Why I don't think that "UPF" is a very useful term

[andrewducker]

I've just been having a not-very-helpful conversation about Ultra-Processed Foods (UPFs) which mostly just reinforced my view that the UPF category is not a very useful one.

It contains so many different things that are so very different to each other - thousands of ingredients, thousands of processes. And occasionally someone comes along with a new study that says "UPFs kill 4% of people!" and it's incredibly frustrating because what I want to know is *what* kills 4% of people. Is it baking? Is it emulsifying? Is it flavourings? Is it baking powder?

Telling me "Any pre-prepared food you buy is killing you!" isn't very actionable for most people. If we actually care about people's health and its relationship with food then some large-scale research into exactly *what* foods are causing increased deaths, and which preparation methods, so that people can avoid those specific things.

Additionally, if we want to save live then a systemic fix is probably a good approach - regulate the use of things that kill people. And "Restrict the use of High Fructose Corn Syrup" sounds like something you could regulate. "Restrict the use of UPFs" is something so general as to be unmanageable.

Comments

[juan_gandhi]

Makes sense to focus on corn syrup and similar things. Not just "anything that was already prepared for you".

[wychwood]

It drives me mad as a category because it really does just seem to mean "foods we think are unhealthy already" - oven chips (potato and oil!) are somehow UPF but yoghurt (complex fermentation process) isn't! etc.

[gingicat]

There's also a big divide between yogurt that has gums and fillers, and yogurt that does not. I've been buying full-fat yogurt because it tastes better than the low-fat yogurt that uses gums for texture.

[gingicat]

Ultra-processed seems to *generally* (not always) mean a combination of number of steps plus adding things that are heavily refined.

Heavily refined can mean white flour and stevia to some people, and xanthan gum or inulin to others. If I have the option of xanthan gum or not xanthan gum, I'll tend to go with the not. However, as someone required to eat a low-sugar, high-fiber diet for diabetes, I eat a lot of sugar alcohols and inulin and xanthan gum because that's what's available and tastes good. Monkfruit powder is always bulked up with am additive that falls in my personal category of heavily refined, but it tastes a whole lot better than stevia to me. Hopefully someday it will cost less - a 2lb bag at Costco is USD10.

I follow a rule of "if I'm not sure what it is, I'll look it up on my phone and then decide" when in the grocery store.

[andrewducker]

Sadly, sugar alcohols totally wreck my stomach and trigger my IBS.

[gingicat]

That's sadly common.

[hilarita]

As someone who can't have gluten, avoiding xanthan gum or similar is not feasible, unless I really limit my diet in a way that's also not great.

[gingicat]

Baking for gluten-allergic people is how I discovered that one can buy xanthan gum for home use.

[cmcmck]

And let's also talk, shall we, about the lack of knowledge of food preparation and the lack of affordability of 'healthy' food to those not middle class and comfortably off?..........

[andrewducker]

That's one of the big reasons I don't think that saying "Just cook better!" helps.

[nancylebov]

Throw in lack of time and kitchen facilities.

[gingicat]

Indeed!

[aldabra]

There's also an issue of inadequate regulation of opaque systems. You can still broadly trust that if you buy a tomato you're getting a tomato (although you can't can it according to a twenty year old recipe because it has more sugar in it, and Tesco is defecting appallingly by selling non-vegan citrus fruits). But if you're buying reconstituted powder there is great scope for things to have gone wrong invisibly. Your illicit horsemeat is more likely to be sold in pre-made pies than as roasting joints.

A lot of the UPF problem is that the more stuff has been processed before you eat it the faster you digest it, with consequent blood sugar issues. But some of it is that you can't tell what you're getting by looking at it, and so there is great scope for optimising for colour and taste and smell and cheapness rather than healthiness. The US is about to run into great problems by dismantling its food safety systems at the behest of the commodity manufacturers; that looks like a great efficiency briefly and then it doesn't.

If you were a local socialist council trying to optimise for your local production you could arrange for the production to be done in local facilities, so you could see the ingredients going in as well as the product coming out, and you could deploy your sixth-formers and students on work experience gigs to observe the processes. But global capitalism doesn't lend itself to that.

[gingicat]

And of course there's the issue that processed foods are the ones most likely to taste exactly the same every time, which is important to those with high taste sensitivity.

[aldabra]

But also, there's the question of how this research is done, which is to find a sample of 100,000 people like me and ask them to fill in a food diary for a week once every fifteen years, and then see how often we die. Those food diaries can't ask you to transcribe commercial ingredient lists otherwise everyone would drop out instantly, so fifty years later the person doing the stats can't pin it down to individual ingredients. They can broadly label bacon and sausages and ham and ready-made pies as processed, and home-roasted meats and sliced chicken as not-processed, and even that is a bit imprecise. These days they can ask whether your fruit squash has sugar or sweeteners but fifteen years ago they probably couldn't because they weren't filtering the questions electronically in real time.

So if they're getting a UPF signal at all their UPF may be 1990s-era Pot Noodle rather than 2025 veggie turkey-substitute. There hasn't been time yet to do longitudinal health studies on contemporary processed foods. They can't tell you in any detail what it is that the signal shows, but you don't want them suppressing the fact that they've found a signal.

They'll be designing the studies now that would pin this down further by 2060, if they still have funding for this kind of thing in 2060. I think they're sure enough that there's a problem with high fructose corn syrup that they're testing it on hamsters now; perhaps they'll do that on processing in general.

[bens_dad]

They can broadly label bacon and sausages and ham and ready-made pies as processed

I've seen suggestions that english sausages are very different from continental sausages (preservatives IIRC) and don't count as processed for this sort of discussion of food health.

[hilarita]

The very best food diary people cut out the ingredient list on processed food, and sent it in with their diary, and then it could be stored along with the diaries and evaluated as well.(Absolutely not everyone used to do this, I hasten to add.)

(I was a data entry minion for the EPIC Norfolk project for about 9 months.)

Well!

[lsanderson]

Fresh from the BBC:

Ultra-processed foods may be linked to early death
Philippa Roxby Health reporter
https://www.bbc.com/news/articles/crm30kwvv17o

Re: Well!

[andrewducker]

It was that study I was discussing!

Re: Well!

[lsanderson]

Shocked! I'm just shocked! (Although I thought your lead time might be longer.) There's not many places that do controlled human food studies for fairly obvious reasons, so it's a bit of a crap shoot when somebody wanders into these spaces. And so much of the money is on the side of the processed foods people, and not the health foods people -- plus some of the health foods champions present as not playing with a full English breakfast.

[magedragonfire]

The category is so broad that it’s functionally useless. Sourdough bread and tofu both fall under the UPF umbrella, but they’re not harmful to you unless you have an allergy, and are beneficial in a lot of ways too. Even additives to bread like “dough improver” aren’t all scary when you start looking at them - stuff like ascorbic acid (vitamin C), citric/acetic acid (powdered lemon juice or vinegar), salts that help yeast rise better, etc.

It’s like saying something’s organic (there are plenty of organic pesticides); the term itself doesn’t automatically mean “healthy” or “unhealthy”, although there’s a huge push by influencers to make it seem so.

Found on the Interwebs

[lsanderson]

Is ALL cooking "ultra-processed" food?

https://youtu.be/OhA3T60PtSM?si=fAk9u7oSbHNWVaor