Regulars here have probably noticed that I'm a bit flippant and dismissive when it comes to people discussing CICO or calories. Those without extensive backgrounds in exploring different dietary theories probably scratch their heads when they see this, since the discussion of calories is so pervasive in popular diet and weight loss culture and is so strongly pushed by nutritionists and the medical community at large. I believe that I'm dismissive of it for good reason, but thought others who haven't explored the dark corners that I have may wonder what my rationale is and whether it's grounded in logic or magical thinking. Hopefully this post will give the reader something interesting to chew on (and maybe even swallow). Note that I'll be purposely oversimplifying several things here and leaving out certain details to avoid making this more complicated than is necessary. I want my descriptions to be something a non-technical person can follow without having to look up and learn unfamiliar jargon.
In general, energy conservation principles have been highly successful in physics, chemistry, and represent one of the core assumptions behind thermodynamics. If energy conservation were to be broken then thermodynamics would fail and that hasn't happened thus far. When people speak of calories, they're talking about a measure of energy stored in the foods we eat. Specifically, if you take that food and combust (burn) it, a researcher can measure how much heat energy is released using a device called a bomb calorimeter. They start with a piece of food, apply a certain amount of energy to it (in order to burn it), and then measure how many degrees the whole thing heated up. If it heats up more than would have been expected from just the energy applied to it then that additional energy is said to be the "food energy" or calories that were contained in the food. For example if enough energy was applied to heat up the sample by 50 degrees, but it instead heats up by 70 degrees, the extra 20 degrees of heat would be assumed to have come from the food burning. That 20 degree difference would be what the "calories" in a food are calculated from.
If our stomachs were blast furnaces and the goal was only to keep our bodies warm then measuring calories consumed would actually make a lot of sense, since converting fuel to heat is a 100% efficient process, which is to say that every calorie of energy stored in something can be converted to heat if desired. However, since our bodies also do so many other things beyond staying warm, it becomes necessary to unpack the concepts of work and efficiency. Driving a car could be a helpful analogy here. It turns out that of the energy stored in the gasoline, when it is burned in your engine, only about 20% to 30% of it can be captured in the form of useful work (e.g. making your car move), with the rest of the energy being "waste heat" that must be eliminated through your car's radiator (to avoid the engine heating up too much). If you study thermodynamics, you'll discover that any process where energy is used to perform work has that same problem in that you'll never get 100% efficiency out of such a process. Some of that energy will always be "wasted" as heat. There's a concept in thermodynamics called "entropy" that partially represents that frustrating state of affairs. For every CICO genius that likes to glibly spit out that "CICO is just the first law of thermodynamics" (conservation of energy), they conveniently seem to be ignorant of the second law of thermodynamics (entropy).
To unpack that last part a little more, if the food you ate just went into a "calorie bin" (perhaps a secret compartment underneath your stomach) and then your body burned it when necessary (maybe at 50% efficiency), then CICO would be a highly useful description. You'd just apply the 50% correction factor and it would work beautifully. In reality, different macros come with different overheads. For example, protein (despite being clocked at 4 calories/gram, just like carbohydrate) takes significantly more energy to process in your body, such that when it comes to doing work your body can't do as much work with 100 grams of protein as it could with 100 grams of carbohydrate. Now, this isn't shocking news to the CICO people, who fully acknowledge that to be the case. They choose to account for it as an increase in energy expenditure, which is a perfectly acceptable way to account for it within the CICO framework. The reason I bring it up is because it's only one minor example of the calories number printed on the side of the box not properly corresponding to the net result of you eating it. In fact, there's a phenomena known as "rabbit starvation" (Google for more info) which can occur when a person consumes too much protein (relative to carbohydrate and fat). In rabbit starvation, no matter how much protein that person consumes, their body is unable to effectively use it or store it, leading to them wasting away. This is a simple example of what I would dub hormonal influence overriding CICO. Although all energy is ultimately accounted for (nothing magical is happening here), the body is running in a way where too much energy is being wasted, leading to a starvation effect.
Another clue that the CICO story might not be the best framework for predicting weight changes would be type 1 diabetes. In this condition (prior to insulin injections being available), the body becomes unable to produce insulin. This leads to a vicious cycle in which other hormones cause the body to rapidly waste and dump energy. Specifically, blood sugar goes sky high, but none of it can get into the cells of your body to feed them (due to the lack of insulin). Your cells signal to the liver that they're starving for sugar, so instead of diverting into other fuel types, the liver just keeps making more sugar. Meanwhile, the kidneys filter it out of your blood and into your urine to try to save your other organs from the risks of excessively high blood sugar. If your body was instead smart enough to stop making more sugar and instead meet its energy needs using fat and ketones, a type 1 diabetic could likely survive much longer without insulin. Instead, their hormonal signaling is trapped by high glucagon levels and rapid weight loss and wasting away ensues. This claim (that high glucagon is ultimately the trap) is proven by recent studies involving the use of glucagon suppressing drugs in type 1 diabetic rats. When given the drug those rats to survive and live normal lives at normal blood sugar levels!
