[lolinder]

You're obviously right that processed foods are far worse, but the study the article cites[0] doesn't say what the article says it does, not at all. The article deliberately misconstrues the phrase "weak evidence" to mean that the authors are attacking the past studies, when in fact they're just saying more research needs to be done. Here's the opening paragraph of the Discussion section:

> We evaluated the relationship between unprocessed red meat consumption and six selected disease outcomes following implementation of a meta-analytic approach. We found that unprocessed red meat intake had weak evidence of an association with increased risk of colorectal cancer, breast cancer, IHD and type 2 diabetes and no evidence of an association with ischemic stroke and hemorrhagic stroke. In other words, given all the data available on red meat intake and risk of a subsequent outcome, we estimate that consuming unprocessed red meat across an average range of exposure levels increases the risk of subsequent colorectal cancer, breast cancer, IHD and type 2 diabetes at least slightly compared to eating no red meat (by at least 6%, 3%, 1% and 1%, respectively).

[0] Full text here: https://www.nature.com/articles/s41591-022-01968-z#Sec2

[cogman10]

> You're obviously right that processed foods are far worse

I see this claim semi frequently online. So I'll just ask.

What is meant by "processed"? What does it mean to be "worse" (Worse than meat? Regular foods?) What is meant by "whole foods"? And finally, what's the evidence for these assertions?

Is black tea a processed food? How about coffee grounds? When does something cross the rubric of processed vs unprocessed?

To me, it's not obvious that "processed is bad" particularly because I don't know what "processed" means.

As a side, how do fortified foods fit in with processed foods? Iodine in salt, for example, has been a huge public health success. Goiters have practically been eliminated as a result.

[CadmiumYellow]

Every time processed foods come up online someone makes this argument. I will assume you're being genuine but a lot of times it's done in an annoying "gotcha" way and it drives me crazy. Regardless, the NOVA classification system is helpful for talking about this topic: https://regulatory.mxns.com/en/ultra-processed-foods-nova-cl...

When people use the term "processed foods" colloquially they are generally referring to NOVA group 4, ultra-processed foods. It seems that that group is implicated in worse health outcomes and it is also that group that has proliferated the most significantly over the past 40 years in tandem with rising rates of obesity and poor health outcomes.

[cogman10]

Wasn't really trying to gotcha. Just trying to get a good understanding on what's actually meant when someone says "processed foods are bad".

The term is quiet loose.

And I'd agree, that category does generally seem to be an issue. Though, I have to wonder if it's more a problem that that category typically has foods with an absurd amount of calories that are easy to consume.

[kerryritter]

The easiest way to understand unprocessed food is "nothing bad added, nothing good taken away".

For instance, pure peanut butter is unprocessed. While it runs through a mechanism to change it from peanuts to peanut butter, there is no oil added, nor are any of the healthful nutrients of peanut butter extracted. Opposite to this is peanut butter like JIF which is processed - the mix is diluted with sugar and vegetable oil to make the same amount of food for cheaper.

Same with tofu - it starts as soybeans and is ran through a mechanism to turn it into the blocks of tofu we see in stores, but we do not add sugar or oils to change the contents of that block, nor do we remove nutrients from the soybean (this may not be 100% accurate, but generally speaking, nutrients are not removed in this). However, many vegan meat products put a lot of unhealthy additions into the mix, thus making it processed.

[AstralStorm]

Generally we put a curdling agent into tofu, so it usually has more of either calcium or magnesium than unprocessed soybeans.

[type-r]

the easiest way to understand processed is has it been changed at all since being harvested? peanut butter and tofu are both processed. raw peanuts and legumes are not. cooked legumes are also technically processed. does not mean that it's always bad per se but that is the definition.

[badcppdev]

Is low-fat milk processed or unprocessed?

[Scalestein]

I think they go hand in hand to a degree. There aren't many "non-processed" (picked off a tree, animal products, etc.) foods which are as calorie dense as processed foods. Having access to something like twinkies or a powerbar makes it way easier to consume lots of calories.

[CadmiumYellow]

Exactly. Ultra-processed foods are a supernormal stimulus that drives an exaggerated response in humans, e.g. overeating.

[DanBC]

For "processed red meat" people mean red meat that has been preserved by smoking, salting, curing, or by adding preservatives.

[bluGill]

While you might mean the NOVA classification, the vast majority of people have never heard of that, much less read it. As such the original question still stands because most people have no idea that any definition exists, and in any case if what they mean happens to fit that, well that is just coincidence.

