Hacker News Reader

Depression impairs motivation, decreases tolerance of physical discomfort, and is also associated with impairment of planning abilities, and also has a correlation with eating disorders.

Occam's razor: the depressed can't plan and then cook a good meal for themselves, and so buy ready-to-eat processed food.

Anecdotally, when I was younger and very poor, I could not eat well unless I did everything from scratch -- several hours of work a day in the kitchen. When I got depressed, I'd try to cook while hungry, and get more and more agitated at my discomfort until I lost emotional control, and nothing got cooked. Cry. Eat instant noodles or crackers. Repeat the next day. When I finally made or got my hands on a proper hot meal with protein, my mood would noticeably improve.

My mood is still very susceptible to my diet, but these days I can afford stop-gap solutions to having no energy to cook, like just getting some reasonably healthy takeout.

Oh well in that case yeah throw out any concerns or criticisms. `retrac`'s anecdote and invocation of the law of Occam's Razor is proof that the study is iron clad.

Or realize that on such a painful and personal subject that some people will share their personal anecdotes—totally unscientific, oh my!

Also impulse control and relationship to instant gratification and addictive behaviors.

Processed foods are formulated to induce addictive behavior.

"We used 2 definitions for depression:

(1) a strict definition requiring self-reported clinician–diagnosed depression and regular antidepressant use and

(2) a broad definition requiring clinical diagnosis and/or antidepressant use.

We estimated hazard ratios (HRs) and 95% CIs for depression according to quintiles of UPF intake using Cox proportional hazards models, with adjustment for known and suspected risk factors for depression, including age, total caloric intake, body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), physical activity, smoking status, menopausal hormone therapy, total energy intake, alcohol, comorbidities (eg, diabetes, hypertension, dyslipidemia), median family income, social network levels, marital status, sleep duration, and pain."

https://jamanetwork.com/journals/jamanetworkopen/fullarticle...

I know that when my mental health starts to become a problem again the first thing that goes is any desire to cook. Which often leads to gettin take out, frozen meals, or just constant snacking.

I have had basically zero desire to cook (or really do much of anything in complete honesty) for the last couple weeks and I am super thankful for "processed" things like Huel.

Edit: Also I am 100% going to reach for that frozen meal from Trader joes that gives me a brief amount of joy and makes me feel good.

Hope you feel better soon :)

But this shows that theres no eating processed food --> get depressed causation, just a correlation. Otherwise a perfectly happy person who eats a lot of tv dinners would inexplicably get sad with nothing else in their life changing.

Thank you!

I have an appointment with my psychiatrist tomorrow! Yay mental health.

Exactly, I don't want to reject the study personally. Mental health is a complex beast and it was worthwhile to look at, but there is also a lot of stigma around processed foods that I think make something like this very complicated and it doesn't address why people may use the processed food in the first place.

> I have had basically zero desire to cook (or really do much of anything in complete honesty) for the last couple weeks and I am super thankful for "processed" things like Huel.

I'd never heard of Huel. Just looking at the info on their bag of pasta bolognese:

* bag size is 721 grams but serving size is 101g!

* they put the "per day" measurement at 505g? But even there, you're be consuming 3200mg of sodium-- 139% of the daily recommended value

Honestly, eating a 101g serving sounds to me like attempting to eat 1/8 of a bag of M&Ms.

If I open that 721g bag for lunch, it's going to be gone by end of day, guaranteed. (Probably with half a baguette for dipping.)

Edit: I forgot it's for lunch and dinner in my example. So make that a whole baguette. :)

Yeah so I don't do the Huel hot meals, I just never liked them.

I do the Huel Black which is a meal replacement shake with extra protein. Which if I did 5 servings of would be about 2,250mg of sodium a day. However personally I do it 3 times a day (the extra protein helps) and then a couple snacks like an apple or something else throughout the day.

