SUNDAY, 27 OCTOBER 2024

I’m something of a sucker for reading or watching anything to do with nutrition and whether a given food is good or bad for us. I’m not alone in this, I’m sure, and the internet has done a lot to feed this obsession with the effect of food on our health. But this fascination with nutrition is not simply a contemporary trend fuelled by a life spent online, as is made clear by a book I recently read called Eat Your Heart Out by Dr. James le Fanu. This was written as long ago as 1987, but in my opinion its core questions are still relevant today. Le Fanu contends that the range of vastly different cuisines around the world proves that there is no single diet which guarantees better health and wellbeing, but that most diets are adequate as long as they have contain enough calories, vitamins, minerals and essential amino acids, and that expecting more than this from a diet is misguided. There is no such thing as the healthy diet and certainly no such thing as a superfood, and searching for them is a waste of time.

The main evidence le Fanu offers for his argument is the contrast between the nutritional advice given by experts in the 1930s and the contradictory advice offered in the 1970s and 1980s which was de rigueur at the time he wrote his book. In the earlier period, there had been a focus on protein and calories and the key message was to eat more meat and dairy. In contrast, the post-war years saw a reverse of this advice as saturated fat became seen as the main cause of higher rates of cardiac arrest, strokes, diabetes, and even cancer. Red meat became a dangerous food which should be eaten sparely and replaced when possible by less fatty meats like chicken, full-cream dairy should be jettisoned for low-fat or fat-free alternatives, salt was starting to become the bad guy it is still often seen as today, and the food pyramid emerged, with lots of basic carbs from things like pasta bulking up the low-fat diet.

Nutritional science has obviously moved on a lot since le Fanu’s time (for example, there is no mention of the microbiome in his book) and expert advice has changed in response to this. So for almost a century now, the guidelines of reputable organisations like the NHS or the American Heart Association have often fluctuated wildly, and this has not only been true of the foods which science never seems sure about (e.g. eggs and coffee) but of the broader macro-nutrients. The worries about meat, dairy and salt of le Fanu’s days have, if anything, grown stronger and have now been joined by concerns about sugar, carbohydrates, ultra-processed food (UPF), and sometimes even fruit.

These changes in the official guidelines have created a scepticism about nutritional science among many people. As the 180-degree turn within the period 1930-80 highlighted by le Fanu shows, fashions come and go in the world of nutrition with very little consistently agreed on except for the health benefits of vegetables. When I read his book, I was struck by the confidence with which scientists and dietitians expressed their advice, and it feels as if the advice has changed but not the certainty with which it is expounded. Also, the official guidelines of the medical establishment all tend to say the same thing at any given moment in time, with little recognition of new or alternative ideas until the paradigm shifts and the herd gradually moves to its new collectively agreed position.

This scepticism has been exacerbated by the rise of internet gurus, the host of doctors and nutritional gurus offering advice to the general public on sites like YouTube. I have to admit that there is something addictive about these ‘experts’ and their pronouncements even though I realise that I have no way of knowing whether any of them are in fact qualified doctors. I suspect that most of them are, and they are not just a quack putting on a white coat and wrapping a stethoscope around their neck, but even if they are genuine medical practitioners, so little of medical training involves learning about diet that they are unlikely to have much more knowledge in this field than many laypeople. These self-declared gurus are a cacophony of voices saying opposite things: for every person arguing against meat and dairy, there is a person arguing for them; for every vegan there is a carnivore; for every expert saying that carbs are the core problem there is another placing all the blame on UPFs. Even foods which are usually beyond question (e.g. avocados, olive oil, fruit) have their occasional detractors. Recently, for example, I watched a woman screech into the camera that ‘fruit is evil’ like some deranged religious believer warning us about the temptations of the Devil.

But it’s too easy to lay all of the issues surrounding research into nutrition at the door of internet quacks with their weird obsessions or to blame the media for sexing up scientific research to turn it into clickbait (scientists themselves are not above this kind of sexing up in their press releases in their search for future research funding). A quick look at the history of nutritional guidelines by official sources based on research suggests that a dose of humility on their part would not go amiss. Le Fanu’s question about whether a specific diet can really make such a huge difference to human health and wellbeing remains a valid one in my opinion, and even if it seems reasonable to assume that diet can have this effect, how can we ever confidently identify what that diet is?

