Giles Yeo: ‘Let’s consider the type of food we eat, not fixate on calories’

Since the dawn of the 20th century, almost all weight loss guidelines have used calories as a simple measure of how much energy we’re consuming from our food. But according to Giles Yeo, a Cambridge University research scientist who studies the genetics of obesity, there’s one problem: not all calories are created equal. In his new book, Why Calories Don’t Count, Yeo explains that what really matters is not how many calories a particular food contains, but how that food is digested and absorbed by your body.

Can you explain why you feel calorie-counting is a flawed approach to weight loss?
There was an American chemist in the 19th century called Wilbur Olin Atwater who calculated the calorie numbers for different foods, by working out the total energy intake you get from them. But his calculations never took into account the energy it takes our cells to metabolise food in order to use it. This is important. It’s why for example a calorie of protein makes you feel fuller than a calorie of fat, because protein is more complex to metabolise. For every 100 calories of protein you eat, you only ever absorb 70.

So we need to consider the type of food we’re eating, rather than fixating purely on the calorie content. You could be aiming to eat 800 calories a day, but how much energy your body actually absorbs will depend on whether you’re eating sugar, celery, or steak.

How could people use this information practically when trying to manage their weight?
What really annoys me is when weight loss gurus say things like: “Oh, just replace the chocolate bar with a banana.” Now that’s a dumb thing to say, because sometimes life demands a banana, and sometimes life demands a chocolate bar.

But you can make a better practical choice in those situations by considering how the body processes food. It takes a lot longer to digest food elements like protein and fibre, so you absorb less of those calories, so if you choose a chocolate bar that’s higher in nuts or dry fruits, that will make it a better chocolate bar.

Speaking of more nuanced approaches to managing weight, recently, body mass index has come under fire as being an inherently flawed measure of obesity. Is it still appropriate for doctors to use BMI?
I would define an obese person as someone who is carrying too much fat, so that it begins to influence their health. But then the question is, how much is too much fat? Your fat cells are like balloons, they get bigger when you gain weight, they shrink when you lose weight. But everyone’s fat cells expand and contract to different amounts, so everybody has a different safe, fat-carrying capacity. Famously, us east Asians cannot get too big BMI-wise before we increase our risk of diseases such as type 2 diabetes, because we have a lower safe fat-carrying capacity.

BMI is very useful for looking at population trends, because it’s free to measure, and you only need your height and weight. But at an individual level, this is not useful to predict health because it doesn’t take into account genetic and ethnic differences in fat capacity. You could be a rugby player, or the same size as a rugby player, and be all fat, and yet have the same BMI. So I think healthcare professionals need to consider the whole, rather than simply BMI itself.

The FDA’s approval of semaglutide – the first new drug to be approved for obesity since 2014 – has made a lot of headlines this year. Is obesity something we should be treating with drugs?
I see it as being part of the toolkit. A lot of people consider obesity as a lifestyle disease; they perceive that being larger than someone else comes down to choice, when this is not the case. For many people, it’s more difficult to say no to food purely because of their genetics.

Semaglutide makes your brain think you’re slightly fuller than you are, so you eat less, and if someone is finding it difficult to lose weight, why not use it? I’m not countenancing drugging the entire population, but I think semaglutide should be one of the tools we use, alongside lifestyle intervention, exercise, all of those things.

You research genetic factors that control susceptibility to obesity. What do these genes do?
We now know that more than 1,000 genes are linked to obesity, and the vast majority of them influence pathways in three different parts of the brain that ultimately influence your feeding behaviour. One is the brain’s fuel sensor called the hypothalamus, then there’s the part of the brain that senses how full you get, and then there’s the hedonic part of the brain that makes eating feel good. All of these regions speak to one another, and some of these 1,000 genes influence these pathways. A mutation that causes a slight insensitivity in your brain to how full you are could influence how much you want to eat, making it more difficult to say no to temptations.

For example, we know that one in 330 people in the UK have mutations in a particular pathway called the leptin-melanocortin pathway, which renders your brain slightly less sensitive to how much fat you’re carrying.

Will we be able to use our genetic knowledge of obesity to develop new treatments that regulate appetite?
Yes, there’s now a drug called setmelanotide, which targets the leptin-melanocortin pathway. It has just been approved in the US for rare genetic causes of obesity, and the European Medicines Agency should be approving it within weeks. Eventually it will be trialled in the larger population, much like semaglutide was, to see whether you can fool your brain into thinking that you’re carrying more fat, and so you end up eating less. This is an example of how understanding the basic biology of obesity, the wiring, the genes involved, can result in cogent and hopefully sustainable ways of tackling it.

In your book, you touch on how obesity isn’t solely genetic, and factors such as socioeconomic status play a key role in determining how those genes are expressed, and how much they matter.
The heritability of body weight – the percentage of the trait that is down to your genes – is a range, which lies somewhere between 40 and 70%.

A colleague of mine, Professor Clare Llewellyn at UCL, has a twin cohort called the Gemini study, where she’s looked at the role of socioeconomic status on body weight. She’s shown that if you take middle-class families, the heritability of their body weight is roughly around the 40% mark. But you go to households with the lowest socioeconomic class, and the highest levels of food insecurity, suddenly it jumps to 70%.

Genetically, there’s no difference between people who are poorer and those who are richer – it’s an accident of birth. But because of your socioeconomic situation, your risk of obesity can jump from 40% to 70%. It shows that if we manage to cure poverty, childhood poverty in particular, we can drop the heritable risk of obesity from 70% to 40% without even touching the biology of the system.

Exercise has been getting a bad rep for its weight loss powers. Is that really deserved? You run a lot: are you doing that to keep trim or for other reasons?
In and of itself, exercise is not effective for weight loss. It is in principle possible to lose weight through exercise – Tour de France cyclists eat at least 5,000 calories a day and still lose weight during the three-week race. The problem is most of us mere mortals don’t exercise anywhere near enough for this to be effective. It is, however, useful for weight maintenance after you have achieved weight loss. So I stay active on my bike and running, to stay fit, but also to maintain my weight.