Fibre Optics Part 6

Due Process

Some unprocessed potatoes frying in beef dripping

I’ve now written 5 lengthy blog posts on the new science of fibre, probably about 8000 words or so. In truth, this has barely scratched the surface of this exciting area of nutrition science, but it is still worth congratulating those of you who have got this far. You probably deserve some sort of badge. Stool Warrior. Fecal Champion. Commander of the Poo.

The question is, now we have all invested so much time and mental effort, why is it important? And why is it particularly important right now?

To be honest, I’ll always be interested in discussing fibre and spend a lot of my working life on fibre related projects. Trying to get more into people’s diets, trying to make it taste good, and particularly trying to increase my understanding of how it functions and behaves, both in the body and within food matrices. Fibre is very much my thing, and I genuinely believe that some of the new understanding outlined in these blog posts could have a positive impact on people’s dietary health. It opens up some interesting directions for the creation of healthier food products, and making better food that still tastes good is exactly what all responsible development chefs should be doing with their time.

But there is something more. In recent months, debates around food and health have lurched towards an obsession with so-called ultra-processed foods, resulting in the demonisation of a swathe of manufactured products. This has largely been the result of a couple of popular nutrition books written by media savvy advocates of the NOVA food classification system, which lumps most of the food people eat into the category of ‘Ultra Processed’ and suggests that we legislate against it on that basis.

The main thrust of this argument is that the healthfulness of a food is largely independent of its nutritional composition, and it is better to categorise a food by the level of processing it undergoes, rather than the amount of sugar, salt, fat, saturated fat, fibre, protein and calories it contains.

There is an important but rarely made distinction to make here. Pretty much nobody sensible believes that we should not act to make our food system healthier, and that will always require us to rate and classify foods in some way. Historically, this has largely been done by assessment of nutritional composition, leading to systems such as Nutri-Score or HFSS, both of which are based on standard nutritional measures. But the NOVA system prioritises the level of processing a product has been through, meaning that bacon and beef dripping sit in ‘healthier’ categories than a loaf of sliced wholemeal bread or a flavoured yoghurt. The UPF discussion is not an argument about whether the food system should be regulated for safety and healthfulness – of course it should. It is an argument about whether this particular classification system is of any use. My view is that adopting NOVA into guidelines and legislation would not make our food system healthier and would create a load of loopholes that retailers and manufacturers would exploit to the detriment of consumers (this is a whole other blog post).

Although I am not a fan of the NOVA system, which to me seems arbitrary and unscientific, it does open up a few interesting questions. Unless we are to abandon logic and reason completely, NOVA relies on there being an unknown mechanistic reasoning for processed foods being less healthy than unprocessed equivalents, even when those foods are matched by all standard nutritional measures. So, even when two foods have the same amount of fat, sugar, salt, saturated fat, calorie density and fibre, the less processed one should be healthier. If this is true, then even if NOVA is not the answer, something about our current classification systems is not working, which probably means it is in need of change.

At first glance this seems crazy, relying on some sort of magical essence of caused by a food being made in a factory. But in the only significant Randomised Controlled Trial on Ultra Processed Foods, there did appear to be an effect. In a much cited and discussed experiment led Kevin Hall’s team at NIH, significant differences were found in the number of calories people consumed when placed on an Ultra Processed diet or a less processed diet, even when those diets were carefully matched for sugar, salt and all the major nutrients. Those on the Ultra Processed diet ate far more and gained weight, despite not reporting any differences in the enjoyment of the food they were eating.

Without this experiment, it would be easy to discount most of the research into Ultra Processed Foods, which is almost entirely based on poor quality nutritional epidemiology, often mining the same data sets for correlations, without considering some obvious confounding. But the existence of a well conducted experiment requires us to consider what mechanisms might be driving these observed differences. It also makes us wonder if our current systems for classifying the healthfulness of foods are adequate.

