Processed food followup - Ultra Processed Food

Consider a cocaine or smoking habit instead, it's measurably healthier

This post is inspired by Chris van Tulleken's Ultra Processed People and Mike Moss' Salt, Sugar, Fat.

This is one for the libertarians, and I consider myself to usually be among your number on most economic and social arguments.¹

So I got onto this processed food kick by reading more KD Hall studies - he started as a skeptic of “ultra-processed food” too, and then started banging out some awesome studies that basically changed his (and my) mind.

I went over the mechanisms by which processed foods make people fat, and one of KD Hall’s best studies on this front, in my first processed food post here.

The essential dynamic - Molochian optimization

You know how whenever companies put 10k phd’s and a strong optimization process on the other end of their product, it turns into an adversarial relationship with you, their customers’, long term happiness and well being?

Google going from two dudes who’ve genuinely revolutionized search to “sponsored results” and YouTube desperately scrabbling with clickbait headlines and fake reaction pics to keep you engaged for just one more click?

Facebook going from “hot or not” for Ivy students to Insta influencers driving girls to new heights of depression and algorithmic feeds that deliberately feed you an endless stream of polarized political hot takes, AI slop, and clickbait?

Sports gambling going from “you can bet on the game with the bookie down the street in that one bar” to mass advertised apps that literally ban anyone who knows what they’re doing, then milks the remaining rubes into bankruptcy?

Health insurance going from “we pool the risks together and that way if you get sick you’re taken care of” to algorithmic denials that deny 91% of claims automatically, and are so bad + profitable that the only customer feedback mechanism that gets anyone’s attention is assassination?²

Food is no exception.

Ultra processed food is another of these industries. The things being optimized are creating superstimuli with tastiness and "morishness" where people want more, because those are the things that sells more units, and selling more units is what the companies employing food scientists want and need to do.

It's not that they WANT to make their customers fat and unhealthy, that's just a side effect of the tastiness and morishness right now. When we get to a point we can legitimately create truly biologically-neutral taste-conveying "food," I'm sure they'll pivot to those.

In this case, it starts with making tasty stuff that people like (the halcyon days of M&M’s and Reese’s and Pringles³), and ends with everyone gulping down chemically disassembled and reconstituted industrial protein, fat, and carbohydrates infused with unregulated carcinogens and additives, and eaten in such massive quantities that 75% of people are overweight or obese.

What is ultra processed food?

And I’ve used the term “ultra-processed food” (henceforth UPF) a couple of times now, but what do we mean by that? Dr. Carlos Monteiro, the Brazilian doctor and researcher who came up with the NOVA 4-box food classification system, characterizes them as:

‘formulations of mostly cheap industrial sources of dietary energy and nutrients, plus additives, using a series of processes and containing minimal whole foods’.

Van Tulleken gives us another useful heuristic / definition:

“UPF has a long, formal scientific definition, but it can be boiled down to this: if it’s wrapped in plastic and has at least one ingredient that you wouldn’t usually find in a standard home kitchen, it’s UPF.”

And given that it encompasses junk food and fast food and pre-prepared food from nearly any source (including your grocery store’s freezers and most chain restaurants), UPF makes up the majority of people’s diets in America and the UK, with between 60-80% of average people’s calories coming from it (and it’s that high for children as well).

And “food” back there is very much in quotes for a reason: One thing our author, van Tulleken, makes very clear is that ultra processed food is less “food” than it is “industrially processed ingredients that are technically edible.”

What does ultra processing food consist of?

Essentially, ultra processed food takes in the cheapest possible raw feedstocks (usually plants, but sometimes meats or milk), disassembles them into their raw components (fat, carbohydrate, protein), refine, bleach, and deodorize those components, then reassembles them with additives, emulsifiers, natural and artificial flavors, and industrial processes, ultimately creating “food” products sold to consumers.

“Take palm oil, for example. When freshly pressed, it’s an almost luminous crimson, highly aromatic, spicy and flavourful, and full of antioxidants like palm tocotrienol. But, for UPF manufacturers, all that flavour and colour is a problem rather than an advantage. You can’t make Nutella with spicy red oil. Oil for UPF needs to be bland, plain and flavourless, so that it can be used to make any edible product – thus the use of RBD.” [Refine, bleach, deodorize]

“The long lists of fats (many of which have never previously been in a human diet) that you’ll start noticing on everything from biscuits to ice cream are the legacy of this technology for processing fat: shea fat, palm fat, mango kernel fat, palm stearate, coconut fat. Once they’ve been given the RBD treatment, they are essentially interchangeable.”

“[C]rops like corn and soy are turned into oil, protein and starch, which are then further modified. The oils are refined, bleached, deodorised, hydrogenated and interesterified, the protein may be hydrolysed, the starch modified. These modified food fractions are then combined with additives and assembled using industrial techniques like moulding, extrusion and pressure changes. This is a pattern I would encounter throughout my diet. Ingredient lists, from pizza to chew bars, started to look the same.”

