Children’s Health: Ultra-processed Foods

Lord Krebs Excerpts

Wednesday 25th October 2023 Lords Chamber

[Lord Krebs (CB)]

My Lords, I declare my interests as recorded in the register.

I shall start by saying what I agree with in relation to the previous two speakers: namely, that unhealthy diets are a major cause of ill health in the UK, estimated to cause 90,000 deaths a year. The risk is increasing—for example, obesity levels have doubled in the last 30 years—and is strongly linked to deprivation. Children at the age of 11 are twice as likely to be obese if they are in the poorest decile of the population than if they are in the richest decile.

The key question for this debate is: what is the cause of dietary ill health? Is it the content of certain foods—too many calories, too much fat, sugar and salt, and not enough fibre—or the way in which they are made? The UPF concept says it is the latter, but I beg to disagree. Although the noble Baroness, Lady Jenkin, said there is agreement in the scientific community, that is simply not true. I have studied a number of the original papers and shall highlight two problems.

First, numerous studies have shown that there is no agreement on what UPF is. For instance, in a study published in 2022, a panel of nutrition experts agreed on classification on the NOVA scale for only four out of 231 foodstuffs that they were presented with. Even when the experts were given the ingredients of those foods, they agreed on only three out of 120 items.

Secondly, there is no scientific evidence to show that processing is harmful to health. Most of the evidence that claims to show this comes from epidemiological studies following a large cohort of people over many years and looking for correlations between diet and health outcomes. Correlation does not prove causation, of course, so there are criteria, known as the Bradford Hill criteria, that have to be met before one can infer causation.

I have looked in detail at recent very large epidemiological studies conducted in the UK, France, Spain and the United States. Although they all show correlations, they do not meet the Bradford Hill criteria for inferring causation of a link between consumption of ultra-processed food and ill health. Furthermore, these studies, which were set up many years ago—sometimes decades ago—were not designed to look at UPF, so the dietary records of participants did not include references to it. The researchers had to retrofit a categorisation, and I have said how little agreement there is on the NOVA classification scheme.

The noble Baroness, Lady Walmsley, also cited an experimental study carried out by Kevin Hall, but when one reads the paper that study does not show that UPF is harmful. It shows that when people are offered unlimited hyperpalatable food that is high in fat, salt and sugar, they overconsume. They consume more calories; no one disagrees with that and it is hardly surprising, but it has nothing to do with UPF. I therefore agree with the Minister, who has said many times that the real danger in our diets is HFSS, not UPF. For me, the real pity is that the Government, having correctly identified the problem, are doing so little to tackle it. Here I agree with the points made by the noble Baroness, Lady Walmsley.

Finally, I have tried to take a dispassionate look at the evidence, and one should always be open to the possibility that the evidence will change. But for now, I conclude that UPF is a red herring in the diet and ill-health debate. Since the noble Baroness, Lady Jenkin, mentioned addiction, the paper published in the British Medical Journal on 9 October does not say that the UPF category in NOVA is addictive. First, it acknowledges that the DSM-5 manual—the standard manual for mental health problems—does not recognise food addiction. Secondly, it says:

“The UPF category … captures foods that may be unlikely to have strong addictive potential”


and that homemade foods, such as homemade cookies,

“may … be addictive but would not be considered a UPF based on the NOVA classification”.