Illustration: Penny Whitehouse


Why can’t science make up its mind about what we should be eating?

Angry Chef, aka Anthony Warner, is on the warpath – this week over confusing food headlines

Added on

By Anthony Warner on

The American Journal of Nutrition recently published a studydetailing the results of a number of medical studies into different food items, looking at whether they cause or protect against cancer. Eggs for instance have four studies indicating a role in cancer prevention, and six that indicate a causal link. Coffee has four on each side and one in the middle (you might have missed that one, I do not remember the ‘Coffee not linked to anything’ study getting a lot of media coverage).

It will come as no surprise to anyone who reads a newspaper that for the vast majority of foods there is much evidence on either side of the fence regarding cancer risk. The website‘Kill or Cure?’ helpfully details the Daily Mail’s seeming quest “to classify every inanimate object into those that cause cancer and those that prevent it”. From apples, cabbage, pastry, yoghurt and tofu it is an astoundingly comprehensive list, with a huge number of foods falling into both camps.

It is understandable that this might just cause a little confusion. As each new study contradicts the last, the public are left not knowing what to believe. Official dietary guidelines are almost constantly under fire, contradicted by new studies and swathes of anecdotal evidence from researchers and health practitioners keen to counter them. To an increasingly disconsolate public, the messages of the “so-called experts” seem based on the flimsiest of ground, riven with infighting and profound disagreements. 

Why is this? Why can’t science just make up its mind and tell us what we should be eating?

Simply put, when it comes to food and nutrition, hard science is a really hard thing to do and rarely produces clear, unambiguous results. 

If you want to test the effects of a drug, then it is fairly straightforward to design an experiment. You will be able to perform a randomised controlled trial against a placebo (sugar pill), and produce some pretty good quality evidence as to how your new drug affects your test group. 

Interesting results will always be the most likely ones to reach the newspapers, but it is worth remembering that interesting results are highly likely to be wrong

If you want to test the effects of food however, things are considerably more difficult. The effect of diet on the body is astoundingly complex and much evidence will come from fairly reductionist laboratory experiments into the effect of single nutrients on animals or cells in test tubes. These experiments can give scientists important clues, but they are notoriously unreliable. 

If you want to study the longterm effects of a specific diet on real people, accurate experimentation is even harder. It is incredibly hard to accurately control the food intake of large groups of people over a long period of time. Usually you will have to rely on epidemiological data (based on statistical observation of populations) rather than controlled trials and this is further complicated by the fact that much of the evidence will be based on self-reporting. 

As anyone who has ever been asked by their doctor how many units of alcohol they consume per week will attest, we all have a tendency to be a bit creative when it comes to self-reporting anything.

These difficulties can lead to nutritional science producing some unexpected and interesting results. Interesting results will always be the most likely ones to reach the newspapers, but it is worth remembering that the more interesting a result is, the more the public should try to ignore it. In an area of study like nutritional science, prone to inaccuracy and randomness, interesting results are highly likely to be wrong. 

For the outside observer, the constant disagreements between scientists is unsettling and difficult. We are left not knowing what to believe. But this uncertainty and disagreement is exactly how science proceeds. Argument has always been the driving energy of progress, ensuring that we never settle for easy consensus views. It has led to us discard bloodletting in favour of germ theory and antibiotics. It drove Einstein to create of a theory of Relativity that contradicted Newton’s Universal Gravitation.

Given this, the world of science will always struggle to deliver a single consistent voice telling us how we should live our lives. As newspapers continue to report single experiments as gospel truths, we will continue to receive conflicting messages. 

Whenever you see the headline ‘Too much XXX is bad for you’ remember that too much of anything is bad for you. The clue is in the word ‘too’ 

So what should we do? Is our quest for certainty hopeless? Perhaps not. Public health information, provided by the likes of Public Health England or the NHS is constructed by a process known as “systematic review”. This means all the best available evidence is combined and reviewed by experts in order to create guidelines as to how we should proceed. Any significant, well conducted piece of research into diet and nutrition (including the one that said coffee doesn’t cause cancer) will be looked at. 

This sort of complex review is used to produce guidelines like the Eatwell Guide, and because it encompasses all the available evidence, it is the source that we should have the most trust in when it comes to making decisions about the food we eat. It is not always perfect and sometimes guidelines are changed, but they are by a long way the best information we have.

Because of how science works there will always be single experiments that contradict these guidelines, and plenty of individual scientists with an interest in challenging them. But in general these scientists will be the ones whose research has produced an interesting result. And interesting results are usually wrong.




A study by scientists at the University of Bath showed that drinking sugar water during exercise stopped glycogen stores being depleted. The researchers were a little surprised to see the experiment reported in newspapers and media outlets around the world, with headlines proudly claiming that sugar water is more effective than sports drinks in maintaining performance. The study had not used sports drinks as a comparison and had not measured performance at all, only glycogen storage in endurance athletes. Somehow, ‘Sports Drinks are a Waste of Money’ is a better headline than ‘Sucrose Solution Affects Glycogen Depletion’.


A widely reported German study made headlines around the world when it claimed that eating bars of chocolate can actually increase weight loss as part of a low carb diet. The study from The Institute of Diet and Health tested a number of volunteers and found that those who ate two bars every day actually lost more weight than the control group. The only flaw? It was actually a hoax designed to show how easy it was to manipulate research results and generate headlines around the world. Although the study was real, the conclusions were based on clever manipulations of data and The Institute of Diet and Health is just a website created by the hoaxer, a journalist called Johannes Bohannon. Somehow, perhaps due to the lure of a ‘Chocolate for weight loss’ headline, none of the journalists who reported the story bothered to check.


After a recent Danish study on the health benefits of exercise "Too much jogging is bad for you" was the line taken by most newspapers. The actual study, looking at the long term health implications of exercise, looked at over 1000 people. Once they had been broken down into specific categories (non-joggers, light joggers, medium joggers etc) the numbers in each group were quite small. The group of ‘strenuous joggers’ on which many headlines were based was only 36 people, 2 of whom died during the study, nowhere near enough to draw any firm conclusions of increased risk. And whenever you see the headline ‘Too much XXX is bad for you’ remember that too much of anything is bad for you. The clue is in the word ‘too’. 


One of the most widely reported studies this year was the Public Health Collaboration’s report entitled Eat Fat, Cut the Carbs and Avoid Snacking to Reverse Obesity and Type 2 Diabetes. Almost all newspapers ran headlines claiming that the current dietary guidelines are fundamentally flawed, especially when it comes to recommendations on fat. The unfortunate reality for those looking forward to piles of bacon fried in butter, is that in scientific terms the report was largely meaningless. Despite its official sounding name, The Public Health Collaboration is not a Government body and its source of its funding is not known. Dr Mike Knapton from the British Heart Foundation was quoted as saying “this report is full of ideas and opinion…it does not offer the robust and comprehensive review of evidence that would be required for the BHF to take it seriously’. Allison Tedstone, the chief nutritionist at Public Health England went further describing it as "irresponsible" and "a risk to the nation’s health". 


Catch Angry Chef's blogs here

Illustration: Penny Whitehouse
Tagged in:
food honestly

Tap below to add to your homescreen

Love The Pool? Support us and sign up to get your favourite stories straight to your inbox