Mechanisms are usually where the Ultra Processed Foods hypothesis falls down. There is sometimes vague handwaving about emulsifiers or additives, something about trans fats (a valid point if they hadn’t been entirely absent from manufactured foods in the UK, US and Europe for many years), and some bizarre theorising about the intension and marketing of a food, as if that is something your metabolism is aware of. But given the considerable differences in a well conducted trial, which saw participants on the Ultra Processed Diet arm eat around 500 calories more per day, there must be some sort of mechanism at play.

Many, including myself, believe it has a lot to do with fibre. Because even though the two diet arms in that experiment were carefully matched for the amount of fibre they contained, because of the difficulties of finding ultra-processed foods with enough fibre in them, a type of soluble fibre had to be added to drinks in the processed diet arm to create a match. And as we have seen, the differences between soluble fibre in a drink and the fibre found in broccoli, oats or brown rice, is considerable. It impacts on food structure, eating rate, the way the body metabolises it, how it affects the microbiome, the production of hormones in the gut, and probably a thousand other things we don’t yet know about. Fibre is important, and it is only classified as all the same because of a quirk of early twentieth century nutrition science. And, as we found out in Part 4, accurately measuring how much fibre a food contains is incredibly difficult, introducing even more uncertainty. It is entirely possible that the NIH experiment was simply revealing some of these limitations.

And then there is food structure to consider. Processing is known to impact on the cellular structure of foods, with heavily milled pulse flours having very different impacts on the body when compared to whole chickpeas, despite being nutritionally identical by all common measures. It is also likely that fats contained within the cellular structures of legumes, nuts or seeds have a different impact on the body than oils that have been pressed, separated and refined. Even though it is not always well understood, it is well known that processing can impact on how foods interact with the body, independently of nutritional composition. Could it be that because of a general shift towards consumption of more processed foods, we are now eating way more of our fibre, carbohydrates and fat in disrupted cellular structures, something that might impact health, but is unlikely to be registered in any dietary surveys? This is a potentially important unknown in the debate about food processing and could explain an awful lot of what is going on.

This should be a win for the advocates of NOVA. At last, a mechanism that can explain some of the effects of processing. But strangely, this understanding of the microstructure of foods is rarely mentioned in the Ultra Processed literature. Why could that be? Why does the focus more often fall on emulsifiers or enzymes, additives that have been extensively studied for safety and are just as common in non-processed foods as ‘ultra-processed’ ones.

Perhaps it is because this new understanding of fibre and food structure opens up avenues to manipulate ultra-processed foods to improve their nutrition. If we definitively prove that it is the micro structures of fibre, fat and carbohydrate that impacts eating rate in the Kevin Hall study, then the Ultra Processed classification is reduced to a objectively measurable difference, so losing its power to control the categories of food that some campaigners seem determined to demonise. If we can make cheap manufactured bread that is demonstratively healthier than an expensive artisan loaf, then to me that would be a huge win for public health. But to an awful lot of food campaigners, it is something that they do not want to consider.

For those of us who wish to make commonly eaten foods healthier, this new science of fibre and food structure opens up a whole range of possibilities. We can produce convenient and delicious foods that retain cellular structures, so retaining many of the health benefits of unprocessed foods. And as we develop our understanding, we can manipulate all sorts of ingredients to increase the healthfulness of our food system. In some cases, we might find that less processing does not constitute an improvement. As we saw in Part 5, a little bit of processing can help beta glucans to form gels. And work on wheat has shown that more intensive micro-milling can greatly improve the bioavailability of iron and other micronutrients. As we gain a more detailed understanding of the impact of food structure on nutrition, we well be able to make a host of foods healthier in a myriad of ways, many of which we do not currently measure.

In the past 100 years or so, nutrition science has come a huge distance, transforming the health of the world. It is sometimes tempting to think that there is not much left to discover, but there always seems to be new leaps to be made, many of which can lead to novel ways of helping people live healthier lives. I am no fan of the current discourse on Ultra Processed Foods, which I feel is being used to make a few campaigners very rich and push certain political narratives, whilst creating an awful lot of unnecessary guilt about perfectly sensible food choices. But if a focus on UPF increases our understanding of how the structure and processing of food might impact on health, and leads to a better food system as a result, then perhaps some of the arguments will have been worth having.