[tsimionescu]

The claim about processed foods has usually become very much accepted because of the failure to correlate any other aspect of food convincingly with the obesity epidemic. That is, there are populations much less obese than others eating all variants of macronutrient distributions (high fat, high carbohydrate, high meat, etc - you can find a low obesity population eating like that somewhere in the world). Processed foods are more or less the only remaining food-based explanation.

[cogman10]

This seems a bit like a non-sequitur to my comment.

Even if I accept that "Processed foods are more or less the only remaining food-based explanation." My questions really boil down to "what are processed foods" and "What's the evidence they are bad"?

[tsimionescu]

I was trying to answer the question "why do people think processed foods are bad, and in what way?". My answer was "because processed foods are the only remaining strictly diet-based possibility for a cause of obesity". That is, if we were to ascribe obesity exclusively to a particular diet choice, the only plausible diet choice is "processed foods" - all the others we've tried have clear counter-examples.

Note that I'm not saying we should do this - I believe obesity is much more likely caused by a wide array of lifestyle factors that no study to date accurately controls for (including diet, exercise, stress, environmental factors, mental status, medication etc), not a single dietary style.

Now, "processed foods" is indeed a very vague term, and people tend to include/exclude different foods based on their pet theory of what may lead to the link "observed" in the previous paragraph.

For example, some people strongly believe that glycemic spikes are strongly coupled with diabetes and obesity, and they would include things like fruit juices (as opposed to eating whole fruits) into this category, as well as high glycemic index foods such as bread (in both cases, even home-made ones).

Other people believe that certain additives are likely to have undocumented side-effects, so they will tend to only include foods with synthetic additives, such as preservatives and food colouring, but exclude traditional highly processed foods such as bread or butter as long as they are home-made without additives.

Yet other groups believe the correlation is related to palatability and/or satiety, so they will consider processed foods to be any foods which contain high ratios of palatable substances (like fat, sugar, salt, MSG, other flavour enhancers) to less palatable macro-nutrients. These people would probably include home-made fruit juice or ramen into the processed foods category, but may exclude things like pickles.

[cogman10]

This sort of gets to the heart of the problem I have with "processed foods are bad". It doesn't feel meaningful because of how vague it is. If I point at butter and ask "is this processed food" the answer is yes or no depending on the person I'm talking to. That makes it hard to trust the claim "processed food causes obesity". Well, maybe it does? Maybe it doesn't? Pickles don't likely cause obesity (they are very low calorie), so it feels unfair to lump them in the same category as cake, for example.

If it's unreasonable to pin down what is processed, it seems unreasonable to make a claim about the effects of processing.

[lobocinza]

In a way everything we eat is processed and processing is not necessarily harmful. To take your example of butter, organic butter may be completely different than mass produced butter regarding it's health implications. And this applies to nearly all food. Mass produced tomatoes will be poor in micro-nutrients and contaminated by pesticides. Et cetera.

Our understanding of nutrition and health isn't keeping up in pace with innovations in the food industry. And due to the scale and depth of it I think it's unavoidable that many small bad things will sum up over time leading to health problems without a clear root cause.

Reasons why "processed" food might be bad for health:

- skewed omega-3/omega-6 ratio

- micro-nutrients depletion

Also it's just not economical or environmentally sustainable to feed everyone with "real food". At least not without radical change and trade-offs in other areas.

[DanBC]

https://www.hsph.harvard.edu/nutritionsource/2015/11/03/repo...

> Processed meat – meat that has been transformed through salting, curing, fermentation, smoking, or other processes to enhance flavor or improve preservation

> Red meat – unprocessed mammalian muscle meat such as beef, veal, pork, lamb, mutton, horse and goat meat

> Consumption of processed meat was classified as carcinogenic and red meat as probably carcinogenic after the IARC Working Group – comprised of 22 scientists from ten countries – evaluated over 800 studies. Conclusions were primarily based on the evidence for colorectal cancer. Data also showed positive associations between processed meat consumption and stomach cancer, and between red meat consumption and pancreatic and prostate cancer.

When they talk about causing cancer they're talking about two things: the quality of the evidence, and then the strength of the effect.

For processed red meat the evidence is very strong: processed red meat does cause colorectal cancer.

But how much cancer does it cause? It doesn't appear to cause much cancer. If you have a genetic predisposition to colorectal cancer, or if you want to be very cautious, you might want to think about avoiding processed red meat. But otherwise, well, the small increase in risk might be worth it for the benefits you get from eating food you enjoy.

[CadmiumYellow]

It's quite hard to study this stuff but here's an example: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7946062/

The study includes an appendix with photos of all the meals - it's very interesting!

