The Science of Everything Podcast - Episode 153: Low Carb Diets

Episode Date: June 9, 2025

A review of the carbohydrate insulin model of obesity and the related proposal that low carbohydrate diets are superior to traditional low fat diets. I begin with a discussion of the theory behind the carbohydrate insulin model, and then assess its plausibility using relevant evidence from rodent studies, dietary studies, ahd controlled feeding experiments. I argue that the scientific evidence does not support the alleged benefits of low carb diets for weight loss in otherwise healthy adults. Recommended pre-listening is Episode 151: Diet and Nutrition, and Episode 152: Obesity, Diabetes, and Hypertension. If you enjoyed the podcast please consider supporting the show by making a PayPal donation or becoming a Patreon supporter. https://www.patreon.com/jamesfodor https://www.paypal.me/ScienceofEverything

listening to The Science of Everything podcast, episode 153, low carb diets. I'm your host, James Fodor. So in this episode, we're going to conclude our three-episode arc looking at diet, nutrition, obesity and such. And in doing so, we're going to discuss the evidence and research on the difference between low-carb versus low-fat diets. obviously there are a wide range of diets that people try to use to lose weight or to maintain weight. But what we're going to do in this episode is focus on one of the major contentions in the literature, which is essentially relaying to the macronutrient balance of an ideal weight loss diet. That is whether to go with the standard recommendation of most health authorities around the world, which is to reduce fat and have the large proportion of calories come from carbohydrates,

or the alternative is to instead reduce carbohydrates in the diet and have the large proportion of nutrients come in the form of fats and proteins. So this is what I mean by a low carb diet. And the contrast, again, is usually the low fat diet. So in this episode, we're going to discuss the theoretical basis of why people have argued for the advantages of a low carb diet. So this is the carbohydrate insulin model of obesity. And we're also going to discuss.

talk about the ketogenic diet, which is a extreme form of low carb diet. We're going to then discuss the evidence for the carbohydrate insulin model and the advantages of low carb diet. So we're going to talk about the evidence for physiological mechanisms and rodent studies and so forth. Then we're going to move and look at dietary studies on humans. And we're also going to look at the specific claims relating to the alleged advantages of the low carb diets with respect to resting energy expenditure and also satiety. and we'll conclude by an analysis of the evidence whether or not it favours low fat versus low carb diets. So recommended pre-listening is episodes 151, diet and nutrition, and episode 152 on obesity, diabetes, and hypertension,

because we'll be building on the concepts that we discussed in those episodes. So without further ado, let's get started and begin by talking about the carbohydrate insulin model. So this is a proposal, the origins of which go back over a century, and it's been sort of developed in various forms. So I won't go through sort of the detailed history of it, but this is a sort of contemporary formulation of, again, a fairly old idea. The core claim of the carbohydrate insulin model is that diets that are high in carbohydrates and particularly simple sugar, so that is carbohydrates with a high glycemic index. We talked about glycemic index in the previous episode. diets that are high in these high glycemic index carbohydrates cause increased food consumption and reduced energy expenditure, thereby causing weight gain and ultimately contributing to obesity and

other issues such as diabetes and so forth. So the core claim here is that unlike the conventional wisdom, the mainstream view, which is that it's really just the balance between consumed nutrients versus expended energy, that energy balance model. Contrary to that, the carbohydrate insular model says that actually the macronutrient balance, so how much of your diet comes in the form of carbohydrates versus fats and proteins, that actually matters a lot. And that's very important for maintaining weight or for losing weight. And so the idea here is that when you consume a large meal that's high in carbohydrates, again specifically or more especially, high glycineic index carbohydrates, so relatively simple sugars or processed carbohydrates, what this is supposed to

the theory says that this increases your blood glucose, which then leads to a spike in insulin, because insulin responds to the increase in blood glucose, and the insulin tells yourself to take up the sugar that's in your bloodstream and to store or use it in your cells. So this results in uptake of glucose by skeletal muscles, the liver, and also adipose tissue. And furthermore, this spike in insulin results in a stimulation of lipogenesis. So this is essentially deposition of fat deposits in your adipose tissue because of the excess energy that's available, because of the high glucose in your bloodstream. So the idea is then that there's this initial spike in blood glucose and insulin,

