Do "Calories" Explain Weight Gain? Response to Science is Dope.

[TLDR: Is weight loss really just a matter of energy balance? Let's really test this hypothesis. For those who are interested in joining the April Reversing Autoimmune cohort, sign ups are now open.]

A while ago I released an interview with an Indian YouTuber called Pranav Radhakrishnan. In that video we talked about the persistent popularity of pseudo science. We especially picked on people like Sadhguru, partly because they’re super easy to pick on and partly because some of the dietary advice they’re peddling is unproven and potentially dangerous.

A few weeks ago Pranav released another video called "How I lost 10 kilos while eating more and staying healthy". As a health person myself and as someone who has a lot of respect for Pranav I was eager to watch it. But when I watched it my jaw dropped. How can one person get so many things wrong? Does this guy really know nothing about the scientific method after publishing videos about science for years?

My first inclination when I saw it was to just ignore it. There’s a lot of garbage on the internet and why should I pick on this one video? But this might be a good opportunity not only to debunk the “calories in, calories out” fallacy, and also to talk about elementary mistakes that even good scientists make. Pranav is not, as far as I know, a scientist. But he is effectively a science journalist and in order to be a good science journalist you have to understand the basics of science. What he says at the conclusion of his video is inexcusable, not just from a scientist but from anyone who has any knowledge of rising rates of chronic disease globally and the role of the food industry in those diseases. 

You Can’t Eat a Calorie!

To be clear (and Pranav states this clearly) there’s nothing original in this video. If you’ve ever watched anything by Layne Norton or talked to your local gym bros, you know what’s in this video.

The basic hypothesis is as follows: If I eat fewer “calories” I’ll lose weight. Straight away we have a problem. What’s a calorie? According to Miriam Webster, the definition of a calorie is as follows:  the amount of heat required at a pressure of one atmosphere to raise the temperature of one gram of water one degree Celsius that is equal to about 4.19 joules.

So calories are heat. I challenge Pranav or anyone to eat heat. We eat food (which is matter), not heat (energy).

If we didn’t want to eat the food, I suppose we could set fire to it, thus converting it to energy. If we have access to a bomb calorimeter we can set fire to the food in a way that will raise the temperature of the surrounding water. By the above definition, if burning that food raises the temperature of the water by one degree, it has one calorie. In this way, we can determine how many "calories" a given food will have in a bomb calorimeter - in other words how much heat it will give off if you use it to toast marshmallows. 

Is that relevant to the human body? Do you have a bomb calorimeter inside you that cares about the number of calories something has? Do you burn everything you eat in some inner bonfire? Of course not. There some things - like say cardboard - that do have calories, in fact a lot of calories, cardboard burns pretty well, but those calories are not accessible to humans. (This is why Pranav talks about fibre as essentially "free"; fibre has calories but can't be digested by humans.) And there are other things, like say plutonium, that have a whole lot of calories - something like 20 billion calories in a single gram - but will not result in any equivalent mass gain in a biological system. (Disclaimer: please don't test what happens when you ingest plutonium.)

Fixing the “Calories” hypothesis

So calories are the wrong term. But we all (sort of) know what he means. So how can we steel man this calories thing and make it make sense?

A more accurate way to think about the energy in food is as potential chemical energy. All matter - grass, cardboard, plutonium or whatever - is held together by bonds of various kinds. When those bonds are uncoupled, energy is released. The extent to which that potential chemical energy can be converted into actual energy depends on the machine that’s using the matter. For example, I have no ability to use the potential energy in petrol, but my car does. My car has no way to access any of the potential energy in a banana, but my body does.

While my body can access some of the potential energy in a banana, it can’t access all of it. Some of that banana is fibre which I can’t digest and may or may not be turned into fatty acids (from which I might be able to derive some energy) depending on the state of the microbes in my colon. I also need to use some energy in order to digest the banana, so even if I could access all the energy in it, I’d have to subtract the energy that my body is using to digest the banana in order to get a complete picture. This is related to the thermic effect of food (essentially our body heats up slightly during digestion) and that’s an effect which is known to change depending on who is doing the eating of that food. In the case of the banana, much of that potential energy comes in the form of sugars which enter my blood stream during the digestion process. Because sugar is highly toxic above a certain concentration (we have about two teaspoons of sugar in our body at all times - a banana has six) the body has to get rid of that sugar quickly using a variety of hormones but especially the hormone insulin which binds to the sugar and takes it for storage in the glycogen which is either in the liver or the muscle tissue. If those storage depots are full, the sugar gets converted to fat and stored in our fat stores.

I hope it’s clear from the above that metabolism is a complicated process and reducing it to “eat less, move more and you lose weight” is at best a gross oversimplification. Pranav is being unscientific about his hypothesis, but let’s give him the benefit of the doubt. Let’s try to formulate his hypothesis for him correctly.

