Ultra-Processed People: Why We Can't Stop Eating Food That Isn't Food

by Chris van Tulleken

The more I spoke to experts, the more disgusted by the food I became. I was reminded of Allen Carr’s best-selling book, The Easy Way to Stop Smoking.

The idea is that you keep smoking while you read about how bad smoking is. Eventually, the cigarettes begin to seem disgusting.

give in – allow yourself to experience UPF’s full horror. I’m not urging you to binge or to overeat, but simply to stop resisting UPF. I did it for four weeks – if you feel like trying this then do it for as long as it takes to finish the book.

about 6,000 years ago, a North African pastoralist decided to store milk in an animal stomach and ended up accidentally inventing cheese,

Dividing food into processed and unprocessed is an impossible task.

‘Food, not nutrients, is the fundamental unit in nutrition’, and it pointed out an as-yet unexplained phenomenon: that a number of good studies had identified foods, such as whole grains, nuts, olives and oily fish, that seemed to reduce chronic disease risk, but that the benefit of the relevant nutrient – beta-carotene, fish oil, vitamin B, etc – vanished as soon as they were extracted from the food and taken instead as a supplement.

In short, there aren’t any supplements that work for healthy people.

Beneficial nutrients only seem to help us when we consum...

Fish oil doesn’t benefit us, but ...

There’s no supplement, vitamin or antioxidant that decreases risk of death, or even of disease of...

The contrarian journalist Christopher Snowdon made this point exactly when, in January 2022, he wrote a blogpost titled ‘What is “ultra-processed food”?’.

Snowdon had particular beef with one of the op-ed’s rules of thumb for identifying UPF: that UPF is likely to contain more than five ingredients. ‘What kind of ludicrous, arbitrary threshold is that??’ he wrote. ‘None of this has any scientific standing whatsoever!’ It is easy to see what he means. Why not six ingredients? Why not four? But arbitrariness simply doesn’t matter if you’re a scientist.

imagine Monteiro’s team had started with something explicitly arbitrary, like star signs. Instead of UPF, they might have suggested that being a Leo is the cause of obesity – in scientific terms, it just doesn’t matter, so long as you can back it up with evidence.

imagine someone had observed that Leos did have more obesity. Well, then the researchers would have to build an intellectual model to explain why that might be: seasons, weather at conception, maternal diet, circulating viruses at birth and so on. They could have done animal experiments, breeding mice to be born ...

Real-life science often starts with something arbitrary. Sticking things in boxes. Grouping things together. Naming them. We have to draw a line somewhere and describe the object of interest.

Obesity in adults is defined, arbitrarily, as having a BMI of 30 or above. It wouldn’t matter if the threshold was 29 or 31 instead.‡‡

Almost all biological measurements – blood pressure, haemoglobin, lung capacity – fall along a continuum. At some point we draw a line, somewhat arbitrarily, and say the people on one side of the line have high blood pressure or anaemia or obesity and those on the other side don’t.

Gary Taubes is probably why you’re aware of the idea that carbs are the problem. He’s also probably why you’ve heard of keto diets (high in fat, low in carbs) and might even have tried to cut down on sugar and other carbohydrates

Taubes’ alternative hypothesis runs something like this: Americans (and everyone else) are eating more because they’re hungrier, and they’re hungrier because of the hormone insulin. Insulin is secreted into the blood by the pancreas, and removes sugar from blood and allows it to enter cells as fuel. If you eat carbs, your blood sugar starts to rise, but insulin brings it back down to normal. When insulin is high, like after a meal, it reduces appetite and turns sugar into fat for storage. When insulin’s low, like when you haven’t eaten in a while, you start burning fat instead.

The idea is that, when we eat lots of carbs, they cause a rapid spike in our insulin levels to cope with the sugar. The spike not only promotes the storage of fat, but it drops sugar levels to lower than they were before the meal. This starves our muscles of energy, meaning that we’re less active as well. Moreover, the fact that the muscles feel starved sends signals to the brain to consume more food.

Taubes suggests that, if we avoid carbs, the opposite will happen: our insulin won’t spike, we’ll store less fat, our energy expenditure will go up and our appetite will go down.

But insulin has a huge range of other functions in multiple body tissues, and there are many other hormones in the mix that help determine whether we store fat or whether we burn fat or protein or carbs for our fuel.

There were always huge holes in the evidence that sugar might be the sole cause of overweight and obesity. Taubes’ theory depended on the idea that, since a low-fat diet had been recommended, everyone had not on...

Taubes used low-fat yoghurts as an example, because they were often sweetened with sugar and thickened with carbs to make them palatable. He quoted US Department of Agriculture economist, Judith Putnam, about increases in carb consumption: that, between 1980 and 2000, the average person increased their an...

although it was true that Americans had been eating more refined carbohydrates, they had not been eating les...

According to the US Department of Agriculture, consumption of fats had increased between the late 1970s and the publi...

In a report on Taubes’ piece published by the Center for Science in the Public Interest, Putnam stated that she had explained all this to Tau...

Taubes presented the possible causes of obesity as a dichotomy: fat or carbs. Other possible explanations – exercise, the role of industry, processing, air quality, or some comb...

