Mobile Menu

Down the Rabbit Hole: Genetics and Obesity – Giles Yeo

Giles Yeo is a Principal Research Associate at MRC Metabolic Diseases Unit and the Scientific Director of the Genomics/Transcriptomics Core at the University of Cambridge. Yeo’s main research focus is exploring brain control of food intake and the role genetics plays in appetite behaviour. Yeo is not just a researcher; he is also a published author and broadcaster. 

Please note the transcript has been edited for brevity and clarity.

FLG: Thank you for joining me today Giles as we go down the rabbit hole and take a look at some of the niche and unusual aspects of genomic research. Today, we are going to be talking about obesity and the role that genetics plays in this. So, if you could start off by introducing yourself and your background.

Giles: My name is Giles Yeo. I’m a geneticist and I work at the University of Cambridge on body weight, of which obesity happens to sit at one end of the spectrum. So, I am an obesity geneticist – that’s what I do.

FLG: Obesity is a major public health crisis, which has been emphasised by the COVID-19 pandemic and the associated increased risk of severe disease. There are always interviews and debates on TV where everyone is passing the blame. So where does genetics fit into this? How does genetics influence our body weight and influence obesity, and how much of a role does it play?

Giles: I think that’s a great question because whenever I do mention that I study genetics of body weight people always look at me with deep scepticism and the reason is because they feel that if you’re saying genetics, then it’s like you’re trying to magic someone’s weight – you are trying to say that this person magically, because of their genes, will become smaller or larger. They think surely you know it’s because they eat too much, and they don’t move enough or vice versa and that’s what they need to do in order to lose weight – they need to eat less and move more. There is a nuance here and at the end of the day our body weight has to be a function of how much we eat and how much we expend energy. It has to be, because it’s physics. So whatever explanation there is, biological or otherwise, it has got to come down to influencing the physics.

And when we talk about the genetics of body weight, we’re actually talking about the genetics of feeding behaviour and specifically the genetics of how our brain influences our feeding behaviour. In other words, why we behave differently around food, why some people respond to stress by eating and other people respond to stress by not eating, why some people love food whereas other people use food for fuel, why some people eat without knowing that they are actually eating. These different behaviours around food influence the physics, meaning some people eat more or less on average and so therefore, why some people are small, medium and large. So that’s what we talk about when we talk about the genetics – not some weird phenomenon that happens to bypass physics. We are now understanding how feeding behaviour specifically influences our food intake and therefore influences the physics

So – how much? That is another great question. So, for the vast majority of human traits and behaviours that we know have a genetic influence, a lot of this was worked out using twin studies. So, there are identical twins who, for all intents and purposes, are genetic clones of each other – they share all their genes, 100% of their genetic material. Then, there are non-identical twins that will share as much genetic material as you would with your own brother or sister or for that matter your parents – 50%. So, if you study enough twins, identical and non-identical, 100% versus 50% of genetic material, then you can take any given human trait and ask – well, what happens if you share all your genes versus half of your genes? And then work out what we call the heritability. This is the number you’re looking for – so, what percentage of this particular given trait is going to be down to your genes versus the environment. And if you do that for body weight then you end up with a range of numbers between 40 and 70%. So, between 40 and 70% of the variation of our body weight is going to be down to our genes. Then you can look at the Ying to the Yang to that, which would actually mean that between 30 and 60% is going to be down to the environment.

FLG: Does that differ in regards to monogenic and polygenic types obesity?

Giles: So that is a big average. Because we’re talking about the average – we’re talking about what affects most people. When we actually looking at monogenic disorders, otherwise known as mendelian disorders (Mendel and his peas), what we’re talking about is causal – where if you have a mutation, like muscular dystrophy and cystic fibrosis, and certain obesity disorders, you are going to end up being severely obese because of circuitry and therefore you eat way way way too much. So, for the vast majority but not all monogenic syndromes this would be 100% heritability, which means that the moment you have it, you will actually be larger. Whereas what we’re talking about as geneticists with regards to heritability is the average heritability of body weight across the population.