A scientific version of the CICO hypothesis would be as follows: If a given human consumes less energy, measured in the potential chemical energy (not calories) of human-appropriate foods (not plutonium) than the amount of energy that the human expends (BMR + NEAT + Thermic effect of food + Exercise), that person will lose mass (especially fat mass, because that's what most of us are interested in). Conversely, if a given human consumes more potential chemical energy than they expend, they will gain mass. 

Does that sound like a fair version of Pranav's hypothesis? Do you think it’s true?

One thing that Pranav doesn’t go into details on here is the expenditure side. As I’ve done above, you calculate the expenditure side by adding the total of the Basal Metabolic Rate (BMR), the thermic effect of food (energy expended in digestion), exercise, as well as something called Non-Exercise Activity Thermogenesis (NEAT). Most people focus on exercise, but when you look at the data, exercise doesn’t count for much.

Most of the energy expenditure for humans is something called BMR or basal metabolic rate, meaning how much energy our body is using at rest. This is just to keep our heart beating, our temperature steady and so on.

One reason that calorie restricted diets tend not to work for most people in the long term is because of changes to our BMR. When we restrict food, the body actually lowers its BMR to compensate. In other words you could be consuming the same amount of potential energy “calories” as you did when you were losing weight and now be gaining weight because your body is running “colder” as a result of the diet. The body is burning less energy than it used to at rest because the body’s trying to maintain your same weight even though you’re consuming less food. The body’s tendency to maintain weight is called homeostasis, something else Pranav doesn’t get into. One of the most revelatory papers for me on this was the Biggest Loser Study which showed contestants' metabolisms took a nose dive after losing weight (through counting calories) which lead to nearly all of the contestants regaining the weight (and sometimes more). 

Scientific Method 101

All of this is interesting, but does it disprove Pranav’s basic point?

So Pranav has a hypothesis that weight is related to energy intake and expenditure. We’ve done our best to formulate it in a way that it at least makes some sense. Once Pranav (or anyone) has a hypothesis, what does the scientific method tell us to do? The tempting thing would be to look about for evidence that confirms the hypothesis. Looking around, I can certainly find people who’ve been able to lose weight by following Pranav’s advice. Does that make it true?

To understand the problem with this process, let’s consider a different hypothesis, namely the following: “All birds fly”. Looking around my environment here in South Africa I can easily find evidence to support this hypothesis. Doves fly, pigeons fly, eagles fly, owls fly and so on. So can I therefore conclude that all birds fly?

If you want to prove that all birds fly, finding evidence of flying birds does not help you. The way your hypothesis is formulated, even one example of a creature that is a bird and doesn’t fly disproves your hypothesis. You therefore must make a good faith attempt to disprove that hypothesis. If you make no attempt to disprove your own hypothesis you’re not engaging in science at all. In fact when we look at some of the biggest scientific disasters of the last century - most notably the vaccine and autism related (fraudulent) work of Dr. Andrew Wakefield - was the result of people looking to prove (instead of disprove) their hypothesis. And of course once you make a good faith attempt to find birds that don’t fly, you will quickly come across ostriches and emus and penguins, any one of which would have been enough to disprove your hypothesis.

So in the energy balance model, the question we have to ask is the following: is there data in the literature talking about even one patient who loses weight even though they are documented to be eating an energy (or a calorie) surplus?

And the answer is absolutely there is data like that, not just about one patient but about a whole class of patients. I’m talking here about people with type 1 diabetes (T1D). It doesn’t matter what or how much you feed an untreated person with T1D. They will waste away and eventually die from malnutrition. This is because they have an autoimmune disease whereby the beta islet cells in the pancreas, the ones that produce the hormone insulin, have been knocked out. Feed a person with untreated T1D seven thousand “calories” per day or more and they will still lose weight.

Now take that same person, inject them with insulin, and feed them only three thousand calories per day. What happens? They gain weight. Sometimes a lot of weight. So “calories” had nothing to do with weight gain or loss; food mass (potential chemical energy, if you like) can only be converted to human mass in the presence of the hormone insulin.

And those of us who work with people in the hospital system see this every day. When people are given steroids or other classes of drug which increase insulin levels, they immediately gain weight. In some cases you can be sure that nothing has changed in their diet and exercise routines because they’re bed ridden. Yet when the doctor puts them on a medication that is known to spike insulin, they gain weight. This is no secret; in fact there are academic papers such as this one that specifically look at the interaction between certain drugs, weight gain and insulin.

So why does counting calories work for Pranav? First of all, not to be unkind, but he doesn’t actually look healthy in this video. He seems to have a few skin issues going on, which is a sign that his gut health isn’t great. So if I were him, I wouldn’t be so quick to brag here. He might be losing weight, but there might be other things going on that indicate a diseased state. I was thin when I was his age too, but I wasn’t healthy.

But I do believe him when he says that calorie cutting works for him in terms of weight loss. Why does it work? Because every diet, including “calorie” cutting diets, tends to decrease insulin levels. You can look for example at the Gardner 2007 study I blogged about a few weeks ago.