One thing that’s pointed out to van Tulleken several times by the food scientists and other people he’s interacting with when learning all this - those industrial components - the bleached and deodorized and hydrolized and modified fats, proteins, and starches - are not something you would want to eat without a bunch of added artifical flavors and sugar.

“But the part of the discussion with Avena that stuck with me most was a casual aside about the food itself. Paul Hart had explained how most UPF is reconstructed from whole food that has been reduced to its basic molecular constituents which are then modified and re-assembled into food-like shapes and textures and then heavily salted, sweetened, coloured and flavoured. Avena speculated that without additives these base industrial ingredients would probably not be recognisable as food by your tongue and brain: ‘It would be almost like eating dirt.’ “

And this indeed is the advantage of this system! The fact that ultra processing enables companies to tune their “food” offerings to the utmost degree, in taste, texture, and moreishness. Because the competition in grocery store aisles is a furiously competitive Red Queen’s Race, fought by goliath companies with thousands of scientists to a side, immense marketing budgets, and cut-throat work cultures.

The Race is so furious, in fact, that most of the food companies have arrived at an equilibrium maximizing addictiveness in their foodstuffs, to ensure repeat customers, and to claw whatever market share they can from their competitors.

The “bliss point” and UPF addictiveness

This is done fairly scientifically. One thing Moss is good at is getting high-level insiders to talk to him and walk him through the process, and I really recommend his book if you’re interested in any inside baseball on how the industry works.

Essentially, both the large food companies and individual consultants who do the optimizations follow a process of getting many thousands of consumers to taste many and varied iterations of a given product, and then dial in the precise “bliss point” average that their target demographic most desires.

And it literally gets to the point of imaging people’s brains in real-time as they try different things, this isn’t necessarily your usual focus group with a bunch of people around a table, rating the last sample a 7 (although it was that in earlier and less optimized days).

“He conducted what he calls an optimization, which translates like this: selecting the best element from a set of available alternatives. “What I say is, let’s base it on science,” he said. “Let’s make twenty or thirty or forty variations. When you do that, you’ll see that we like some of the variations more and like others less. And you can build a mathematical model that shows you exactly the relation between what’s under your control and how consumers respond. Bingo. You engineer the product.”

As one example, for sweet foods, adults generally prefer ~12% sugar by weight, and children prefer ~24-36%. It gets much higher with children’s breakfast cereal, which generally range from 40-70% sugar by weight.

Salt, sugar, and fat are the three big guns when it comes to finding this bliss point. In brain scans, both sugar and fat light up reward centers like a Christmas tree. Interestingly, sugar caps out - it follows an inverted u-shaped curve where after a certain point, adding more decreases the liking of the food - that peak is where the bliss point will live. But fat has no such cap! You can keep adding fat to food, and the “like” will keep going up, and it’s a non-virtuous cycle too - the more fat you add, the more sugar and salt you can generally add to “balance” it, and the more sugar / salt something has, the more fat you can add without seeming fatty or oily.

This process continues until you’re basically at cheesecake, which is usually 18-25% sugar by weight, and 25-35% fat by weight (the remainder of the weight is water, with 5-10% protein, and ~1% salt by weight).

And don’t think they don’t try to optimize further than this! Because as many know, addictive drugs get more addictive the faster you can deliver the drug to the bloodstream / brain.

And indeed, this is explicitly being tweaked by the food scientists at these companies - from Moss:

“I would find out that one of the most compelling, and unsettling, aspects of the role of salt, sugar, and fat in processed foods is the way the industry, in an effort to boost their power, has sought to alter their physical shape and structure. Scientists at Nestlé are currently fiddling with the distribution and shape of fat globules to affect their absorption rate and, as it’s known in the industry, their “mouthfeel.” At Cargill, the world’s leading supplier of salt, scientists are altering the physical shape of salt, pulverizing it into a fine powder to hit the taste buds faster and harder, improving what the company calls its “flavor burst.”

“Sugar is being altered in myriad ways as well. The sweetest component of simple sugar, fructose, has been crystallized into an additive that boosts the allure of foods. Scientists have also created enhancers that amplify the sweetness of sugar to two hundred times its natural strength.”

Did you know Philip Morris and RJ Reynolds both bought UPF companies? General Foods and Kraft for PM, and Nestle for RJR, and they ran them quite profitably for a number of years. They eventually got out, because the business was so viciously cutthroat that even cigarette executives had reservations (at least according to some folk Moss talked to).