FLG: Are the monogenic forms of obesity more easily identifiable compared to polygenic forms?

Giles: It depends how severe the obesity is, because you could have monogenic obesity that makes you quite heavy but not heavy enough for the fire truck to come and try and bash down your door to crane you out. So, I guess it depends on the severity of the phenotype. But there are monogenic obesity syndromes in which the effects are so severe – okay not fire truck severe, that was a slightly facetious comment – but it is severe enough that at childhood you would have been taken to the doctor. We’re talking about 3-year-olds that are 40-50 kilos alright, so look I am 5 foot 9 and I’m 75 kilos so we’re talking about a 3-year-old whose two thirds, maybe half my body weight – it is not normal. So, I think the child would have been taken to the doctor to find out what is going on.

Today, if you end up with that kind of severe obesity, early onset in particular (so that’s the other thing severity and early onset rather than something that’s later onset), in this country, if you go to somewhere like the GP and they know what’s going on, you would be referred to the right person, to the medical geneticist, to have your genes screened. It is by no means done automatically at the moment and we, and other people within the field of genetics, are trying to solve this so that there is a better understanding of trying to identify – much like you would do with the other conditions that are actually out there (muscular dystrophy and cystic fibrosis), so that some of the rare obesity syndromes would also be screened. But you could have a monogenic condition, you know carry it, and have a phenotype that’s not super super severe. I’ll give you one example that’s been recently accepted. I don’t think that it is published yet, and it involves a gene called MC4R – the melanocortin four receptor. It sits in one of the fat sensing pathways in our brain. Your brain needs to know two pieces of information to influence food intake – it needs to know how much fat you have, because how much fat you have is how long you would last without food in the wild, and it needs to know the short-term signals, so what you’re currently eating and when you have just eaten (and these signals are going to come from your gut). So, your brain has, in effect, a fat sensing pathway and a short-term gut signal sensing pathway. This is very simplistic but broadly speaking this is true.

So, MC4R sits in the fat sensing pathway, and we know that mutations in MC4R are the commonest monogenic form of obesity that’s out there. Now, you might say well how common is common. So, in our paper, we report that certainly within the UK, one in 330 people within this country (and it’s probably safe to say that similar numbers will be true throughout, but let’s just focus in this country) will carry at least one copy of a mutated MC4R gene which makes them at 18-years-old on average 18 kilos heavier than a person not carrying the mutation. Eighteen kilos is a lot of kilos. Like I said I’m 75 kilos, so 18 kilos is nearly a quarter of my body weight. So that’s a lot of weight and that’s one in every 330 people so you’d be wondering around being 25% heavier than me, but not really know why. And now we know that a likely reason is that you have a mutation in MCR4.

FLG: For those examples where you know it’s a single defect in a single gene – are they easier to treat, for example leptin deficiency.

Giles: You might think that there is this treatment and yes, it’s true for leptin deficiency. So, leptin is one of the key hormones produced from fat and it is probably the key hormone that your brain senses to determine the amount of fat you have. Leptin is a hormone and so if you don’t have it, you can replace it. But leptin deficiency is incredibly rare – I mean since it was discovered in 1997, there has been around 100 families diagnosed – that’s not a lot of people. But those individuals can be treated with leptin, correct. But leptin can only be used for treating leptin deficiency and nothing else – and that is part of the problem. So there is a treatment for 100 families but what about everyone else?