and that's followed by a relatively rapid, like over a few hours, then crash or decline in blood glucose and also insulin levels. And the idea is that this is interpreted by the brain as what sometimes called cellular semi-starvation. So essentially a relatively low availability of nutrients because of the reduction in blood glucose and also reduction of ketone bodies and fatty acids. We'll talk a bit more about ketone bodies in a minute. But the idea is you have this initial spike with the rush of glucose that's available as the carbohydrates are processed. And then you have a relative crash where the nutrients are much less available. And your brain, so the theory goes

response to that by releasing regulating hormones that promote energy intake and also reduce activity to conserve energy and reduce the metabolic rate. So the idea here is that your brain responds to the changing or rapidly changing blood glucose and general energy availability levels by modulating energy intake and activity in order to conserve energy. So your brain thinks, there's this sort of cellular semi-starvation state. So I need to conserve energy and I need to also intake additional energy. Now, according to this theory, voluntary restriction of intake that doesn't pay attention to the carbohydrate load, but just looks at caloric intake. Again, and that's the standard recommendation is focus on reducing caloric intake. According to the

carbohydrate insulin model, a focus on reduction of food intake without changing the macronutrient balance will only lead to further reductions in energy availability, or at least will primarily lead to reductions in energy availability because of this glucose and insulin, spike and then crash, and will therefore drive further increases in hunger and reductions in energy expenditure lowered metabolic rates. The proponents of this theory argue that traditional diets don't work precisely because in high carb diets, you have this spike of energy availability in the few hours as the glucose is rapidly, simple carbs are processed and glucose is made available, and then you have this crash leading to cellular semi-starvation, which then promotes further

energy intake. So you become hungry, you eat more, which brings up energy availability, but then you're accumulating more calories than you need, and so you put on weight. And that's facilitated by the high insulin, which leads to lipogenesis, as we talked about depositing that extra energy in the adipose tissue. The overall result of this process is that there is a relative shift of body mass away from metabolically active tissues, like muscle protein, to less metabolically active tissues, like adipose tissue. And this reduces resting. and the expenditure. So your body essentially adjusts away from high metabolically active tissues, the skeletal muscles. It's moving mass to the less metabolically active, the out of post tissues,

storing that energy and reducing energy overall energy requirements to conserve energy. Because of this sort of sense of the brain that there's not enough energy available, because of this, essentially the crash after the initial influx of glucose, then the crash following that with a decline glucose and insulin levels. So the carbohydrate insulin model of obesity runs contrary to the conventional wisdom. The conventional wisdom is that excess caloric intake combined with insufficient physical activity is what drives obesity. But the carbohydrate insulin model actually sort of reverses that or substantively reverses that. It actually says that high carb diets, which lead to an increase in insulin sensitivity, that spike and then crash of blood glucose and insulin, which then promotes fat

storage, which in turn leads to a shift in body mass away from more metabolically active cells or tissues and sort of tricks the brain into thinking that there's a relative dearth, a relative lack of nutrients, which then leads to increased energy intake and a reduction in basal metabolic rate, also a reduction in activity. So it's the fat storage in a sense that leads to the increased energy intake. That's the sort of reverse. The standard idea, again is that too much energy intake leads to obesity, or too much energy intake leads to deposition of that energy in adipose tissue, which is essentially gain of fat, whereas the carbohydrate insulin molecule says it's actually storage of excess fat in adipose tissue,

which leads to increased energy intake and also reduced energy expenditure. The ultimate driver of all of this is the excessively high glycemic load of the diet. So too many carbs and too many simple carbs, highly processed carbs. So that's what the carbohydrate insulin model says. It's a model of the regulation of metabolic activity in the body, which fundamentally says that the standard model of obesity caused by too much energy intake is wrong because it's actually the glycemic load that's the core driving factor, the glycemic load leading to increased insulin sensitivity, increased deposition of lipids and adipose tissues, which then tricks the brain into thinking because of the glucose and insulin crash, that there's an insufficient energy availability, which then causes increased energy intake and reduced energy expenditure. So the energy intake increase and reduced expenditure, reduction in basal metabolic rate is actually a consequence of the gain in fat and the insulin sensitivity rather than the cause of the deposition fat and the insulin sensitivity.