We are always told that randomised controlled trials (RCTs) are the gold standard of any research into nutrition but the problems associated with them are far from trivial. Most trials usually have a small sample of subjects, which becomes even smaller once they are split into two groups so that there is a control group. They also tend to take place over a very short period of time, but it seems logical to assume that any effects of diet on our long-term health are unlikely to show after just a month or so. RCTs almost always depend on markers for health (blood pressure, cholesterol levels, blood sugar, and so on) rather than actual health outcomes. And how do we know that heightened levels of these markers after eating a certain food (for example, the sugar spike that follows eating foods with a high glycemic index) is genuinely harmful or just an unimportant temporary effect? After all, blood pressure will often rise to extremely high levels during vigorous exercise, but no one is suggesting that exercise is bad for us. Then there is the fact that we are talking about statistical significance which is often very small and could possibly be the 5% of times when we have an outlier. Meta-analyses are a common way of trying to see past all these issues but it is rare to get two or more pieces of nutrition research which are exactly the same, so there is a lot of room for confounders. The problems of RCTs are considerable.

In addition, is it really useful to look at specific foods or even types of food and study them in isolation? If bread is bad because it causes a sugar spike, is this true whatever else we eat with the bread? We know, for example, that eating fats at the same time as vegetables helps with the absorption of fat-soluble vitamins, so shouldn’t we be focusing on combinations of food rather than single items? In practice, though, this sounds totally unrealistic. Also, nutritionists are realising that no two people are the same and what is good for Jack may not be good for Jill. In extreme cases, some people can die from eating a peanut while other people derive great benefits from them, so aren’t much more subtle effects of specific foods going to happen? Zoe, an organisation which makes some of the most convincing internet videos on nutrition, are talking more and more about the need for a personal nutrition profile, but is this a realistic aim for anyone except the rich in a world where health services are coming under increasing financial pressure?

We are often warned that epidemiological studies are much inferior to RCTs, mainly because of the sharpshooter fallacy. This means that if we have a mass of data there are bound to be clusters that look meaningful but are actually random since these clusters will appear purely by chance if our data is big enough. However, the sheer size of this data presents opportunities because some of those clusters may identify genuine phenomena which can then be explored in RCTs. For example, this has led to the identification of ‘blue zones’ where people regularly live long and healthy lives – real outcomes and not markers, based on a lifetime and not a few weeks or months  – and some scientists are investigating the possibility that these zones have things in common which lead to these healthy outcomes. But from what I have read, the diets in these blue zones are incredibly diverse and seem to support le Fanu’s thesis that human beings can be healthy and have long lives on completely different diets and that the search for the perfect diet is a chimera.

I’d like to turn briefly to the role of the media in all of this. From a combination of self-interest (a desire to get more readers and therefore more advertising revenue) and lack of understanding or time to study the press releases of researchers (and therefore poor analysis of the research), they massively simplify and exaggerate its conclusions. To this we must add our own role as the general public. Put simply, we will tend to believe the research that we want to believe, and to cherry-pick that which suits us. Vegans will focus on research suggesting that meat is deadly; carnivores will root out that which refutes this idea. I’m as guilty of this as anyone. A few years ago I found some research by a Finnish scientist which suggested that wine only had serious negative effects if we drank more than one bottle a day. I jumped on this at once, and even though I consciously realised the dishonesty in what I was doing, at some deeper level I at least half-believed this research because I wanted to.  

I know I’m in danger of coming across as a smartass in this essay, someone who knows better than the scientists when I list these problems, but I don’t think for a moment that nutritionists are unaware of them: they are an integral challenge of their field which they have to negotiate all the time. My concern is that these factors added together – the need of scientists for research funding, the role of much of the media, our own bias towards what we want to believe, plus the bewildering complexity of the field – is making us neurotic about food and turning what should be one of life’s greatest pleasures into a source of fear and loathing. So even if le Fanu’s thesis turns out to be flawed, and the intuitive feeling that we are what we eat is at least partly true, I think he encourages a healthier attitude to food which will benefit us in the long term.