here's a ~1h video https://www.youtube.com/watch?v=5QOTBreQaIk talking about UPF. You don't even need to watch it, just listening is enough to make you think twice about buying many different kinds of food.

These foods tend to be consumed because of convenience (read: to save time).

But is it necessary to save time because they don't have time to spare (e.g., works 2+ jobs) and are under stress.

Along the same lines, if a person is time-challenged, then I would imagine odds are they don't have time to exercise either.

I avoid ultra-processed foods. But I'm curious if the lifestyle (?) such foods cater to might also contribute to the health issues (i.e., anxiety, cancer, etc.)

HN is really on a roll lately, with the ultra-processed food stuff. [1,2]

[1] https://news.ycombinator.com/item?id=39711527

[2] https://news.ycombinator.com/item?id=39675111

My correlation vs causation sense is kicking in here - people eating "ultra processed food" (whatever that means) in the U.S. are usually going to be lower income people who can't afford a fancy Whole Foods diet, or don't have the time to prepare those foods. People who are poor and short on time are probably also more likely to have stress/anxiety and poor health. Skimming through the article there's no mention of controlling for those factors.

Lastly, Amazon sponsors NPR and owns Whole Foods. No footnote mentioning the conflict of interest in this article? :)

Yea, the paper makes clear it's a associative meta-analysis, and the conclusion is basically "we should figure out what the mechanism is"

Plus, even if it is causal, it's not really new information: we know high-salt, high-calorie, nutrient-deficient foods (which ultraprocessed foods tend to be) DO causally drive the negative health outcomes described in the paper.

So like, this is basically a study that goes "yup, all the small studies are still saying this know-to-be-bad thing is still bad, despite recent interventions - let's go figure out if there's more mechanisms we missed here, or if it's just a matter of more interventions/individual choice at this point"

I disagree. We can only buy what's being offered. You joke about not knowing what ultra processed means, well, most consumers probably don't know orange juice from orange drink, the difference between chocolate candy and dark chocolate, etc. The question is, why is that stuff even available for sale if it is so bad?

Even whole foods sells a lot of ultra processed stuff, believe it or not. I wish I could go to a store and be confident that the bottle of orange stuff I pick up is just freshly squeezed orange juice and not... something else. That's just not the world we live in though.

Stressed people eat processed food. Stress causes health problems.

Processed food isn't the same as ultra processed. This is specifically about the NOVA classification of foods.

Yet the NOVA classifications are inherently contradictory.

"Processed foods" involve salt, sugar or oil added to unprocessed or minimally processed foods. Many premium ice creams are merely milk and cream (considered unprocessed or minimally processed), sugar and salt. Yet all ice creams are considered ultra-processed. Take that, ice cream!

Fruits in sugar syrup are considered "processed", yet sweetened juices are considered "ultra-processed". Huh?

Salted nuts are considered "processed" foods. Yet "salty packaged snacks" which would include salted nuts, are considered "ultra processed". Do I have to roast and salt the nuts myself? Where is the processing here? Is the package the problem?

Soups are considered "group 2" foods, which are unprocessed or minimally processed foods combined with oil, sugar or salt, but canned soups are considered ultra-processed. Yeah, OK.

Like, you can kind of see what they're going for, but when you start looking at all the edge cases, it devolves into lunacy. The actual mechanisms of ultra-processing need to be defined, not these examples which simply don't hold.

> Is the package the problem?

It very well might be.

McDonald’s is a solid part of my diet (mostly double cheese and sometimes chicken sandwich, never sodas). I’m in best health I’ve ever been.

The gold standard is large scale human randomized control trials (RCT). All of it shows that calorie balance (with sufficient protein), and strength training is all you need for a healthy system. Everything else is cope or people trying to sell you snake oil.

Go lift and count your macros. Do this for 30 days you’ll see night and day results if you’re just starting out.

> Go lift and count your macros. Do this for 30 days you’ll see night and day results if you’re just starting out.

You're talking about your muscles, which tells us nothing about symptoms such as cancer and anxiety. Your anecdote doesn't even approach the questions raised in the study.

>> count your macros

Hes also talking about a whole diet. Making sure that he gets everything he needs and in the right proportion. He's getting exercise.

This is miles away from "I only eat processed food" that some people have moved too. Some without even realizing how much processed food they are eating.