That's why it's called a carbohydrate insulin model, because it's high glycemic load diets leading to insulin sensitivity and the consequences of that insulin sensitivity on a deposition of lipids in the adipose tissue, which is the core driver of obesity, rather than increased energy intake and reduced expenditure, which is what the standard model says. And so given this diagnosis or this theory about what causes obesity, is to reduce the glycemic load of the diet. So you can do this by reducing the proportion of carbohydrates and also reducing the glycemic index of the carbs that you do consume.

So basically moving from simple carbs to compost. complex carbs. So less processed carbs, whole grains, and so forth. But generally, low carb diets recommend a low overall proportion of carbohydrates in the diet and shifting those calories over to proteins and fats. And in fact, they also kind of say that total caloric intake is not as important as the macronutrient balance, so how much is in carbs versus proteins versus fats. So what I've discussed so far is all essentially theory. What we're going to do is then go through the claims that are made and see what the different lines of evidence have to say about that. Before we get to that, however, I do want to explain a little bit about something called the

keto diet or the ketogenic diet, because this has become popular in recent years, and it's relevant to our discussion here, so I want to give a bit of background on that as well. So we've mentioned ketone bodies in the previous episodes, but just to refresh your memory, a ketone is a type of functional group that's found in many organic molecules. It consists of a single, oxygen, double bonded to a central carbon atom, which is then bonded to other components as well. So ketone groups are very common in many organic molecules. Ketone bodies are a type of small organic molecule, or relatively small organic molecule, which has ketone groups on it, so that's why it's called a ketone body.

The ketone bodies are produced by the liver from fatty acids, basically as a way of transporting energy around the body. and in particular ketone bodies are useful for as an energy source for the brain when carbohydrate availability is relatively low. The brain prefers to get its energy from glucose, but when insufficient carbohydrates are available to produce glucose, the liver can break up, can extract energy, can oxidize fatty acids to extract their energy. However, fatty acids are quite large molecules, so the brain can't utilize those directly, that the liver can extract the energy. from the fatty acids for the brain, but it has to transport that energy to the brain in some form, and it uses ketone bodies. So there's smaller molecules they can pass through the blood-brain barrier, and they basically serve as a way for the liver to transport energy to the brain of a form that the brain

can use when there's insufficient glucose available or carbohydrates to produce the glucose. Remember that in humans, most fatty acids cannot be converted directly to glucose, and likewise, half of the amino acids can be converted back to glucose. And so if there's insufficient carbohydrates in the diet for gluconeogenesis to produce all the energy needed by the brain, then the liver can step in and oxidize fatty acids into ketone bodies, which it then sends to the brain. So the ketogenic diet is a high-fat, very low carbohydrate diet, which is designed to put the body in this state where it doesn't have enough glucose or carbohydrates to make glucose. And instead, the body has to meet its energy needs by metabolizing fat.

And specifically, the brain needs to, it doesn't have enough glucose to, for its, to meet its energy requirements. And so it needs to rely on these ketone bodies produced by the liver. So ketogenic refers to, the term ketogenic refers to the fact that the diets are deliberately designed to trigger ketosis by strictly limiting carbohydrate consumption. They're by depleting glucose and causing the body to enter a state of ketosis, which just means that these ketone bodies, are being produced and being utilized by the brain. The primary medical use of ketogenic diets is for treatment of epilepsy and children, where it's proven to be extremely effective at reducing the number and severity of seizures. So there are certain cases where ketogenic diets are recommended.

However, recently ketogenic diets have become a popular diet, not necessarily on medical advice, though there are some doctors who have advocated it, but it's alleged to be superior to traditional low-fat diets in, eliciting weight loss and maintaining a healthy weight. As I indicated before, a ketogenic diet is really just a more extreme version of traditional low-carb diets, like for example the Atkins diet, which is a more moderate form of a low-carb diet. So a standard diet might have 50 or 60% carbohydrates. A moderate low-carb diet might reduce that down to 20 or 30% or something like that, whereas a ketogenic diet generally reduces the proportion of energy coming from carbohydrate to 5% or below.

So like it's very, very low carbohydrate. And nearly all of the caloric intake comes in the form of fat in a ketogenic diet. The objective of a ketogenic diet is to maintain that state of ketosis. So your liver is metabolizing fatty acids and sending the ketone bodies to the brain. And that can be checked through urine tests. So there are studies that have compared ketogenic diets with otherwise comparable low carbohydrate but non-ketogenic diet. So there's not much carbohydrates, but just enough to keep your brain going so you're not in a state of ketosis. And generally these studies haven't really found much of a difference.