Overall, it’s a Molochian “optimize at all costs” dynamic going on for decades, and that degree of competition and optimization has driven UPF’s addictiveness higher and higher, with probably little actual malice involved, just a desire to maintain a market position and not lose money. For some of the inside baseball on specific battles waged, I highly recommend Moss’ work. The Secret Life of Groceries is another good one, that I’ll probably review in a few months.

I do believe that if food scientists could give you “tasty” and “moreishness” without obesity and health problems, they would be happy to do this - it’s just not possible at our current level of development.

But let’s continue - who are our authors?

For Ultra Processed People, Chris van Tulleken is an Oxford-educated British doctor and television presenter who’s appeared in documentaries and children’s shows. Of note, he has a twin and has taken part in various twin studies, and they’ve both appeared in various shows, documentaries, podcasts, and articles - he says that they are the twin pair with the greatest weight discrepancy of any twin pairs, with his brother Xand ~20kg heavier at his peak.

For what it’s worth, he is a man who understands the siren call of UPF, and you can tell this in his narrations in the book. It seems his twin gave in to these siren calls a little more often than he did.⁴

An “after” and a “before” shot of the van Tulleken brothers, Xand on the left, Chris on the right

For Salt, Sugar, Fat, Michael Moss is a Pulitzer winning (and many-time finalist) journalist, author, and public speaker, an adjunct professor at the Columbia business school, who also wrote a NYT best-seller about processed food from these big companies.

So one of the gimmicks van Tulleken does is he “eats like an average British child” for a month, eating such that ~80% of his calories come from ultra-processed food, with comprehensive health measures taken before and after.

Now, why is this a bad thing?

Well, at the end of van Tulleken’s mini “super size me” experiment, this was the result:

“When I went back to UCL for testing at the end of the diet, the results were spectacular. I had gained 6kg. If this rate had continued for a year, I would have nearly doubled my body weight. Additionally, my appetite hormones were totally deranged. The hormone that signals fullness barely responded to a large meal, while the hunger hormone was sky high just moments after eating. There was a fivefold increase in leptin, the hormone that comes from fat, while my levels of C-reactive protein, a marker that indicates inflammation, had doubled.”

They do a brain scan before and after and claim to see differences in reward center and hypothalamic signaling, but I don’t think they could really tell this with any rigor using the method they used (an MRI scan), so I’m not going to bother to include any quote.

It’s bad at a population level

But there are many signs that UPF is bad for people in the aggregate, overall. For one thing, even after controlling for fiber and macros, the top quartile of UPF eaters by percent of diet have a 1.3 - 1.7x all cause mortality Hazard Ratio compared to the bottom quartile, and the amount of UPF eaten follows a direct dose-reponse curve in terms of the incremental all cause mortality.⁵

“But wait,” you might say, “you can’t just control by macros, UPF is eaten by fat poor people, so they have many other negative health confounds!” And you’d be right!

Happily, many researchers thought of this. In fact, the 1.7x study (the SUN study in Spain) controlled for education level, SES, obesity, marital status, smoking status and historical smoking, age cohort, and a double handful of specific health conditions. Then because there could still be residual confounding, they did sensitivity analysis and calculated Vanderweele E values to get the minimum strength of association before calculating the final impacts.

A Swedish study with n=27k⁶ measured actual UPF metabolites in blood and found a 1.23x HR in all cause mortality for each additional sigma of UPF consumption.

They also controlled for BMI, marital status, smoking history and status, educational level, alcohol use, exercise, coffee drinking, familial health history, and more in this analysis. Their finding is in line with several other studies: “Our observed positive association with all-cause mortality was supported by a recent meta-analysis [40] and four subsequent cohort studies [17-19,21]”

Why might this be? What is actually driving the mortality if all this other stuff is controlled for?

We don’t really have direct physiological smoking guns. But I can explain why we don’t have them, and point you to an overall dynamic.

For one thing, just like we discovered in my Poorly Made in China review, not only do consumers have no idea what’s in the products you buy, often even the company selling the product itself has no idea, and the manufacturers also have no record of this (intentionally, to avoid liability).

Food is basically the same deal.

In Europe and the UK, they have some limitations on the additives you can put into food. For the US, there’s more than ten thousand additives that manufacturers can use, and there’s essentially zero regulation or policing of them.

You know how the FDA kills tens of thousands of people a year on the pharmaceutical side of things by having impossibly high hurdles that cost billions and a decade to surmount? And those end products are gate-kept by doctors with a lot of education and involve a patient whose history they know and who will follow up with them if anything bad happens.

What if I told you the FDA takes literally the exact opposite stance to the “Food” part of the FDA mandate? In the sense that nearly everything is “self-regulated” by companies affirming that they’re safe and then putting them in the food supply with essentially zero barriers, auditing, or policing. At which point the food is in stores, being eaten! There is ZERO gate-keeping, zero knowledge or limitation on who will be eating them (e.g. kids), and zero means to follow up if something bad happens (like, say, 75% of the US being overweight or obese).