Nonetheless, you can count macros and lift to your heart's content but if your diet includes lead paint chips your physical and mental health will ultimately suffer. I'm not saying that ultraprocessed foods are as bad as lead paint chips, just that this line of reasoning falls short of addressing the concerns raised in studies on ultraprocessed foods.

I like how you picked lead paint and not doom scrolling on your phone while sitting on the couch instead off going out and getting exercise and being social.

God forbid our own industry hold up a mirror and admit we might be part of the problem.

It seems that you really want this to be about exercise. It isn't. Don't get me wrong: every minute doomscrolling on your couch will certainly take years from your life; somebody would have done the science but they died already. But, yet again, this is entirely tangential to the article.

The issue with RCTs is that you can only run them under tight conditions for so long on humans, so it's difficult to measure long-term outcomes like lifespan.

(Not to disagree with you on exercise. All data seems to point towards exercise being the first lever people should look to pull---as opposed to nutrition issues that people will have endless debates about---in terms of improving health.)

Keep it up for 20 years and let us know how it's working out for you.

Exactly, CICO seems like a great approach until you realize it means counting calories for the rest of your life. I think keto is the way for me. Eat to satiety, exercise, live life with an even energy level.

    git grep '#define' | wc -l

"evidence piles up" well of course it does. Who is paying for studies to show the benefits of processed foods?

The time individuals don't have to spend preparing their meals, the energy they don't buy and the fuel they don't burn to cook these items; surely those social and environmental factors should be added to the balance, when we weigh the costs and benefits?

Imagine someone does start studies to publicize the benefits of industrial foods. Imagine the NPR headlines about "Science for Sale" and how the industry is buying propaganda.

I guarantee the industrial-food-complex is paying scientists to identify potential benefits of ultra-processed food.

> I guarantee the industrial-food-complex is paying scientists to identify potential benefits of ultra-processed food.

I just don't think that this particular mode of "money corrupts science" is happening much. I mean the kind of thing where a big industry facing an image problem would say: Hey, let's bribe some scientists to do bogus studies, and use the "findings" as ammo in a public relations war.

Maybe that kind of thing was happening when the tobacco industry was lobbying against regulation. But nowadays, I'm having a hard time believing that. Because, nowadays, science is the weakest it's ever been in history (going back, maybe to the middle ages, where it might have been even worse), as a method of actually changing minds. So public relations wars would be fought differently, in a way that didn't involve science much.

Second, I think it doesn't jibe with the personality structures of most scientists, as far as their motivations go. They've already decided for themselves that money is not all that high on their hierarchy of motivations, otherwise they wouldn't be scientists.

What I think is happening instead is that money is an instrument that corrupts science in a direction of confirmation bias.

Essentially you have to ask yourself: What happens if scientist X tells us about their research on YouTube? Would that person be able to build a successful following? -- Even if that scientist despises YouTube, there is a ton of institutional pressure on scientists these days to communicate their work publicly, so to a first approximation, thinking of the "YouTube court of public opinion on science" to decide a scientists' career fate, is an approximation that's not too far off the mark.

"Ultra-processed foods are bad for you" is currently very much the way the wind is blowing. Any scientist who claimed otherwise would be seriously sticking their necks out with regard to the institutional context they're embedded in. Any scientist who goes an YouTube and tells people "my research shows that ultra-processed foods are bad for you and, oh by the way, here is a food product I really recommend" can become world-famous and rich. You do the math.

https://www.scientificamerican.com/article/food-industry-bac...

Companies like Coca Cola overplay the benefits of exercise as the solution to health problems so they can continue to sell ultra processed "food" with an extremely healthy profit margin.

UPF isn't food. It's "edible food-like substances" and it exists to make money, not to keep you in good health through your life.

It's why nearly every soda vending machine has a sticker like this:

"Balance what you EAT, DRINK, and DO"

https://www.bevindustry.com/articles/89678-aba-addresses-obe...

I'm not sure I believe that companies making ultra processed foods are victims of a catch-22 where they are unable to reveal the truth of how healthy their ultra processed food is because of the social shame that would come from funding a study.

That was an npr headline. https://www.npr.org/sections/thetwo-way/2016/09/13/493739074...

I can't identify your point. Studies that show harms/benefits of processed food are good, because they let us make decisions based on facts (as opposed to emotions). Of course there are tradeoffs involved.

Reply on news.ycombinator.com