So the state of ketosis itself is not, doesn't appear to be that important. I'm not going to go into a lot of detail comparing ketogenic with non-ketogenic but still low-carb diets because there's not really much evidence that this state of ketosis makes a huge difference. and the main focus that I want to have in this episode is on the carbohydrate insulin model and low carb versus low fat diet. But because ketogenic diets have become more popular lately, I thought it was important to discuss them. One thing that I want to say about the ketogenic diet is that in the popular discourse, it's often claimed that the ketogenic diet turns the body into a fat-burning machine,

which helps promote rapid weight loss. And I think that this is an instance of science communication being sort of technically correct but very misleading because the whole point of the state of ketosis is that your liver is indeed oxidizing lipid. So it is burning fat in that sense. However, the ketogenic diet is designed so that your fat intake is much, much higher than it would be under a normal low-fat, higher carb diet. So the fact that your body has become a sort of fat-burning machine, so to speak, doesn't actually tell you anything about how effective you're going to be at losing weight because you're also consuming more fat.

And really what's just happened is that the macronutrient balance has shifted, and now your body needs to burn all that fat because that's where the energy is. Your body is going to deposit excess energy in the form of adipose tissue, regardless of whether that energy initially came in the form of carbohydrates, fat, or protein. And so the mere fact that your body is burning fat and your liver is shipping the ketone bodies to the brain in a ketogenic diet doesn't tell you by itself whether it's better than other diets for losing weight. So this is a very misleading statement that's often used in marketing for ketogenic diets. So that's what I wanted to say about ketogenic diets. Now we're going to go back and talk about

general low-carb diets, but this also applies to ketogenic diets as well. So let's talk about the evidence for the carbohydrate insulin model and what different lines of evidence have to say. So first we're going to focus on studies that have been done in rodents and looking at the underlying metabolic processes and see what plausibility there is to the claim that it's macronutrient balance and specifically high glycemic loads in the diet, which result in increased energy intake, reduced expenditure, and thereby cause obesity. Some of the claims made by the carbohydrate insulin model are just sort of standard that are not disputed. So it is true that high glycemic foods, particularly simple sugars, are metabolized

very quickly and therefore result in a very rapid rise in blood glucose. I discussed that in previous episodes. So that's not disputed. The rapid rise in blood glucose also results in a rapid spike in insulin levels because insulin responds to blood glucose levels. So that's not disputed either. Likewise, when you have lower glycemic load, or particularly when you have fats or proteins, blood glucose does go up after eating, but much, much more slowly because those things take longer to metabolize and there's in the case of fats that they're not converted directly to blood to glucose anyway so they're converted to they're either used directly or they convert to the ketone body so they don't actually contribute to blood glucose going up and so in low carb diets

and low glycemic load diets blood glucose goes up much more slowly it goes up less and the insulin concentrations of the blood also go up less conversely they also decline more slowly as well so with high glycemic load diets, you have a sharp spike and then a rapid decline in blood glucose and also insulin levels. Whereas in low glycemic load diets, you have a slower spike and a much lower level, so it doesn't go up as much, and it also stays high for longer. So this is kind of what gives part of the justification for the idea that the nutrient availability is kept higher because the blood glucose level and also other nutrients like the ketone bodies and the fatty acids. they take longer to metabolize, and so they're hanging around in the blood for longer.

The idea is that when you have a low-glycemic load diet, the blood nutrient levels are kept at a more consistent level, but also at an average higher level, because with the high-glycemic load diet, they spike up and go down again, whereas with the low-glycemic load, they a lot more slowly and stay higher for longer. So there's a sort of maintenance of a more consistent nutrient availability in the bloodstream.