Well, it’s only very recently I would say 2016/2017 that another treatment has emerged which actually targets this specific pathway, and it actually targets and binds to MC4R. It is a drug called setmelanotide and it binds to MC4R and is in trials at the moment. It has been approved for treatment of leptin receptor deficiency (because obviously if you have a leptin receptor deficiency you’re not going to respond to leptin), POMC deficiency (POMC sits downstream of the leptin receptor and naturally produces the agonist for MC4R – so it sits between the leptin receptor and MC4R) and to treat the enzyme that is required to process POMC into peptides. But those are for very rare forms obesity. What is now under trial is – will the drug work if you only have one working copy of MC4R? Because as I said, one in every 300 or so people will have only one fully working copy of it. So, trials are now ongoing to find out whether setmelanotide is going function properly with only one working copy – you might think yes but the trials have to be done. If you upscale the number of one in 330 people, that’s 200,000 people in the UK, so that’s quite a few people. After they do that and they try it out, they now have a far larger safety profile because then there’s a lot more people than just a couple 100 people. Then, they can begin discussing whether this drug is safe to roll out to people who don’t have monogenic obesity.

And you might wonder – why are they taking these steps and why are they being careful? Look what we’re seeing now with COVID-19 and the fact that there is an accelerated process for actually testing the safety of vaccines. What you now are beginning to see is that you have the safety profile of testing it on 50,000 people, which sounds like a lot, but then the moment you roll out the vaccine to hundreds of millions of people then the very very very rare side effects begin to emerge, and people are beginning to try and understand what these very very rare side effects do. So, I think the reason for companies, like Rhythm Pharmaceuticals who produce setmelanotide, being careful is because they’re saying we’re going to start with the rare because we have a good cost benefit analysis in terms of what are the pros and what are the cons – so these kids are so obese that actually they will really benefit from it. Then if it’s safe, they will move on to something slightly more common and then eventually we will try the whole population which means we’re going to millions of people. So, I think the possibilities of treatment are certainly coming online now.

FLG: If we move away from these rare forms – would you be able to give more detail about the whole pathway and variants within different genes that can cause polygenic obesity.  

Giles: We now know, from a polygenic perspective, of at least 1000 loci that are linked to our feeding behaviour and therefore influence our body weight. And where do these sit? Some of these genes do sit within the fat sensing pathway, so they include MC4R, POMC and the leptin receptor. But we’re not talking about mutations, we are talking about polymorphisms within this pathway. So that’s one classic pathway. Some of them clearly influence very subtly the fat sensing pathway. But that’s just one pathway. You can consider your brain from a feeding and control perspective in three parts. They’re not mutually exclusive from each other. They clearly talk to each other, but I think conceptually is it useful to think of your brain in three parts. The first is the fuel sensing part, so this will include fat sensing and the short-term sensors, and this tends to be an area of the brain called the hypothalamus. So, if you feel where the bridge of your nose is and take it right to the centre of your skull, it sits just above the pituitary gland at the base of the brain – so that’s where the fuel sensor is. Put simply, I have burnt 1000 calories and I need to eat 1000 calories.

The second part of your brain you need to consider is the hindbrain, which is as it says, right where your neck joins your skull. The hindbrain is the more visceral sensing portion of the brain. While hunger tends to be taken care of by the hypothalamus, because it’s there to try and figure out how many calories you have burnt and therefore how many calories do I need to eat, the brain stem is more visceral. It tends to sense fullness, to very full, to oh my God I feel like puking. It actually senses it via circulating hormones, much like the hypothalamus, or direct nerve innovation from peripheral organs. This could be stretch receptors in the stomach or if you have a bad tummy from food poisoning or being too full. Then you can actually have hormonal and direct innovation for the brain to make you stop what you are doing – from feeling slightly ill to actually being ill.

The third part of the brain is the hedonic part of the brain, which makes things feel good, the reward element of the brain. So, you have hunger, you have in effect fullness to pukiness, and then you have the ‘it feels good to actually eat’. And they all talk to each other, and they all influence each other. Clearly how hungry you are will influence how driven you are to eat food, and how full you are will influence that as well. So those 1000 loci that I told you about sit within this entire system and all of us have an individual mix. It’s not purely the fat sensing pathway, or purely the gut sensing pathway, or purely the reward pathway, but some individual mix that makes some people say more yes to food or more no to food. So that, broadly speaking, is how we conceptually think about things.