It’s yet another reason to hate the FDA

Van Tulleken has a fun example in the book, let’s go through it. Corn Oil ONE, an Iowa company, voluntarily notifies the FDA that it intends to market a product called ‘COZ corn oil.’

So this corn oil is made in a novel way - the company mostly uses corn to create ethanol that goes into cars. But they have leftover corn mash after it’s made - this mash contains antibioitics, pesticides, solvents, and other additives, because all that stuff is left behind in the ethanol creating process. You can wash this with more solvents and extract some “distilled corn oil,” and then sell that for livestock feed. But Corn Oil ONE wants to process this oil a little more and feed it to humans at a higher price point than they can get as livestock feed.

So when a company has a new food additive, they can go three routes.

1. They can petition the FDA for a full review of the new additive and have it formally listed as a food additive. They’d have to submit a good amount of documentation, and this would probably take years.

2. Or they can try to get the additive declared as GRAS, or Generally Recognized As Safe. You send some data, whatever data you choose to send, and hope they have no followup questions and rubber stamp you.

3. Or they can literally just start putting the additive in whatever human food they feel like, after “self-determining” that it’s safe.⁷

Wow, I wonder why anyone does anything BUT option 3. This was definitely an unforced error on their part, trying to go the slightly harder route.

So first they go for option 2 - they send 81 pages to the FDA, claiming it’s safe based on two unpublished studies and 4 experts the company elected to collect opinions from. They half-ass this so much they have a molecular structure diagram for the corn oil - but corn oil doesn’t have a given molecular structure, it’s made of many different molecules. The diagram was actually the molecular formula for an HIV drug called Lopanavir. As von Tulleken puts it in the book “including the wrong molecular structure is a clue that this may not be a company with the sort of thorough, detailed approach that we’d hope when they are determining the safety of additives in our food.”

The company was also using a processing aid made by Dupont called FermaSure XL, which the FDA has turned down as explicitly NOT GRAS, and which the FDA objected to in a response to their request.

So they withdrew their petition and went for option 3, and just put it in human food directly.

“If the company that will make money from an ingredient disagrees with the FDA’s concerns, and it believes that its product is GRAS, then it can withdraw the FDA application and put the molecule in food anyway.”

This is completely legal as of 1997, with a reaffirmation in 2016 (largely due to regulatory capture), but had been going on since roughly the 60’s, and was just retroactively validated when the FDA looked at the backlog, shrugged, and essentially said: “eh, nobody’s died en masse yet…”

And once a company puts it in the human food supply, there is no policing, no auditing, and no way to know that you’re getting the “animal feed grade” corn oil full of pesticides and antibiotics and processed with explicitly non-GRAS chemicals. All that will ever be on the label is “corn oil.”

In theory, the FDA could go to the facility or company headquarters to investigate, but realistically, they have ~100 FTE regulating a multi-hundred-billion dollar industry with more than 10k additives, and were so backlogged from the 60’s on they came up with this whole “self-determined” safety regime. They’re not doing anything like that.

Maybe this is a one-off, a particularly egregious example? Well, in the sense that they made the unforced error of even trying for method 2 and leaving a data trail such that we know about it, yes.

But in the sense that most companies do this? Since 2000, there’s been only 10 method 1 or 2 approvals. There have been 766 new food chemicals since then, which means 756, or 98.7%, have been “self-determined” and put in our food supply with essentially zero vetting or oversight, or means of measuring impacts, or means of recourse if they’ve messed up, either via incompetence or malice.

The artifical flavor industry is the same thing, with about 2.6k different flavors self-determined to be GRAS, predominantly self regulated. But consider just one example, isoeugenol, which is self-certified as GRAS. The US National Toxicology Program decided to look into it, because it has a similar structure to other carcinogens. That study found clear evidence that it causes cancer - 80% of the mice got liver tumors, with a direct dose-dependent effect. FEMA, the flavor and extract industry manufacturers association, defended their GRAS certification by estimating a daily per capita intake two thousand times lower than existing WHO estimates and calling the study a “high-dose phenomenon without any relevance for assessing the potential cancer risk of the use of isoeugenol as a food flavor ingredient.”

And once again - do you, the consumer, have any remedy for any of this, even in theory? Nope. You’re eating ten thousand different self-certified things, you don’t know which one is making you fat or giving you cancer, and you’ll never be able to assemble a clear evidentiary trail back to any specific company.

I’ve never been prouder to be an American! Or happier that I’ve been an expat the past year or so. 😂

Remember the “weight homeostat?”

In my first processed food post, I talked about how the likely mechanism behind everyone getting fatter is a problem with damage to weight / appetite homeostats.