All four of those diets including the “calorie” counting ones were able to decrease fasting insulin. And remember that all of the diets did show some weight loss, but only the low carb Atkins diet (which did not involve counting “calories”) was statistically significant weight loss over 12 months. It’s no coincidence that the Atkins diet had the biggest effect on insulin levels as well.

So what’s the takeaway here? Pranav is right in that counting calories can work for some people. But calories are not the mechanism of action. Counting calories is one strategy to reduce insulin levels and high insulin levels are what was necessary for the person to gain the weight in the first place.

Why is it important to understand the mechanism of action? For many people counting calories is difficult. It can lead to the crash dieting problem, where you crash your basal metabolic rate, meaning you’re just going to have to eat less and less every time you diet. It's not fun to go around hungry all the time. Pranav's answer to the hunger problem is to eat more fibre which may work for some people. My guess from looking at his skin is that his high fibre diet is destroying his digestive health.

Once we understand that it’s insulin and not food energy that’s at work here, then we can find other diets, including but not limited to high fat low carb diets, that can reduce insulin as effectively or even more effectively than calorie counting. And that 2007 Gardner study indicates that High Fat Low Carb diets are a more sustainable and a more effective way to lose weight than calorie counting diets.

The broader lesson here is this: If you want to be good scientist you look at the outliers. You can lose weight by eating less. That’s common sense and in keeping with most of our personal experience. But are there populations that eat more and can’t gain weight? Are there populations - for example women going through menopause, pregnant women, or maybe even some subsets of children - who eat less or the same amount of calories but gain more weight? And what’s going on with those populations?

If your theory cannot account for the outliers, it’s not correct. In the case of the insulin theory of weight gain, it is simply impossible for someone to gain weight without insulin and it’s also impossible to lose weight if your insulin levels are stubbornly high, which is something I see with my clients.

If you want to lose or gain weight, you should really be focussing on your hormones, in particular insulin, and not so much the quantities on your plate. Those things are related - if you eat less you most likely will decrease your insulin levels. But that’s not the only way - in my opinion it's a pretty bad way (though it may work for some people). 

At the end of the video, Pranav says the one thing that’s actually dangerous. He basically says you can eat the candy, you can eat the junk food you can eat anything you like as long as it fits your macros. The junk food is not the problem, he says, it’s the quantity of the junk food.

This is insanely bad advice. To give one example, potato chips are designed in a lab by food scientists who are deliberately trying to trigger the parts of the brain associated with addiction. They are trying to make a substance that will make you eat the whole packet and then go out and get another one because you’re addicted. They're doing that because they're trying to make money. Yes there are agencies like the FDA that are supposed to regulate these chemicals, but most of the people who staff the FDA are waiting for a day they can get a better paying job with these industries. As such they keep the bar as low as possible when it comes to additives and other chemicals in food. They certainly don’t study long-term effects of these chemicals, nor do they do a particularly good job of keeping out food that can poison people in the short-term out of our food supply. (If you doubt this, just do a web search for “food recalls in my area”. In my area, a number of children died last year from eating contaminated packaged foods.)

If you’re the kind of person who can moderate your intake of these Frankenfoods, they may be ok in small quantities and as an occasional treat. But for the vast majority of us, myself included, it helps to not even think of these things as food. For me they are industrially processed, food-like substances that are better avoided if at all possible.

By the way that’s not a controversial position. It just makes sense to avoid foods that come from a factory and have ingredients you can’t pronounce. From a weight-gain perspective, these foods are known to spike insulin even more than a banana. So let’s do another thought experiment: What would you do if it was your job to market those industrially processed foods? How would you protect the billions of dollars that food companies make selling those products?

I’m not sure what you would do, but I can tell you what Coca Cola did some years ago. “You can still have some of our industrially processed sugar water,” they said, “you just have to exercise for a bit after you do so.” In other words, it’s not about the highly addictive junk we’re selling you. You’re just eating too much of it and not exercising enough.

If you’re an advertiser, this puts the blame on the consumer and allows you to continue marketing your product. You can find evidence to support it - there are people (maybe Pranav included) who can moderate their junk food intake and seem to do ok, at least in the short term. But would we ask someone who suffers from alcoholism to moderate their intake of whiskey? Of course not. To suffer from alcoholism means that you have lost the ability to moderate your intake. And in a world designed to breed carboholism, “eat the doughnut if you feel like it” is potentially criminal advice.

In fairness, Pranav isn’t the only person giving this terrible advice. Pretty much every mainstream nutritionist and dietician I’ve run into has been handing it out for years. That's one reason the results are so bad - obesity rates and diabetes rates and autoimmune rates are climbing because the nutrition profession has been handing out this terrible advice for decades. 

If you want to heal your body, fix your hormones. “Calories” are a marketing gimmick designed to shift responsibility onto the consumer and justify the production of "foods" that have no nutritive value and are highly addictive.