FLG: Where does the microbiome sit within this pathway?

Giles: I consider the microbiome very much like people consider epigenetics, I put them in the same category. And you might think, ‘Really, how are they the same thing?’. I consider the microbiome and epigenetics as the interface between your genes and the environment. Now remember when I talk about the influence of our genes – it’s not 100%, it’s never 100%. The environment plays such a huge role and I say between 40 and 70%. Part of this interaction is clearly the genes responding to the environment and then trying to adapt. So, the two areas that are measurable in this interaction are, first of all the epigenetics – so how your genes are decorated, influencing whether genes are turned on and turn off and that’s influenced by the environment.

Secondly, our microbiome. There are famously supposed to be more microbiota in our gut then there are cells in our body – that’s a slight overstatement, they are about the same number. But there are still trillions upon trillions, so there’s a lot of them. They are at the frontline because when we eat, in effect our food hits the microbiome before it gets absorbed. So clearly, the microbiome is going to play a huge role in how our food, and the various sources of food, are going be metabolised and digested before they are actually absorbed. So, therefore what we eat, the environment in which we live in, how we are breathing and how healthy we are will influence the microbiome, and in doing so will influence everything else. So that’s where the microbiome in my mind sits – it sits at the interface between our genes and the environment and so therefore, change according to our environment will influence how our genes respond to the environment.

FLG: There are loads of different diets our there – do they have an impact on our microbiome and can long-term dieting impact your epigenetics?

Giles: Both your epigenome and microbiome are volatile, which means they can change, unlike your genes which can’t change. But they change at different rates – so the microbiome is far more volatile than your epigenome which means they can change quite rapidly. Now, do they change long-term? I don’t think there are enough studies that have been done, but I think clearly if you live in a different country for long enough your microbiome will change to reflect the country you’re living in. I am ethnically Chinese. I was born in the UK and raised in California and here, so the diet that I faced and ate, and continue to eat, is going to be very different from the me born in China. So undoubtedly my microbiome, even though genetically I’m Chinese, is going to be very different from the equivalent me sitting in my old fishing village. So, in that scenario it can change and can probably change permanently – so my son’s microbiomes is probably going to reflect me rather than China me’s microbiome.

Now, how quickly does it change? I think it probably changes pretty quickly, the question is whether or not it lasts. So, what would probably change your microbiome quickly is if you make drastic changes – so this is going to be these diets. So, say suddenly if I decided to go keto or paleo or some weird carnivore diet, then the removal of all the fibre and a lot of carbohydrates in my diet will change the bugs in my stomach. Now, what we know from a health perspective is that the healthiest microbiomes that are out there – and you might debate what is the definition of healthy – very variable. The more variation you have in your microbiome the healthier it is (put simplistically) and what gives you the most variation within your gut microbiome is if you eat lots of fibre. Therefore, actually one of the biggest changes that you can make to your diet is if you shift to a largely plant based diet – being vegan in particular – and eating lots of fibre, will also make lots of changes to your microbiome. So undoubtedly changing your diet, big changes, will change your microbiome. Now, if you begin to make more subtle changes, like intermittent fasting or you’re eating less (so you’re not shifting what you’re eating broadly but you’re shifting the amount you eat), then it is less certain what happens to your microbiome. So, you need relatively big shifts in your diet to change your microbiome.