Well, these dynamics behind UPF, additives, and flavoring agents give us a fairly plausible smoking gun. UPF has intentionally been tuned by thousands of smart Phd’s to be more addictive. The lack of oversight and policing in additives and flavoring agents means it is extremely likely that more plastic, antibiotic, pesticide, and other pollutants are getting into those things, and in turn, our food.

The combination of these things is a good candidate for what’s wrecking people’s weight homeostats. They feel hungrier than they “should,” because they’ve been exposing themselves to more unknown-but-probably-negative quality additives and flavoring agents, and large amounts of energy-dense superstimuli food, and together these things have altered the mental and physical feedback cycles that usually keep people a steady weight.

And this was just the half of it - the book goes into specific emulsifiers, stabilizers, and modified starches, as well as some other individual additives that we know cause negative effects, I just try to keep these things capped at ~20 minutes read time, and this is already at 30 min.

If you want some horrifying additional light reading on this front, I highly recommend gut microbiome scientist Stephen Skolnick’s three part series on “the thousand secret ways that the food is poison.”

A brief plea for better FDA regulation

But to the libertarian point - I’m as “people should be able to put whatever they want into their own bodies” as it gets, but A) people don’t know - this is basically Upton Sinclair’s The Jungle level malfeasance, the FDA is clearly exposing the US populace to real risks by not doing their job, and B) this is like 60-80% of kids’ diets, and that’s not right either.

I’m a regulation skeptic, in the sense I think on average most additional regulation will net out to be a mistake, but in this particular case, there seems to be clear possible gains.

Do parents have a right to give their kids whiskey and cigarettes? Because as we’ll see, UPF might be worse than that. I feel like if I tried anything like that, they’d be CPS-d pretty fast. But then again I feel like there’s probably a distressing number of parents doing exactly that with no repercussions at all. Go go, CPS - as valuable as the FDA! Each murdering people on both ends with vastly inappropriate levels of over-and-under reaction!⁸

So what’s the problem with UPF overall?

1. They are products created by a fully adversarial system, optimizing on measures that are directly antagonistic to your health and well being.

2. They’re not food, but are industrially processed assemblies of ingredients that are technically edible, and which largely haven’t been evaluated either individually or cumulatively for safety or dangerous health impacts - which they affirmatively DO have, usually ranging from 1.3-1.7x HR for all cause mortality, even after controlling for essentially everything anyone can think of.

3. They’ve been literally engineered to be superstimuli and to be as addictive as possible, to drive higher consumption and more sales, because that’s what companies want / need to optimize towards.

Hot take: “fun” addictions are better choices than UPF

I would rather my (future) teenage kids take up a cocaine habit than a junk and fast-food habit. You can QUIT cocaine! You can quit smoking. Heck, you can quit drinking or even heroin. But you know what you can’t quit? Eating.

And if you alter your brain and get addicted to ultra processed junk and fast food, you will forever be bombarded with advertising, restaurants, product placement, and similarly addicted friends and family eating that stuff all the time around you. It’s as though a heroin addict were constantly being offered new and improved strains of heroin, seeing advertisements for it everywhere, being given free samples, and having friends offer to pick some up on the way over. You will never be free of it!

And because obesity is a permanent, one-way ratchet, if they ever get addicted, that’s pretty much it. Game over, that’s their life forever. Because “overweight or obese” comes with “sedentary” and when you package those two together, your all cause mortality shoots from 1.3-1.7x to more like 4-5x.

Cocaine is WAY better, by comparison. You’ll stay skinny, you’ll stay active, you can rack up some iconoclast / counterculture cred, it helps you think, it gives you energy, it can even help you relax! It’s pretty much win/win! And when you decide to quit, that’s it! It’s not valorized by anyone, not flaunted by celebrities, it’s not advertised everywhere you go - everyone hides it! Who do you know that’s totally open about their cocaine use? Nobody you admire, if you even do know anyone.

Even smoking is probably better. If you decide to quit, the whole of society hates and derides smokers, forever banishing them to cold and wet corners far from doorways to indulge their filthy habit. Celebrities hide it in shame, rather than just for the illegality. Remember Obama hiding his increased smoking while in office, and only owning up to how much he smoked retroactively, after he left? A much preferable addiction, either one.

You want to talk all cause mortality? If you compare regular exercisers with healthy diets to obese sedentary moderns, there’s a 4 - 5.5x all cause mortality difference.⁹

For smokers versus non-smokers, it’s only a 1.2 - 2.5x all cause mortality difference.¹⁰ For cocaine addicts, as long as they’re not injecting, it’s around 1.2 - 3x, too.¹¹

And cocaine is MUCH easier to quit succesfully than obesity!