However I will say regarding eating the entire bag in a day. It is fairly dense and filling (even though I didn't like the taste). So I would be surprised if you would finish it in a day. (then again it is only 7 meals to a bag compared to the 17 meals in a bag of Huel Black so maybe I am remembering wrong)

> I also don't see any controls for other circumstances, so it can also be explained that other factors could contribute both to eating processed foods and getting depressed (ex: working long hours, having low income, relationship stress)

Of course there were controls.

Directly from the article: “Adjusting for other health, lifestyle and socioeconomic risk factors for depression…”

As noted in another comment, the paper goes into further details on the controls they used.

Honestly, this kind of thing is insulting. If you’re going to levy that kind of accusation of incompetence against someone, you at least owe it to them to read the source material before doing so.

It’s good of you make this correction, we need to be accurate when discussing the research at hand.

That said, it’s worth pointing out that “adjusting for” a factor isn’t quite as effective as it sounds: https://dynomight.net/control/

Having built some regression models in my time, throwing a ton of extra variables in doesn’t fill me with confidence that I am “controlling” much of anything.

Following from the above, I'll presume you've got a specific methodological critique based on the process detailed in the paper?

The paper in question: https://jamanetwork.com/journals/jamanetworkopen/fullarticle...

It’s a research letter, which is peer-reviewed by JAMA but shorter than a traditional, full-fledged research article.

The full text of the letter’s description of the adjustments:

> with adjustment for known and suspected risk factors for depression, including

[followed by a list of over a dozen factors]

The data sharing supplement says that the code is on GitHub (hmm!), but doesn’t include a link (oh…).

That’s all I have to work with. I can’t make a specific criticism of the adjustments in this study. I’m left with my preconceived notions of how this kind of research is often done, even with peer-review.

When I read the words “we controlled for X in our population study of Y”, I usually take the lazy shortcut of not updating my internal model of the relationship between X and Y at all, because I’ve fooled myself too many times before. If they mention a natural experiment or instrumental variable I get excited.

Researchers love dunking on artificial sweeteners. It is one of the guaranteed methods of getting loads of press. Press yields prestige. Given that artificial sweeteners often correlate with people with weight / health problems like diabetes for very functional reasons, it's like shooting fish in a barrel.

Paradoxically this same data source (the Nurses' Health study, which is a continuous series of questionnaires) has had a prior paper claiming "sugar-sweetened soft drinks, refined grains, and red meat" led to depression. Now that "ultra-processed foods" and artificial sweeteners are the public demon, they take centre stage for the next round. I suspect with a bit of p-hacking one can contrive whatever aha result they desire out of it.

Their argument regarding eliminating correlation seems suspect. They analyzed diet in period 1 against reported depression in period 1 and found a given correlation. They then analyzed diet in period 1 against reported depression in period 2 (apparently 4 years later) and claim to have found the same correlation, which is what he cites in defence. Yet they never state that depression in period 2 in those cases is worse than period 1, invalidating it.

The Nurses' Health Study is a really fascinating exercise and certainly holds massive use, but it also has been a source of a lot of incredibly dubious nutritional "science".

https://nurseshealthstudy.org/participants/questionnaires

The NHS II surveys were the ones they used for this "get press" study, and looking at the actual surveys it seems doubtful that they yielded the results they did.

Probably could build a lot of fun blog entries p-hacking the data from these surveys.

There is a lot of evidence that artificial^W non-caloric sweeteners are terrible.

There are correlation studies in humans showing many different health issues (weight gain, cancer, mental health issues, among others) that correct for all sorts of confounding factors.

There are experimental studies in rodents that show all of the above across multiple products.

They're known to interact poorly with certain drugs taken for mental illness, and can cause all sorts of brain-related problems in some people (such as migraines).

They're also known to screw up your gut's microbial community, and that's shown to cause depression.

(I'm not commenting on the quality of this particular study, to be clear.)

>There is a lot of evidence that artificial^W non-caloric sweeteners are terrible.

There is shockingly little good evidence showing this, and these are many of the most studied substances in human history. There are a lot of very flawed studies that got enormous press and attention, however, yielding the "overwhelming proof" illusion.