So all of that is, not disputed and is sort of well-established. Another thing that's well-established is that low blood sugar levels can lead to hunger. In particular, there is a hunger-inducing hormone called ghrelin, which is produced by the digestive system, and that responds, at least in part, to blood sugar level, though it's also a bit complicated. It is true that declines in blood glucose levels, predict hunger, and predict behavioral phenomena like meal initiation or requests for meal, and then predict energy intake. So all of that is true. Where it becomes less well accepted and more speculative is the claim that these changes to blood glucose and blood insulin levels drive increased energy intake and drive reduced energy expenditure, reduction in basal metabolic rate and reduction in spontaneous activity. So that's supposed to be the core driving force by which high glycemic load leads to obesity. It's because it changes energy intake and it changes energy expenditure. which then leads to increased caloric intake, reduction in

chloric expenditure, and therefore obesity, deposition of the excess energy in there. That's the part that's speculative, and so what's the evidence for this part of it? One piece of evidence that is highlighted by proponents of the carbohydrate insulin model is that delivery of insulin to mice and also delivery of insulin to patients with diabetes leads to increased food intake and hence obesity. And the idea here is that, again, according to the carbohydrate insulin model, if you administer insulin, that reduces the concentration of metabolic fuels in the bloodstream because it causes tissues to take up glucose and also encourages adipose cells to deposit lipids rather than to metabolize them. So it reduces the fatty acid concentration and ketone body concentration in the bloodstream as well. So insulin leads to essentially storage of energy rather than usage of energy.

and this leads to hunger because there's lower energy availability, the cellular semi-starvation, hence obesity. So there's some evidence that that can happen, say, when you deliver insulin to mice and some people with diabetes. However, the key link here, which is the idea that insulin leads to increased hunger and reduced satiety, is pretty inconclusive. So one study which looked at the concentrations of insulin in the blood

and compared that to these two hormones, talked about before. So one of them is grelin, which is produced by the gastrointestinal system and leads to increased appetite. And another is leptin, which is a hormone that's produced by adipose tissue and leads to satiety, so a sense of fullness. So this study looked at the relationship between these three molecules and it found that, quote, the concept of insulin-mediated leptin stimulation and grelin suppression seems unclear, and we could not identify a clear relationship among insulin, leptin and grelin, end quote. And overall, what this and other studies have found is that satiety, the sense of fullness,

is a very complex phenomenon and it has many cognitive and social components. There doesn't seem to be any strong evidence that it's much affected by the macronutrient balance. A different study which compared specifically low carb and low fat diets found no physical differences of hunger, fatigue, or stomach symptoms between the groups, and only a very small, so 0.3 points on a 3 point scale, it's like a 10% difference essentially, a small difference in subjective hunger with slightly lower hunger in the low carb group. So this study was specifically looking at the difference between low carb versus low fat diets. And so it essentially found no or only very small differences.

So overall here, what we see is that it's true that insulin leads to reduction in nutrient availability in the bloodstream, but the evidence for that leading to an increase in hunger or reduced feelings of satiety is just not really very strong. It's not clear that it's related to subjective hunger very much. I mean, in that study we talked about, there was a 10% point difference, but then, like, other measures of physical effects of hunger or stomach symptoms and feelings of fatigue weren't significantly different. And then the study that I mentioned that looked at actually hormone levels that relate to satiety, so ghrelin and leptin also didn't really find any significant difference there.

So the evidence just isn't strong for a particular association between insulin and satiety. here. Now, another avenue of evidence that proponents of the low carb diets point to is mice studies, or rodent studies more generally, which they argue, or some studies, they argue, show that high glycemic diets lead to increased weight gain even when controlling caloric intake. So I'll quote from one of these studies, quote, when energy intake was restricted to prevent weight gain, so this is in mice, the high glycemic index group still developed abnormal body composition. Despite consuming fewer calories, these animals had much. more body fat at the expense of lean body tissues. End quote. So there are a number of studies that

have found this sort of effects that if you feed mice a very high glycemic index diet, that they do put on more weight even when you control for caloric intake. However, looking at a systematic review of these rodent studies, which analyzed I think a few dozen of them, they actually found that it's true that there is an effect on body mass as well as insulin levels of the glycemic load of the diet. Now, that is what would be predicted by the carbohydrate insulin model that having a high glycemic load diet promotes fat deposition, it promotes energy intake. And even if you restrict that intake, it still results in a reduction in the basal metabolic rate because of the shift from high to low metabolically active tissues and therefore is going to lead to weight gain. However,

