Ultra-Processed Foods and Chronic Disease
Expressing outrage isn't enough
Ultra-processed foods are “formulations mostly of cheap industrial sources of dietary energy and nutrients plus additives” and there has been a lot of noise about them recently. We should definitely be concerned when Americans and Brits are getting more than 50% of their calories from these products, and it’s even worse when we look at young people.
Their consumption has for years been consistently associated with obesity and cardiometabolic diseases, but the study that really made people sit up and take notice was an inpatient, randomised controlled trial which showed that an ultra-processed diet caused people to gain about a kilo in two weeks, while an unprocessed diet caused people to lose weight. A modelling study based on this trial data suggests that if we could get kids and adolescents to eat a pretty much entirely unprocessed diet, the prevalence of obesity in this age group would be halved.
But while it’s really easy to talk about the negative health properties of these products, it’s much, much harder to develop a plan that will decrease their ubiquity and consumption.
I am sure lots of people reading this might be thinking, “well, obviously we just need to ban them”. It’s important to recognise reality. These products are pervasive in our food systems because people find them tasty/convenient/affordable. We can just bang an 80% sales tax on them though, right? The ultraprocessed food (UPF) industry is worth trillions.
Most* governments are not going to put restrictions and regulations on a THOR-LIKE industry that employs millions of people, generates extraordinary returns to shareholders (this includes government pension funds), and pays millions in taxes to the Treasury [despite the usual strategies multi-nationals use to reduce their tax obligations] without an absolute shit-ton of infallible evidence that UPFs directly cause harm.
And this is the problem. We don’t have a good enough evidence base yet.
This is complicated by grey areas like the fact that some of these products provide key nutrients that can be difficult for people to get from food sources (like folate), particularly for vulnerable groups. In the UK, we get ¾ (yes, 75%!) of our magnesium, phosphorus, potassium, calcium and retinol from UPFs. There are other advantages to these types of products – not least reducing food waste - that others have described.
And the challenge with food and nutrition policy is that you better be sure when you are banning certain products, or advising people to not eat certain things, that you have a good idea of whether the thing they’re going to replace that product with is going to be more, not less, healthy.
So these grey areas are legitimate barriers to developing a logical, defensible, and implementable food policy. Just as important (if not more so): these grey areas are ripe for exploitation by the UPF industry. So, we can whinge all we want about how awful they are, but the only way we’re going to change anything (it’s boring and predictable I know) is more research.
The primary thing we need to understand is the mechanism of action. What is it about these products that causes us to eat more? And are there components of these products which – independent of their effect on energy intake – harm our health in other ways?
Let me illustrate why this is important. In an analysis of secondary outcomes in the clinical trial I mentioned above, the authors noted a significant relationship between the rate of eating, and calorie consumption. In other words, perhaps the reason UPFs cause us to eat more is that we can scoff them down really, really quickly. Not only may this not apply to all UPF (do we eat UPF yoghurts quicker than regular yoghurts? UPF bread faster than non-UPF bread? Genuinely don’t know but I doubt it) but it may also apply to foods which (depending on how you define them) are not UPFs - I’m thinking sweet potato fries with just salt and oil added.
Similarly, another concern about UPFs is the additives in them. In my view, it’s probably not likely that these types of additives are having a significant effect on any easily observable clinical markers. But they may subtly alter metabolic pathways that overtime could potentially lead to metabolic dysfunction. For example, if additive X caused a very small but persistent increase in insulin secretion, it is possible that that a 24-hour chronic elevation of insulin, even by a very small amount could contribute to rises in cardiometabolic disease and adiposity over time. It’s a testable hypothesis. And these types of studies could help us understand which additives might be affecting us in significant ways.
Hopefully you can see from just these two examples that if we can understand the mechanism of action, we can direct our policy tools at the specific properties of foods which are driving excess adiposity, rather than a proposed policy being based on somewhat subjective categorisation. This type of more precise, causal evidence may also be an easier sell in terms of driving political support.
How do we get this research funded? I think it’s going to have to be government that provide this funding (we’re talking >$100M). Media attention about the dangers of UPF will hopefully increase public understanding of the role our dysfunctional food systems play in causing chronic diseases and this may help exert pressure on politicians to actually do something about it (but let me repeat – I very much doubt this will happen without slam-dunk evidence).
What about dietary guidelines?
Sadly, what guidelines say has little impact on what people actually do. In any case, most dietary guidelines already recommend reduction of a large number of UPFs including cookies, cakes, etc. The Brazilian government famously changed their dietary guidelines in 2014 to take into account degree of processing when considering the healthfulness of foods. I have looked, but I don’t think there is any follow-up data on what effect these guidelines had on actual consumption (Bolsonaro did shit like shutting down the National Council for Food Security and Nutrition which probably didn’t help). Nevertheless, Brazil has a very different food culture (they were at about 20% UPF in 2008) compared to countries like the UK and the US, and they probably have an easier time of it moving their citizens towards a more whole food-based, unprocessed diet.
My favourite UPFs**
Let me finish by giving a practical example of how the most super-ultra-processed-food can very likely be part of a healthy diet.
In January 2020 I reduced my meat and dairy intake by about 98% purely for environmental reasons, but I exercise a lot and need a lot of protein. And I like lentils, but not that much. So plant-based meat alternatives have become a staple. I probably eat 1-2 servings per day on average. Am I worried about the effect on my health? No.
A diet comprised of predominantly processed foods (and pretty unhealthy ones at that - an example breakfast in the randomised trial was cheerios and a muffin) is unhealthy compared to a near 100% unprocessed diet. But if just one component of each meal is UPF, and the rest is comprised of whole foods? Or if two foods out of 15 eaten in a day are UPF? In general, individual foods have less of an effect than we think: I would be surprised if a 100% unprocessed diet was demonstrably healthier than a 90% unprocessed diet. So I’ll eat mycoprotein pieces or a soy burger but also try to have a legume at lunch and dinner and also eat loads of fruits, veges, nuts and seeds. (Also, shout out to my other favourite UPF - protein powders).
And I think this example really summarises how useful processing can be in solving a legitimate problem (in this example, how to provide enough protein for 8 billion people while keeping enough land for trees and stuff), but also how to mitigate the potential negative effects in a pragmatic way.
An evidence-based food policy can help us achieve this at a population level, but we need data to do it.
*(I say most, because Columbia has a new president and from November 2023 they’re going to put a tax on UPFs that are high in sodium, added sugar and saturated fat. Personally, I think this is a pragmatic step because they're specifically going for UPFs that have negative health properties anyway. Can I see the US or the UK even doing this any time soon? No #nannystate).
**Just wanna finish this piece by upsetting loads of people.
UPFs are here to stay unless we expect all working Mum’s to give up their jobs and become 1950s housewives. We need to differentiate between different types, baked beans & tinned tomato soup (a can of Heinz is 89% tomato) are not nutritionally equivalent to Wotsits!
Isn’t there good evidence that most people find most of most of these foods less satiating than more whole foods, besides higher protein UPFs? Wouldn’t that be safe for doctors to at least share routinely with patients? And that the patient can expect a transitional time of withdrawal from the dependency on the intense flavors that help drive the overeating behavior? Not just say to lose weight or cut calories. That has not been effective.