After all, by the numbers, only 20% of coke users end up addicted.¹² But a full 75% of the US population is overweight or obese, and essentially every single one of them eats UPF regularly. A majority of obese people try to lose weight every single year,¹³ and they pretty much universally fail to achieve any substantive result - even the optimists say that 80% of people trying fail to lose and keep more than 10% of body weight off for at least a year,¹⁴ and 10% of body weight isn’t exactly going to move the needle significantly, either cosmetically or health wise,¹⁵ even if they were reliably achieving that amount of loss.

I mean obviously, it’s better to have NO addictions, UPF included, and that’s what I’m personally aiming at for the people in my life. I’m mainly hot-taking to show you the extreme degree of downsides and effects we’re talking about here, because it’s not even close, it’s massively better to have a legitimately bad addiction than to be obese and sedentary and eating UPF for most of your calories.

So what’s the solution again? Eat real foods that your great grandma would recognize, that you have to actually cook

Or per van Tulleken’s heuristic:

if it’s wrapped in plastic and has at least one ingredient that you wouldn’t usually find in a standard home kitchen, it’s UPF

So avoid everything that meets that criteria. And this is most restaurant food these days too, not just fast food - if you’re in a chain restaurant, it’s a very good bet they get the majority of their food from Sysco or US Foods and basically reheat it, and that food is predominantly UPF. If you eat in restaurants, eat high end, or eat where you know their supply chain and that they’re using real ingredients.

Everyone hates this answer, but it honestly seems like the best answer to me.

Ain’t nobody got time for that? Preach! I completely agree.

1. Trial “rigorously eating no UPF” for a month, and see if you feel and look any better - if it has no effect on you, there’s your answer, you do you.

2. Do meal prep and save 5-8 hours a week while eating healthier.

3. It’s a dose response thing, so if you mostly cut it out but still meet friends or family at a UPF restaurant occasionally, or just cut it down by 20%, that’s better for your health too.

4. You genuinely love it? That’s fine too, it’s a choice, like smoking, just be aware of the costs for you and your family members.

5. Replace it with a healthier addiction - and you have a LOT of leeway here, because even cocaine or smoking are healthier. But you know, exercise, dance parties, recreational sport, hiking, swimming, coffee, spicy food, socializing - all claimed as “addictions” by some people.

6. If you want to mitigate some of the health costs, exercise and / or get a treadmill desk to make sure you’re not sedentary, which basically doubles the risks (see the linked review or footnote 9 here for an idea of effect sizes).

1. Which I’ll briefly characterize as “vice laws are dumb, let people do whatever they want to themselves, but for god’s sake, make sure economic growth keeps happening, because that’s the biggest lever to actually alleviate poverty and improve lives.”

2. The assassinated United executive, Brian Thompson, was the sub-CEO of the division that was responsible for creating the system that generated algorithmic denials that denied 91% of cases, mostly erroneously.

Brian, a sub-CEO of one division of United, made $10m in 2023. Kim Keck, top-level CEO of all of BCBS, made $4m. That $6M delta came from SOMETHING profitable that his team drove, and that something profitable was increased denials. It's worth it to note that comp comes before margins - the vast delta in comp going to Brian and the other executives like him is part of why United "only" has 6-7% margins.

To save money, this CEO authorized denial of claims by AI which almost certainly denied claims that should have been authorized. In other words, his choices directly led to the deaths of thousands if not tens of thousands of people and brought him (and his shareholders) millions of dollars. More, this is part of a recurring trend of United having more in-network denials than any other company, so it’s basically in the company culture and DNA to routinely do this: https://imgur.com/Jn2pi2N

Yes, it’s not ideal to have vigilantes murdering whatever executives they take a dislike to, because the rule of law is important, but honestly, I think Brian more or less had it coming, and there was literally no legal way to provide this pointed customer feedback regarding United’s business practices in a way that would have affected anything at all.

I honestly wish there were MORE Luigi’s in this specific domain, so we could get real health reform in the US. Do you know if you buy insurance on the Exchange it’s literally impossible to get insurance that covers you in multiple states, at any price? Every available option leaves you uncovered in 98% of states, there is literally no legal way to get actual insurance for anyone that travels between states on the exchange. It’s monstrous, and just the tip of the iceberg.

3. Fun fact, did you know Gene Wolfe, famed sci-fi author of the Emperor of the New Sun trilogy, actually invented one of the machines and processes that make Pringles?

4. Although, somewhat surprisingly from somebody clearly writing an “advocacy” book, throughout the book he steadfastly maintains an “obesity is a disease and it’s literally impossible to control what you eat” take, which I found fairly original in a book that’s literally about persuading you to eat less UPF, and given that both he and his twin give up UPF in the course of things.