Now, a big problem with the epigenome, aside from it being volatile, is that it is organ specific. So, in order to study the epigenetics of whatever trait or behaviour you are trying to study you need to study the organ that it is going to influence. For example, if you are studying the role of the epigenome in exercise – the tissue you are interested in will be the muscle. Now muscles are slightly painful but are available for small biopsies, so you can study the epigenetics of exercise and actually the epigenetics of exercise is quite a mature field. Equally, you could study type 2 diabetes. In type 2 diabetes you would probably be interested in three tissues – the pancreas (that’s difficult to get to but not impossible), your fat (you can definitely get stuff out) and also your muscles. The problem with studying the epigenetics of BMI, and hence food intake, is you have to get to the brain and that, as you can imagine, is difficult. Impossible actually at the moment for living persons. So, the problem with trying to understand the epigenetics of obesity is that it’s hugely difficult. We do have access to post-mortem brains but that is quite a big distance from the nutritional insult in question – it’s quite a big difference in length of time. So, trying to understand that is going to be difficult. Will having a long-term poor diet influence your epigenome? I think the answer is undoubtedly yes. To what degree and to what degree is it reversible, that is yet to be known. But yes, I do think that, as with any other insult that is actually out there, could be stress or trauma or a long-term unhealthy diet, I think all of these can influence your epigenome.

FLG: As you said our diet is like an insult to the body – so surely back and forth switching between diets is unhealthy?

Giles: I think it depends on what you are doing. I think as long as it’s not anything super extreme it’s probably fine for you because then that introduces the variability. I worry about the people who go on a long-term carnivore diet. I can’t imagine why that is a thing, how do you survive? You know we are not carnivores, yet people do do this, and I’m waiting for them to keel over and die because there’s been no long-term safety profiles for that. I worry about that, when you make some big long-term changes and whether that does then influence your long-term health. Whereas I think if you’re flitting between diets, it is not going to help with your weight loss, but from your microbiome perspective that is actually probably good.

Now if your question is what happens if you weight fluctuates a lot – yo yo dieting – is that going to end up with vicious cycle? I think anecdotally evidence certainly points to it. You know people say on yo yo diets that whenever I yo yo in the upward scenario I yoyo higher than I was before. I think there is certainly some data out there that shows that this could be the situation, that you could trend upwards. There’s some kind of vicious cycle that is going on there if you continue yo yo dieting. I think that is an area of hot research, I think people are understandably concerned because a lot of yo yo dieting going on out there. So, people are trying to understand what happens with long-term yo yo dieting.

FLG: Different diets suit people differently. However, I did see recently companies selling these so-called DNA-based diets – what is the validity behind these?

Giles: I will give you the short answer and then I will give you the long answer. The short answer is don’t waste your money. Now before I get sued, there are many genetic companies out there and what these companies are not doing is they are not lying about the genes that are there. Our ability to sequence genomes and genotype our various genes is no longer the bottleneck. The technologies are mature, and they are of relatively low price, certainly compared to 10-15 years ago – they are still expensive but not stupid expensive, not out of reach of a middle-class person. A genetic test costs, these days, anywhere from £100 to £300 – so not cheap but I think a lot of us can afford that. The issue is not the genetic information, the issue is in interpreting the information. And all of the genetic companies make the mistake of taking bona fide associations in genes which have been done through population level studies – these have thousands or millions of people and found that on average if you carry this variant, you are more likely or less likely to do something (whatever that something might be). They take that one piece of information and try and turn it into a prediction. Now no one is lying, nobody’s lying here. They are misinterpreting and using the data for something that it is not supposed to be used for – fundamentally misunderstanding the difference between population level risk and individual prediction.

I’ll give you an analogy – this is the analogy which I use when I teach first year medics. It is slightly facetious, but I don’t think it is because I think it actually hammers home the point. Let’s take pregnancy. So, we understand the biology very well, many billions of women have got pregnant, so we understand reproductive biology. And we know that the younger you are as a woman, the more likely you are, or the higher the risk you are, in getting pregnant and then as you get older the risk or likelihood drops until you get past menopause and then it becomes zero. And it’s a curve and you can draw it because billions of women have got pregnant before – it is very robust. But does this mean that I can pick a random 34-year-old woman, for example, off the street and predict if she is going to be able to become pregnant based solely on her age. The answer is no. I am not going to be able to predict this. Why? That particular woman could be infertile for a myriad of different biological reasons, so the likelihood is zero. Or she could be ovulating that day so suddenly it is very high. So, the point being is that even with a very powerful biological understanding of biological risk – age with pregnancy – we cannot predict whether or not someone will be pregnant without additional biological tests. So, if we take that analogy to what these genetic testing companies are trying to do, they are taking one risk factor (equivalent to age) and trying to make a prediction. It is not possible.