The evidence suggests that various qualities of ultra-processed foods drive overconsumption. There is also speculation that the calories in ultra-processed foods are more bioavailable than those in less processed foods because ultra-processed foods are easier to digest due to their low fiber content, though I don't have any evidence for than on hand (will update if I find it).

[cogman10]

I agree, that was an interesting read.

Definitely send more studies if you find them. This one seems to suggest the main problem with ultraprocessed foods is they are easier to over eat (they taste better? Are less filling?) but it doesn't really show that they are necessarily bad.

Seems to jive with how food works for me. I've been counting calories as of late and it's definitely easy to down a bag of doritos or drink a bottle of soda

[CadmiumYellow]

One thing I found super interesting was that the participants in that study rated both diets equally tasty, which points to some other reason why one diet was easier to overeat than the other! My guess would be that the ultra-processed diet is simultaneously more calorie dense and less satiating. It's true that that's not inherently bad, per se, but if we can prove that those factors cause weight gain and related health issues then I'd be pretty comfortable calling ultra-processed foods bad.

Have you heard of Stephan Guyenet? His book The Hungry Brain is a good read on the neuroscience of obesity. He proposes that eating highly palatable calorie-dense foods changes the brain in various ways that ultimately drive us to consume more of those foods and to have difficulty changing our eating behavior. Unfortunately his suggestion is to eat an intentionally bland diet, which is obviously unappealing, but my anecdotal experience is that highly palatable but less processed foods (e.g. a wheel of brie) are far more satiating than similar but ultra-processed foods (e.g. queso made from velveeta).

I'm not sold on the idea that we need to eat bland foods, or only foods familiar to our hunter-gatherer ancestors, or only plant products...it seems to me that we only need to go back 50 years or so to before the dramatic rise in obesity started, which points toward avoiding ultra-processed foods. I hope we get more studies on this eventually!

[vosper]

> What is meant by "processed"?

In the case of meat I think it's usually referring to added nitrates as a result of processing, which is linked to health problems [0].

> To me, it's not obvious that "processed is bad" particularly because I don't know what "processed" means.

Yeah I think your point is valid - "whole" is a bit like "natural". Too broad and undefined. We're supposed to think one is good and one is bad because marketing of "whole" and "natural" is trying to create that impression. No-one says "chemical" in their marketing, like no-one says "processed". You only advertise as "natural" or "whole".

[0] WebMD, but there's loads of articles about it https://www.webmd.com/diet/foods-high-in-nitrates

[edanm]

I agree with the criticism and hate when this term gets used, for the same reason. Most people just use "processed" to mean "bad" without thinking about it.

I think the closest you can get to a "real" definition is food that has added ingredients, usually oils or preservatives, not found in traditional cooking. Usually these are added more for reasons of shelf-life or in order to cram in extra taste, or both.

[pessimizer]

> by at least 6%, 3%, 1% and 1%

But this is very, very weak evidence.

edit: of course I have to assume something about the structure of the data to say it's very weak evidence. It may be very strong evidence of a very weak effect.

-----

edit 2: Also, how is this, the last "takeaway," listed in bold, large font directly beneath the headline, some sort of exaggeration or mischaracterization?

> They only found weak evidence that unprocessed red meat consumption is linked to colorectal cancer, breast cancer, type 2 diabetes, and ischemic heart disease, and no link at all between eating red meat and stroke.

[noptd]

>just saying more research needs to be done

Right, because the results of the previous studies amount to evidence that is too weak to conclude that red meat is inherently unhealthy to consume. I'm not sure how your comment contradicts this.

[lolinder]

"Weak evidence" is a term of art, not a pejorative. The original article frames this study as an attack on prior studies and vindication of red meat. My point is that this is extremely dishonest, and the authors of the study do not intend to convey that at all. They instead say that the evidence is there, but it's not yet strong enough for them to recommend any policy actions.

> The available evidence suggests that eating no unprocessed red meat may minimize the risk of disease incidence and mortality compared to consuming any, but there is insufficient evidence to make stronger or more conclusive recommendations. More rigorous, well-powered research is needed to better understand and quantify the relationship between unprocessed red meat intake and chronic disease.

[renewiltord]

Weak evidence of a risk increase from 4.3% to 4.6% of colorectal cancer? Okay, if that's where we end on red meat's cancer risk, I think I'll take it.

[notSupplied]

Sadly, news articles always grossly exaggerate research results.

"We found weak evidence of ..." becomes "New study ABSOLUTELY DESTROYS the myth that..."

[themitigating]

"The new study is climbing the ropes, oh my God what is he going to do, ohh he just jumped on the old studies, oh the humanity "