confounding factor of these studies is that the difference between the high glycemic index and the low glycemic index diets was not simply macronutrient balance, but there were other differences as well. And one of the main ones was dietary fiber. So generally, the high glycemic index diets consisted largely of essentially sugar with very little dietary fiber. And it's known that low fiber diets lead to a range of negative health outcomes and metabolic effects. And can also have other confounds like changes in gut bacteria, which may influence the efficiency of nutrient extraction and other things. So it can have a wide range of complex effects on the body. And because these confounds were not sufficiently controlled by these rodent studies, it's really unclear whether

it's the macronutrient balance itself that's leading to this, or whether it's just a nutriently deficient diet that lacks dietary fiber and potentially other micronutrients as well. And another important thing to consider is that these studies that the proponents of low carb diets point to, these experiments involve testing the difference between high glycemic index, so basically like a lot of sugar, and low glycemic index, so more whole grains and things, diets in rodents. They're not actually studies that compare low-carb versus low-fat diets in rodents. So it's not clear whether this actually favors the low-carb diets. In fact, it's not really very strong evidence for much of anything

because of all of these confounds of the fiber, gut bacteria, and potentially other micronutrients are because you're not actually comparing the low-carb versus the low-fat diets. So that's the evidence for the physiological mechanism. overall we can summarize this as saying that once it's true that high glycemic index and carburge diets do lead to a spike in blood glucose and insulin and then rapid decline the evidence for this having an appreciable effect on hunger and satiety is not strong it doesn't seem like that there's much of a consistent relationship between insulin and relevant hormones like leptin and ghrelin it's not clear that it's much empirical evidence for physiological or psychological differences in

hunger between low carb versus low fat diets are only very small differences. High glycemic versus low glycemic diets in rodents don't control for other important nutrient differences such as fiber intake. And so it's hard to interpret the relevance of those studies for the question as to the benefit of low carb versus low fat diets. Okay, so that is the evidence relating to the physiological mechanisms. Now we're going to talk about dietary studies. So before I go through the evidence, let me talk about how these studies are typically done. So many studies have been conducted which compare the efficacy of low carb versus traditional low fat diets for generating weight loss, generally in obese patients. So what they do is they recruit obese people

from an otherwise healthy population. So usually these people don't have diabetes or other health effect. Sometimes they may, but in general they try to recruit healthy obese people and randomly assign them to one of these two diets. Now, because we're talking about outpatient studies, so people who just live their life normally but are participating in the trial, we'll get to inpatient studies a little later, the actual intervention essentially consists of just telling people to consume the low carb or the low fat diet.

And generally, there's some amount of instruction that's given as to like recipes, some amount of follow-up that's given to see whether they're maintaining the diet. And usually there'll be periodic doctor's visits to, check their weight and also check other sort of vital signs. And the exact intensity of those varies across studies. These studies typically last between six and 18 months, I think the very longest are maybe two years. Subjects are not confined. They go about their lives as normal. And critically, caloric intake and dietary adherence is not directly observed by the researchers. So there's various ways that they try to assess that, how well people are adhering to these diets.

They can't do this directly, right? Because they're not monitoring. people, you know, 24-7. Some of these studies, generally the shorter ones, do actually provide meals for the subjects to eat. Usually that's not done. Usually you just tell them what to eat, but sometimes meals have been provided. But again, there's no way to really know whether they're subjects are eating those meals or eating something else in addition to those. At the outset, there's a major limitation of these type of dietary studies, which is that you're relying on the self-reports of participants as to what they've actually eaten. And there's two main methods that are used to assess this because what you want to do when you're doing your studies is you assess,

well, have they adhered to the diet, right? You've assigned them to the low carb or low fat group, but you have to assess whether they're actually adhering to that. There's two main methods to do that. One is a food frequency questionnaire, which asks subjects to report the frequency with which they consume different foods over long periods of time. So over months or years, you'll ask like, how often do you eat this, how often do you eat that? The second type of method is called a 24-hour recall, which asks subjects to report all the food that they've eaten the previous day. The 24-hour recall method has the advantage that it doesn't have to rely on memory nearly as much. It just relies you to talk about or to report what you've eaten the previous day.