I do think there is a value to these genetic scores – but I think they have to be taken as one of the risk factors. So, for example, if you go to the doctor and you say, ‘I want to know if I’m at risk of a disease (e.g. getting fat)’, they can say, ‘What do your parents look like? What’s your blood pressure? What’s your cholesterol levels? What is your genetic risk score?’. I think it’s useful as one of the risk factors where then when you begin to triangulate the issue, you can say ‘Okay Mr Yeo I think you need to be careful because of this, that and the other’. This is going to be true for whether or not I am going to become obese, whether or not I will be responsive to a high protein diet, whether or not I’d be better with a high fibre diet – it’s going be true for all of the above. The genetic testing companies just need to get a better handle on what the data can be used for – it is useful information, but they must not over interpret or misinterpret and misuse the information for what it is not designed to be used for.

FLG: The environment plays such a big role and these companies do not take that into account.

Giles: No, they do not. So, I got paid by a major tabloid newspaper to take one of these tests and write a review. They asked for my name – thank you, Giles – for my age, they asked for my sex (although they can figure that out from my genetic test). But do you know what, that was pretty much it. I did put in my weight and stuff, but there was not a lot of other information. They did not ask for my socioeconomic class, I don’t know what number I can give them for that – middle – that’s not an answer. Then they came back with all of these predictions – some of the predictions were mad! The ability of my liver to detox – stop – now that is entirely BS. And this is an issue.

Now what is interesting is that there are going to be several monogenic traits out there, that are predictive. So, as I have said, I am ethnically Chinese, so I am lactose intolerant and that is going to be down to a single gene and that is measurable. So, if I don’t have the polymorphism that 85 to 90% of White Northern Europeans have (I don’t), then I can’t actually handle large amounts of lactose as an adult – that’s predictive. Another thing that is predictive is the ability to handle alcohol. This is the result of two genes – alcohol dehydrogenase and acetaldehyde dehydrogenase – but single polymorphisms in each of them. The third is the ability to handle caffeine, so that is also predictive. And what these companies do is they always include those three traits and so if you are not a geneticist, which most people are not, and you pay £100-200 for your score and you look and see ‘Oh my god I am lactose intolerant’ or ‘Oh I can’t drink alcohol’, you think they are right. Then you look at all the other traits that are telling you that you should be doing aerobic exercise versus resistance exercise – PLEASE! The problem is that you believe it because you think they have got it right with the milk and alcohol.

They mix in monogenic traits with complex polygenic traits but altogether in the same report, and so therefore the consumer finds it difficult to try and parse out what is genuinely predictive and what is just population level risk. Now you might say ‘What harm will it do?’. Now if they said I have a risk for obesity and therefore I need to eat better – what harm could it to? But what happens if, for example, your genetic score says that you have low risk of obesity or low risk of some other disease – do you then say, ‘Here we go boys let’s just do whatever the hell we want’? If you eat all the pies, you will still end up heavier even if you have no risk factors. And that is the problem. If you look at the data and millions of people take it, how do you interpret the data and how do you use the data? Do you use the data dangerously? In one sense, I know it doesn’t do any harm. The problem is in the reverse, when you suddenly say, ‘I can do what I want, I’m protected’. That is not how genetics works.  

FLG: In terms of obesity, the environment plays such a huge role. Obesity over the past few decades has increased, particularly among the younger generation. What do you think needs to be done to combat this obesity crisis?