5. From: SJ Dicken, R Batterham, The Role of Diet Quality in Mediating the Association between Ultra-Processed Food Intake, Obesity and Health-Related Outcomes: A Review of Prospective Cohort Studies (2021)

n=~106k Adjusting for French dietary guideline adherence or for both French dietary guideline adherence and for Western dietary pattern still resulted in each 10% increment in UPF intake being associated with a 14% (1.04, 1.27) or 19% (1.05, 1.35) increased risk, respectively, of all-cause mortality.

n=105k Srour et al. reported a 12% (1.05, 1.20), 13% (1.02, 1.24) and 11% (1.01, 1.21) increased risk of CVD, coronary heart disease (CHD) and cerebrovascular disease, respectively, per 10% increase in UPF in the diet [117]. Multiple dietary adjustments did not alter these risk estimates.

n=105k Fiolet et al. reported a 12% (1.06, 1.18) and 11% (1.02, 1.22) increased risk of all cancer and breast cancer, respectively, per 10% increase in UPF in the diet [126]. Adjustment for lipids (including fat), sodium and carbohydrate intake had no impact on the risk of all cancer (HR: 1.12 (1.07, 1.18)) or breast cancer (HR: 1.11 (1.01, 1.21)) per 10% increase in UPF in the diet, respectively [126]. Adjustment instead for Western dietary pattern also did not change the 12% (1.06, 1.18) and 11% (1.02, 1.22) increased risks

n=~20k Four studies within the Seguimiento Universidad de Navarra (SUN) cohort have adjusted for fat, added sugar and sodium intake, or for dietary pattern. Rico-Campa et al. demonstrated that the highest vs. lowest quartile of UPF intake was associated with a 62% (1.13, 2.33) increased risk of all-cause mortality [103]. Adjustment for saturated and trans fats, added sugar and sodium intake still resulted in a 69% (1.12, 2.56) increased risk of all-cause mortality. A 58% (1.10, 2.28) increased risk still remained after adjusting for Mediterranean diet pattern adherence instead.

n=~12k In the US Third National Health and Nutrition Examination Survey (NHANES III) cohort, there was a 31% (1.09, 1.58) increased risk of all-cause mortality in the highest vs. lowest quartile of UPF intake, which remained significant after further adjustment for dietary quality score using the HEI-2000 (p-trend = 0.001)

n=22k In the Italian Moli-sani cohort, the highest vs. lowest quartile of UPF intake was associated with a 32% (1.15, 1.53) higher risk of all-cause mortality, 65% (1.29, 2.11) higher risk of CVD mortality, and a 63% (1.19, 2.25) higher risk of ischemic heart disease (IHD)/cerebrovascular mortality [105]. Adjusting for saturated fat, sugar, sodium and dietary cholesterol intake resulted in a 28% (1.09, 1.49), 56% (1.19, 2.03) and 33% (0.94, 1.90) increased risk of all-cause, CVD and IHD/cerebrovascular mortality, respectively, in the highest vs. lowest quartile of UPF intake.

n=3k In the Framingham Offspring cohort, each additional serving of UPF per day was associated with a 5% (1.02, 1.08), 9% (1.02, 1.16), 7% (1.03, 1.12) and 9% (1.04, 1.15) increased risk of overall CVD, CVD mortality, hard CVD and hard coronary heart disease, respectively [118]. Further adjustment for diet quality using the DGAI-2010 still resulted in a 4% (1.01, 1.07), 9% (1.02, 1.16), 6% (1.02, 1.11) and 9% (1.03, 1.15) increased risk of overall CVD, CVD mortality, hard CVD and hard coronary heart disease

n=~98k In the US Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial cohort, the highest vs. lowest quintile of UPF intake was associated with a 50% (1.36, 1.64) increased risk of CVD mortality, and a 68% (1.50, 1.87) increased risk of heart disease mortality [119]. Multiple dietary adjustments did not alter this risk; adjustment for saturated fat, added sugar and sodium resulted in a 48% (1.34, 1.63) and 65% (1.47, 1.85) increased risk of CVD mortality and heart disease mortality, adjustment for diet quality using HEI-2005 resulted in a 48% (1.35, 1.63) and 67% (1.49, 1.86) increased risk of CVD mortality and heart disease mortality, and adjustment instead for red meat, processed meat, whole grains, fruit, vegetables, fibre and dairy intake also still resulted in a 49% (1.35, 1.64) and 66% (1.48, 1.86) increased risk of CVD mortality and heart disease mortality, respectively

6. Du, Zhang et al. Ultra-processed food consumption, plasma metabolite profile, and risk

of all-cause and cause-specific mortality in a population-based cohort (2024)

7. From the FDA.gov site:

“The FDA has established a GRAS Notification Program to help ensure that these ingredients are safe for the ways in which they will be used and to help industry meet its responsibility for ensuring the GRAS status of ingredients they intend to use in food. This notification is not mandatory; however, the FDA strongly encourages manufacturers to contact the agency and follow the available procedures for our oversight of GRAS conclusions by submitting a GRAS notice.”