Yet another clue is the phenomena known as "fat fasting." In fat fasting a person consumes only fat (with very minimal carbohydrate or protein) for some period of time. Because of the lack of carbohydrate and protein, that person's insulin levels stay lower than normal. They wouldn't get as low as a type 1 diabetic, of course. Also their livers would produce some carbohydrate for them, such that their blood sugar levels wouldn't drop excessively low. I don't endorse this particular diet, but just note that engaging in it does lead to weight loss, despite the massive numbers of calories that person consumes while on it.
I could keep going here, but I think I've made my point that hormonal influences are ultimately what drive weight change results. At no point is CICO ever violated in any of those examples, it just fails to be a useful tool in predicting weight changes when hormones are manipulated. Now had those people been hooked up to sophisticated machinery capable of monitoring and detecting changes in "calories out," as I said, CICO would hold and be proven true. Unfortunately, most of us don't have that equipment laying around in our spare bedrooms. If we did, we could try different behaviors or diets and see which ones bumped up calories out, but alas, that is not an option. We're mostly limited to watching a number on the scale and noting what behaviors tend to make it go up over time and what behaviors tend to make it go down. CICO is perfect at explaining weight loss results after the fact. If you lost Y pounds last month then by definition your calorie deficit averaged X calories per day. However, if you tell me that you're going to eat a certain number of calories every day for the next month, without knowing the specific foods you'll be eating, it would be impossible for me to predict the end result from calories alone. For all I know, you could be fat fasting that month, or you could be consuming things that raise or lower your metabolic rate. People encountering weight loss stalls are well aware of this frustrating state of affairs.
One last thing I'll add before closing is that many people assume that GLP weight loss (hormone manipulation) leads to weight loss results fully predicted by calories alone. Although the reduction in how much you're eating certain is a key factor there, it's not the full story. Another major part of the story is that GLPs do slightly tip your metabolism towards a higher rate. This is demonstrated in a rodent study in which the weight of GLP rats VS "calorie matched" rats are tracked over time. The GLP rats were allowed to eat as much chow as they wanted to. Meanwhile, the "calorie matched" rats, instead of getting a GLP, had their meals cut back to match the amount of food that the GLP rats were eating. That is to say that both the GLP and the "calorie matched" rats consumed the same exact number of calories (and it was the exact same food). Despite being calorie-matched, the GLP rats lost significantly more weight (and faster) than the "calorie matched" group. And this was pre-reta, so no glucagon-agonism was occurring, which would have made such an effect even larger.
In general, energy conservation principles have been highly successful in physics, chemistry, and represent one of the core assumptions behind thermodynamics. If energy conservation were to be broken then thermodynamics would fail and that hasn't happened thus far. When people speak of calories, they're talking about a measure of energy stored in the foods we eat. Specifically, if you take that food and combust (burn) it, a researcher can measure how much heat energy is released using a device called a bomb calorimeter. They start with a piece of food, apply a certain amount of energy to it (in order to burn it), and then measure how many degrees the whole thing heated up. If it heats up more than would have been expected from just the energy applied to it then that additional energy is said to be the "food energy" or calories that were contained in the food. For example if enough energy was applied to heat up the sample by 50 degrees, but it instead heats up by 70 degrees, the extra 20 degrees of heat would be assumed to have come from the food burning. That 20 degree difference would be what the "calories" in a food are calculated from.
If our stomachs were blast furnaces and the goal was only to keep our bodies warm then measuring calories consumed would actually make a lot of sense, since converting fuel to heat is a 100% efficient process, which is to say that every calorie of energy stored in something can be converted to heat if desired. However, since our bodies also do so many other things beyond staying warm, it becomes necessary to unpack the concepts of work and efficiency. Driving a car could be a helpful analogy here. It turns out that of the energy stored in the gasoline, when it is burned in your engine, only about 20% to 30% of it can be captured in the form of useful work (e.g. making your car move), with the rest of the energy being "waste heat" that must be eliminated through your car's radiator (to avoid the engine heating up too much). If you study thermodynamics, you'll discover that any process where energy is used to perform work has that same problem in that you'll never get 100% efficiency out of such a process. Some of that energy will always be "wasted" as heat. There's a concept in thermodynamics called "entropy" that partially represents that frustrating state of affairs. For every CICO genius that likes to glibly spit out that "CICO is just the first law of thermodynamics" (conservation of energy), they conveniently seem to be ignorant of the second law of thermodynamics (entropy).