This argument was made (practically verbatim) to defend tobacco, asbestos, and fossil fuels. It's easy to fund and publish a sloppy study that shows no correlation, and then claim the findings in the literature are "contradictory".

The studies I'm referring to go back to the 1970's. I've seen many that show strong negative effects, and some that show no or inconclusive effects. None show positive effects.

At this point, if the negative effect studies are statistical anomalies, then there should also be a roughly equal number of studies showing positive effects too.

>This argument was made (practically verbatim) to defend tobacco, asbestos, and fossil fuels

Sure it was. Indeed, your argument is practically verbatim the rhetoric of antivaxxers, flat Earthers and climate change deniers.

As to your other point, what "positive effect" are you demanding? Controlling blood sugar and controlling calories both have an *enormous*, overwhelming volume of evidence. Sugar substitutes are just tangentially related to that effort.

Sugar substitutes lead to metabolic issues that outweigh the reduction of calories eaten. If you also eat food that contains sugar, then they cravings they cause lead to blood sugar spikes later in the day. This is all settled science.

Anyway, I'd love to get some links to the scientific studies that the flat Earthers have saying the earth might be flat, after all.

Also, I don't think the anti-vaxxer's citation of one retracted study about mercury poisoning is really comparable to the dozens of non-retracted studies showing negative effects of artificial sweeteners across multiple mammals (including humans), or the work showing the chain of causal relationships that lead to metabolic issues.

Similarly, climate change has been scientific consensus since at least 1980, and I'm arguing on the side of scientific consensus, not against it.

The sweeteners cause incorrect hunger satiation signaling (as does HFCS), and also screw up your gut microbiome. The incorrect signaling makes you hungrier after eating the sweetener. If you eat at that point, then there goes the caloric intake benefits.

If you don't eat at that point, your body reacts by slowing down your metabolism, conserving more calories than you saved from avoiding sugar and also discouraging you from exercising. Over time, the lack of exercise and artificially lowered metabolism leads to a lower base metabolism, causing long term weight gain.

This doesn't touch on the problems a disrupted gut microbiome causes, known side-effects / intolerances / drug interactions that various products cause, or that some of them are carcinogens.

Making hugely contentious claims and then stating that it's "settled science" is not a convincing tactic. Again, you are using the anti-science tactic of saying "this one thing kind of said this, and that's what I've selected as my position, ergo it is The Truth". Science doesn't remotely work like that. Claiming you are on the side of "scientific consensus" is so cosmically wrong you're either lying, or staggeringly ill-informed.

There were HCQ and Ivermectin fanatics who were sure it was "settled science" because they read only what they wanted to see, and only accepted results they wanted to be true.

"f you don't eat at that point, your body reacts by slowing down your metabolism, conserving more calories than you saved from avoiding sugar and also discouraging you from exercising."

Ah, the "what's even the point?" tactic, which is the fallback of lazy dieters everywhere. Their metabolism "slowed" and magically the conservation of energy no longer applies.

What if the deep subconscious keeps track of sweetness in the mouth and nutritional requirements (conscience hunger and cravings)? Artificial sweeteners would break the biological programming and cause all sorts of problems including depression and inappropriate hunger.

Edit: I hadn’t fully thought this through, but in my experience sometimes looking at a clock causes the physical hunger sensation. I have also been painfully starving which immediately stopped after the first swallow of a chugging a coke. Artificial sweeteners would have thrown a monkey wrench in my biological machinery.

I wonder if depression simply starts much earlier than diagnostic controls detect it.

Even though this study started years before these people were diagnosed, they may have already had depression.

The counterargument by Chan misses Curtis' point - people with higher RISK of developing depression. Depression goes hand in hand with high trait neuroticism, and both the internalizing and externalizing branches of neuroticism contain subtraits related to lower self-control (though we're mostly with the internalizing one here, I'd imagine)

The correlation might still be useful as identifying how society is failing both in providing (tasty, easy, cost equivalent or better, healthy) alternative foods and how rampant a problem this is.