The downside to that is that it's noisier because you might have had an unusual day. You don't necessarily know how often you eat this sort of thing. And it may not be as accurately indicate how often you eat unusual food. So, for example, if you eat certain foods only occasionally, that's not necessarily going to be picked up on the 24-hour recall method, but that might be important for the overall, that might be important for assessing the overall macronutrient balance and composition of your diet, not just what you wait the previous day, but what you eat in a week and a month. Now, one review of these methods explains, quote, all self-report methods are challenging. People do not commonly attend to the foods that they've eaten, do not remember everything, do not know the contents of the foods eaten, and cannot estimate portion-s-sides. accurately, end quote. Another review of validation studies, which attempt to compare people's self-reported intake with direct physiological measurements. So there's ways to, that directly measure very accurately how much weight people have gained, not just based on weighing them, because that's

not necessarily as accurate either, because there's changes in water weight and other things, but there's ways to very precisely measure people's metabolic rate and changes in fat levels and things like that, which I won't get into. But so some of these, validation studies have been done that try to assess how accurately people report their caloric intake. And they found that on average, calorie intakes are underreported by 28% in the frequency questionnaires. So that's the long time period ones. And about 15% with the 24-hour recall method. So there is a better accuracy, at least with overall caloric intake in the 24-hour recall. But both methods underreported caloric intake by a significant amount, 15 to 30%, is really,

really a huge underreporting. Because if you're reporting that you're consuming your recommended dietary intake of calories, when actually you're consuming 15 to 30% more, that means you're going to actually be putting on a lot of weight, even when you sort of think or report that you're actually consuming the right number of calories. And even worse than this, an average overall under-reporting of caloric intake is a problem. But if it was consistent, you could just adjust for that, right? But the problem is that only around 10% of the variance in actual caloric intake was explained by reported caloric intake, which means that individual subject-level self-reports are almost completely useless at tracking individual difference in the true energy intake.

So what someone tells you will bear some resemblance to what they actually ate, but when you're looking at differences across people, people telling you that they ate more versus less tells you almost nothing about how much they actually ate relative to other people, at least in overall caloric intake. The composition of the diet may be a different question, but in terms of overall caloric intake, which is obviously important for weight loss studies, self-reports are very unreliable. Even worse, there's actually evidence that not all types of foods are reported equally accurately, or equally inaccurately, depending on how you think about it. In particular, there's several studies that have found that there's a tendency to selectively under-report fat intake, whilst people

tend to be a bit more accurate at reporting their intake of protein. One study, for instance, that explicitly compared weight loss in low-carb versus low-fat diets, found that despite the fact that participants reported roughly equal caloric intake in the two diets, there was about a five kilogram greater weight loss in the low carb diet compared to the low fat diet. The authors attributed that to greater underreporting of caloric intake for the low fat diet compared to the low carb diet. They speculate, quote, because the low carb group followed a diet that differed dramatically from their usual intake with more limited food choices that were easier to catalog and record, we think it is probable that their reporting was more accurate, end quote. So the important

here for us when we're considering low-carb versus low-fat dietary studies is that low-carb diets are generally much more different from standard diets than our low-fat diets and also much more restrictive because most of the foods that people eat, especially processed foods in a modern Western context, have a lot of carbohydrates in them. So if you're strictly cutting down carbohydrate intake, it's quite, at least for most people, that represents quite a change to the diet and quite a restrictive diet in terms of the range of foods that you can eat. And so there's evidence that this means that people on a low-carb diet are slightly more accurate in remembering and reporting the food that they've eaten. They're attending more to it. They have to be more selective

about what they eat. And there's fewer different things that they eat, which maybe just helps reporting more accurately. So if there are small differences, even relatively subtle differences in accuracy of reported and self-assessed caloric intake, that could lead to differences in weight loss outcomes between low-carb versus low-fat studies, which actually don't have anything to do with the diet composition itself. It basically just amounts to people finding it easier to track caloric intake on low-carb versus low-fat diets. Now, we'll come back to this question at the end, because that in itself is interesting. However, note that the carbohydrate insulin model, which motivates low-carb diets, doesn't itself saying anything about how easy it is to maintain

a restrictive diet or how easy it is to track your food intake. That's not supposed to be the mechanism of low-carb diets. The mechanism is supposed to be this sort of cellular semisarvation, which triggers deposition of fat and reduces circulating fuel in the bloodstream, which leads to hunger, which increases energy intake and reduced its energy expenditure. All that stuff we talked about. It's supposed to be driven by insulin and the glycemic index loading of the diet. It's not supposed to be driven by how easy it is to maintain a certain diet. In fact, many proponents of the carbohydrate insulin model have said that you shouldn't actually need to try to restrict your caloric intake. All you need to do, according to these proponents, is shift away from high