To unpack that last part a little more, if the food you ate just went into a "calorie bin" (perhaps a secret compartment underneath your stomach) and then your body burned it when necessary (maybe at 50% efficiency), then CICO would be a highly useful description. You'd just apply the 50% correction factor and it would work beautifully. In reality, different macros come with different overheads. For example, protein (despite being clocked at 4 calories/gram, just like carbohydrate) takes significantly more energy to process in your body, such that when it comes to doing work your body can't do as much work with 100 grams of protein as it could with 100 grams of carbohydrate. Now, this isn't shocking news to the CICO people, who fully acknowledge that to be the case. They choose to account for it as an increase in energy expenditure, which is a perfectly acceptable way to account for it within the CICO framework. The reason I bring it up is because it's only one minor example of the calories number printed on the side of the box not properly corresponding to the net result of you eating it. In fact, there's a phenomena known as "rabbit starvation" (Google for more info) which can occur when a person consumes too much protein (relative to carbohydrate and fat). In rabbit starvation, no matter how much protein that person consumes, their body is unable to effectively use it or store it, leading to them wasting away. This is a simple example of what I would dub hormonal influence overriding CICO. Although all energy is ultimately accounted for (nothing magical is happening here), the body is running in a way where too much energy is being wasted, leading to a starvation effect.
Another clue that the CICO story might not be the best framework for predicting weight changes would be type 1 diabetes. In this condition (prior to insulin injections being available), the body becomes unable to produce insulin. This leads to a vicious cycle in which other hormones cause the body to rapidly waste and dump energy. Specifically, blood sugar goes sky high, but none of it can get into the cells of your body to feed them (due to the lack of insulin). Your cells signal to the liver that they're starving for sugar, so instead of diverting into other fuel types, the liver just keeps making more sugar. Meanwhile, the kidneys filter it out of your blood and into your urine to try to save your other organs from the risks of excessively high blood sugar. If your body was instead smart enough to stop making more sugar and instead meet its energy needs using fat and ketones, a type 1 diabetic could likely survive much longer without insulin. Instead, their hormonal signaling is trapped by high glucagon levels and rapid weight loss and wasting away ensues. This claim (that high glucagon is ultimately the trap) is proven by recent studies involving the use of glucagon suppressing drugs in type 1 diabetic rats. When given the drug those rats to survive and live normal lives at normal blood sugar levels!
Yet another clue is the phenomena known as "fat fasting." In fat fasting a person consumes only fat (with very minimal carbohydrate or protein) for some period of time. Because of the lack of carbohydrate and protein, that person's insulin levels stay lower than normal. They wouldn't get as low as a type 1 diabetic, of course. Also their livers would produce some carbohydrate for them, such that their blood sugar levels wouldn't drop excessively low. I don't endorse this particular diet, but just note that engaging in it does lead to weight loss, despite the massive numbers of calories that person consumes while on it.
I could keep going here, but I think I've made my point that hormonal influences are ultimately what drive weight change results. At no point is CICO ever violated in any of those examples, it just fails to be a useful tool in predicting weight changes when hormones are manipulated. Now had those people been hooked up to sophisticated machinery capable of monitoring and detecting changes in "calories out," as I said, CICO would hold and be proven true. Unfortunately, most of us don't have that equipment laying around in our spare bedrooms. If we did, we could try different behaviors or diets and see which ones bumped up calories out, but alas, that is not an option. We're mostly limited to watching a number on the scale and noting what behaviors tend to make it go up over time and what behaviors tend to make it go down. CICO is perfect at explaining weight loss results after the fact. If you lost Y pounds last month then by definition your calorie deficit averaged X calories per day. However, if you tell me that you're going to eat a certain number of calories every day for the next month, without knowing the specific foods you'll be eating, it would be impossible for me to predict the end result from calories alone. For all I know, you could be fat fasting that month, or you could be consuming things that raise or lower your metabolic rate. People encountering weight loss stalls are well aware of this frustrating state of affairs.
One last thing I'll add before closing is that many people assume that GLP weight loss (hormone manipulation) leads to weight loss results fully predicted by calories alone. Although the reduction in how much you're eating certain is a key factor there, it's not the full story. Another major part of the story is that GLPs do slightly tip your metabolism towards a higher rate. This is demonstrated in a rodent study in which the weight of GLP rats VS "calorie matched" rats are tracked over time. The GLP rats were allowed to eat as much chow as they wanted to. Meanwhile, the "calorie matched" rats, instead of getting a GLP, had their meals cut back to match the amount of food that the GLP rats were eating. That is to say that both the GLP and the "calorie matched" rats consumed the same exact number of calories (and it was the exact same food). Despite being calorie-matched, the GLP rats lost significantly more weight (and faster) than the "calorie matched" group. And this was pre-reta, so no glucagon-agonism was occurring, which would have made such an effect even larger.