(bolding mine)

“By 1997 FDA had tentatively concluded that it could no longer devote substantial resources to the GRAS affirmation petition process. The agency published a proposed rule outlining a GRAS notification process to replace the resource-intensive GRAS affirmation petition process.”

“The GRAS notification program provides a voluntary mechanism whereby a person may inform FDA of a determination that the use of a substance is GRAS, rather than petition FDA to affirm that the use of a substance is GRAS.”

8. Foster kids have ~1.45x all cause mortality hazard ratios, the “murdering” bit is closer to the truth than we’d like to think about.

9. 4-5.5x HR Triangulated two ways - first from: Khaw, K.-T., et al. (2008). Combined impact of health behaviours and mortality in men and women: The EPIC-Norfolk Prospective Population Study. PLoS Medicine 5: e12.

“Four health behaviours combined predict a 4-fold difference in total mortality in men and women, with an estimated impact equivalent to 14 y in chronological age.”

The four behaviors are basically healthy diet, exercise, not smoking, and not drinking or drinking only moderately. This is without regard to obesity, and is purely behavioral. Obesity is a fairly significant all cause mortality bump (generally 1.2-2.0 RR), even for obese people who exercise, etc. So conservatively, the number for an obese sedentary person is somewhere around 4x-5.5x.

Second triangulation:

In Dan Lieberman’s Exercised, he has this chart comparing exercisers to sedentary people with a 4x all cause mortality difference between the endpoints. That was just referring to activity with no reference to weight - so if you consider a sedentary obese person with the same RR adjustments above, this would shake out to 4-5.5x also.

Ain’t nobody exercising the 30 hours a week at the end of the chart? Well, about 1% of US adults are (mostly professional athletes, a handful of military guys, and some triathletes and ultramarathoners), but sure, that does seem a tad unattainable - and it should be noted that the 650 min point (~11 hours a week - about what I do) is still basically at 4x.

But we can do better with a good diet. So the top quartile of ultra-processed-food eaters by percentage of diet have all cause mortality RR between 1.3-1.7x compared to the lowest quartile. But those guys in the bottom quartile are still eating crap! So then if you follow a good diet with lots of organic fruit and veg and fish and with junk or fast food less than 5% of your calories (ie the bottom 1-5% vs the bottom 25%), you get an additional .75 all cause mortality RR (a 25% bump). So plausibly you could cut end up at the 350 min a week point (~6 hours, or an hour a day with one complete rest day), and still have the 4x all cause mortality benefit. And at the 90 min a week point with a good diet, you’re still cruising at about a healthy 2x all cause mortality benefit!

10. Miranda Jones et al. Smoking, menthol cigarettes and all-cause, cancer and cardiovascular mortality: evidence from the National Health and Nutrition Examination Survey (NHANES) and a meta-analysis (2013)

11. Barrio et al. Mortality in a cohort of young primary cocaine users: controlling the effect of the riskiest drug-use behaviors (2013)

For teenagers specifically, cocaine addiction was a 2.33 RR: Markota M, Croarkin PE, Bobo WV Increased 5-year all-cause mortality in youth with positive urine cocaine drug screens (2016)

Funnily enough, there’s about half as much relative risk for male cocaine users vs female for some reason, so if you’re a guy, cut that RR in half. Also, nearly all these studies are confounded by recruiting new moms who pissed hot for drugs in the hospital, recruiting HIV positive needle drug users, or recruiting people on the street with $15 weekly study offers and things like that - it’s a pretty safe bet they’re sampling exclusively from the highest risk, lowest SES populations out there, and if you’re high SES, your overall RR from cocaine use is likely even less than the 1.2x bottom end.

Smoking, amphetamines, opiates, and even marijuana has the male / female risk split too! Men have a roughly 1.9 HR for smoking, vs 2.3 HR for women, and women enjoy very little benefit from quitting smoking in terms of HR, whereas men drop to ~1.3 HR after quitting. Women who do any drugs seem to universally have much higher (generally ~2x higher) mortality RR than men that do, which I did not know before doing this literature dive.

12. Although for smoking, it is around ~2/3 that end up addicted.

Lopez-Quintero et al. Probability and predictors of transition from first use to dependence on nicotine, alcohol, cannabis, and cocaine: results of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). (2011)

13. Martin CB et al. Attempts to lose weight among adults in the United States, 2013-2016. (2018).

14. Wing, Phelan Long Term Weight Loss Maintenance (2005), DOI: 10.1093/ajcn/82.1.222S

15. Most of the negative health effects of obesity - like cardiovascular disease, risk of stroke, cholesterol and blood pressure impacts - follow a dose-response curve with incremental weight / BMI.