Giles: So, we can’t change our genes – yet. I am a geneticist, so I’m interested in biological variability in response to the environment. But I think the real long-term way we are going to solve the problem is to fix the environment – it’s to change our food environment. I’m not countenancing going back to the 1950s, this is not what I’m saying. But I think we do need to have better policy decisions and a better built environment in order to sort out the food environment. The other thing which is crucial, and I think it be is illustrative of the role of the environment, is socioeconomic class. If you are in some lower income country and you are in poverty what happens is, if you don’t have enough food, you’re skinny. So now you’re at risk of all kinds of diseases because you are malnourished. But we are now in a situation where they’re actually more people dying from eating too much than dying from eating too little. So, in this country, same as it is in Western Europe and in the US and Australia, is that the lower you are on the socioeconomic ladder, the more likely you are to have obesity – so it’s become the other way round. This is because of the access to cheaper calories.

I’ll give you one clear example of the interaction between socioeconomic class and genes. A colleague of mine, Clare Llewellyn at UCL, runs a classic twin study called the GEMINI study, and she’s interested, like me, in body weight. But the unique characteristic of this study is that Clare has food security measures in her twins. So, she knows about the food security and how good, in effect, the access to quality diets is within their households. So, here’s the really interesting thing which even as a geneticist and understanding how it works, it still makes me pause for breath. So, what happens here is using her numbers and using her twins, is that in middle class households, her twins heritability of body weight is 40%. In order to max out the heritability of body weight she needs to go to the households with the lowest food security, with the highest food insecurity, so these are people that tend to be on the lower end of the ladder of socioeconomic class. Suddenly, you put yourself in a poor environment and these twins max out their heritability of body weight at 70%. So, what you see there is that if you happen to be susceptible for all the genes that I just spoke about and you place yourself in a poorer food environment, you max out your genetic burden or potential. Whereas if you are able to make better decisions, or you are in a position to make better decisions, then you are able to mitigate against some of your genetic risk.

So, to put simply – if we solve poverty in this country, we could fix up to 40% of our heritability of obesity – far more than any drug that’s out there and we can’t change our genes. I think that should give everyone pause for thought -that if we could fix poverty, we could fix a large amount of obesity and associated diseases that are out there. But it doesn’t mean that if you have the susceptibility to be obese, you won’t be more obese if you are rich or poor. But being rich or poor plays a huge role even if you might be protected or you might be susceptible to being obese.

People have done the calculations. In some places, you can get up to 900 calories for under a pound. Many of us do have a choice – like I think we need to eat more lentils or more tofu, whatever. I can drive down to the Chinese supermarket or whole food shop and buy lentils and cook it. (A) I have the time, (B) I know what to do with the lentils and (C) I have the money to go to the whole food shop and actually pick up whatever lentils I want. But if you are Mrs. Smith and you need to work two minimum wage jobs in order to put food on the table, so therefore you have no time because you are working two jobs and you need to feed your kids and a pound can buy you 900 calories (you can go to Iceland to get four frozen pizzas for £4.99) – who are we to judge? And this is the problem. We have our choices; we are here and you’re absolutely right – we can make the choices. But if you don’t have the choices and if you can’t make the choices – then what the hell are you meant to do?

So, Michael Marmot has written a lot about food security in the country and has actually calculated that for some people in the bottom decile, in the bottom 10% in terms of socioeconomic class in this country, if they followed Public Health England’s “healthy diet”, they would spend nearly 70% of their income on the food. Michael makes it very clear that he’s not saying that we shouldn’t be eating healthier, but if we’re giving advice to people who will never ever be able to follow it because of their economic situation, what the hell are we doing?! So, I think we need to consider our whole food system for everyone of all socioeconomic classes. And the bottom line is this: in order to really solve obesity, we need to make the healthier choice, the cheaper more convenient choice. Human beings will always take the easy way because that’s what we are designed to do. Until you make the healthy choice the easier and more convenient and cheaper choice, we will not fix the problem.

FLG: Thank you so much. I think in society we are also so quick to blame (and judge) other people and I think as a society we need to take a step back and really look at the wider picture.

More on these topics

Down the Rabbit Hole / Interview / Obesity