In 2012, he partnered with a charismatic Canadian physician called Peter Attia and together they set up NuSI – the Nutrition Science Initiative – and raised millions in funding. The plan was to solve the problem of obesity in the USA. They would conduct a series of experiments to demonstrate, once and for all, that calories from sugar promote weight gain more than calories from fat.

they recruited excellent scientists who were sceptical of the whole hypothesis. One of these scientists was Kevin Hall.

The low-carb diet did result in decreased insulin in all volunteers which meant that the experimental conditions were adequate to test Taubes’ hypothesis that insulin levels were important. But when the overall data were analysed, there was a surprise: there was no difference between the groups in terms of the effect of fat or sugar on metabolism. A calorie was a calorie, regardless of whether it came from carbs or fat. It was a small study, but a rigorously conducted small study can still disprove a hypothesis. Hall published his findings3 and reviewed them in the European Journal of Clinical Nutrition.

This article was as much about the philosophy of science as it was about nutrition.

Hall reiterated that we can’t definitively prove any scientific model. Instead, scientists run a series of experiments and make observations. Only if the model stands up to these tests does it become widely accepted. But the most important part of any model or theory is that it should make predictions that, if wrong, will invalidate the model. As much as we all like to be right, good science is about trying to prove yourself wrong. Hall felt that, whether NuSI could admit it or not, important aspects of Taubes’ carbohydrate-insulin model had been shown to be incorrect – the model was too simplistic.

I asked if he was bored of writing about carbs. ‘My wife says if I saw someone hit by a car crossing the road I’d find a way to blame carbs,’

it wasn’t just Hall’s pilot study that seemed to have contradicted significant parts of the carbohydrate–insulin hypothesis. There’s no shortage of other evidence that contradicts it too. The hypothesis has been tested many times, and the longest-running study in people living in the real world found no sustained differences in calorie intake between low-carb and high-carb diets.

In another experiment, volunteers ate two diets in a random order – one 10 per cent carbs and 75 per cent fat, the other 75 per cent carbs and 10 per cent fat. Contrary to what the carbohydrate–insulin hypothesis would have predicted, it was found that participants actually ate 700 fewer calories per day on the high-carb diet, and that only the high-carb dieters reported a significant loss of body fat.

a large US study showed that the reduction in energy expenditure between 1960 and 2006 nearly perfectly accounted for weight gain over the same period.

If people are eating fewer calories but still gaining weight then they must be far more inactive than before.

average calorie expenditure for people in the UK is around 2,500 calories. Yet the survey data Snowdon quoted suggested that we’re eating fewer than 2,000 calories per day, which would mean an average daily deficit of around 500 calories.

study showed is that people in surveys were underestimating their calorie consumption by more than 30 per cent. US studies back this up. A comparison of the US National Health and Nutrition Examination Survey’s doubly labelled water data with the actual survey data showed consistent under-reporting of how much people eat and over-reporting of how much they move.

the doubly labelled water experiments show that we’re actually underestimating how much we eat to a much greater extent today than we did a few decades ago.

People regularly under-report things they find shameful.¶

studies have shown that a desire to lose weight increases under-reporting,

Third, people are eating more snack food outside the home than previously, and this food is easily forgotten and harder to capture.

Fourth, we’re getting worse at responding to surveys in general. Economists are upset about this, so they’ve studied it.

Fifth, the portion sizes of our foods are bigger, and their energy higher, than the reference databases say they are.

In Christopher Snowdon’s case, he is paid a salary by the Institute of Economic Affairs, a free-market think-tank. Their finances are largely opaque, but they have received funding from sugar giant Tate & Lyle.57, 58 The sugar industry has an interest in promoting the narrative that inactivity, rather than food, is the problem. How much this arrangement informed Chris Snowdon directly isn’t clear, but we know from research in other areas that people are often unaware of how they are influenced. Doctors consistently deny that drug company funding affects our practice or our research,

Snowdon’s paper made an impact, but it wasn’t published in a proper academic journal with expert, independent peer reviewers. What about those papers that were though? The ones by professors Steven Blair, Peter Katzmarzyk and James Hill, among others, that emphasise the role of inactivity? Remember Blair saying that there was ‘virtually no compelling evidence’ that fast food and sugary drinks are to blame?

I went back to papers by these authors to look for conflicts of interest and found papers from 2011 and 2012 that stated clearly that there were none to declare.59-61 But in 2015, some scientists sent freedom of information requests to the universities that employed these and other scientists.62 In response they received 36,931 pages of documents, including emails between the scientists and Rhona Applebaum, then chief science officer of the Coca-Cola corporation. Thanks to these requests, and a huge amount ...

Coke helped Blair establish that non-profit group, the Global Energy Balance Network,63 which promoted the message that there was no compelling evidence of a significant link between sugar-sweetened beverages and obesity.64 Coke funded all those papers I listed earlier, by Blair, Hill and Katzmarzyk.65, 66 Coke even funded an entire national programme, run by the American College of Sports Medicine, called ‘Exercise is Medici...