glycemic index foods. When you do that, you'll naturally be less hungry and you'll naturally have a higher metabolic rate. You'll naturally lose weight. You won't actually need to, at least for most people, you won't need to sort of consciously try to reduce your caloric intake. The weight loss will sort of happen by itself, so to speak. So I'm emphasizing this because this potential mechanism of a difference in diets of making it easier to track and restrict your dietary intake is not supposed to be the mechanism. It's not the alleged mechanism behind the efficacy of low carb diets. And so it's really a distinct issue. So all of those caveats being made, let's have a look at the results of six different meta-analyses of outpatient studies of low-carb versus low-fat diets.

If you're not familiar, a meta-analysis is essentially a re-analysis of a bunch of different individual studies, which tries to make their results, It sort of analyzes them in a consistent format. Usually what they do is that they standardize the results from all of the studies and compare them on a key outcome. Often it's fat loss or changes in body weight. And then they aggregate them together by weighting the studies in accordance with their standard deviations. It's a weighted average of the effect size of each individual study, weighted by the inverse of the standard deviation of each of the trials.

So don't worry if you don't know exactly what that means. I'm just sort of explaining how they aggregate them together, but for our purposes, you can just understand that a meta-analysis is basically a re-analysis and averaging of multiple trials. Most of these meta-analysis have about a dozen to maybe 20 trials in them, some a few less, and it's important to bear in mind that many of the individual studies within the meteringalysis, like they're considered by each meta-analysis, are the same, but not completely. So like there's overlap. So these meta-analysis results are not independent of each other, but they are a bit different from each other,

because the inclusion criteria are a bit different, some are a bit older so they have, so they don't have some of the newer studies, and there's always some slight differences in how the methodology is applied. So we're going to look at six, right? So, Mansour at Tal, effects of low carb diets versus low fat diets on body weight and cardiovascular risk factors, a meta-analysis of randomized controlled trials. This study found a two-kilogram difference in body weight, favoring low-carb diets. Bueno etal, very low-carbohydrate, ketogenic diet versus low fat diet for long-term weight loss and methamalysis for randomized controlled trials. This study found 0.9 difference in weight, so in body weight,

and that again favored the very low-carb ketogenic diet. Huetal, effects of low-carb diets versus low-fat diets on metabolic risk factors and meta-analysis of randomized control trials. This study found a 1-kilogram difference in changes in body weight, favoring the low-carb diet. Hesha Natal, systematic review of randomized controlled trials of low-carb versus low-fat-slash-low-calorie diets in the management of obesity and its comorbidities. This study found also a one-kilogram difference in weight loss over the course of the trial, favoring low-carb diets.

Bernstein-Natalysis, Dietary intervention for overweight and obese adults comparison of low-carb and low-fat diets and meta-analysis. This study found a 1.9 kilogram difference between the two diets, again, favor the low-carb diet. Chorla et al, the effects of low-fat and low-carb diets on weight loss and lipid levels. So this one was a little bit different because it looked at the, or the differences in weight loss between the two conditions, at different periods of time. So it follow-up time over the course of the study. And this is one limitation of the other manner analysis, because the studies actually have different periods of time. So you shouldn't necessarily compare weight

loss after three versus six versus 12 or 18 months, because it could be different. So this study found that there was greater weight loss in the low carb condition of about 1 to 1.5 kilograms over up to 3, 6 and 12 months. But they actually found that in follow-ups over 12 months, so a year or more, that there was actually a 0.8 kilogram greater loss in the low fat, in the low fat condition, although this was not statistically significant. So overall, they found a one kilogram greater weight loss in the low carb condition, but it was different depending on the time span you looked at. And in the longest time span, it actually favoured the low fat, though not statistically significant difference. Younger Tal, the effects of low-fat high-carbohydrate diets versus low-carbohydrate high-fat diets on weight, blood pressure, serum liquids and blood glucose, a systematic review and meta-analysis. So this one found a one-kilogram difference favoring the low-carb diets. And finally, Tobiasetal effects of low-fat and other diet interventions on long-term weight change in adults,

a systematic review of metering. This study found a 1.1 kilogram difference in body weight favoring the low-fat, favoring the low-carb diet. So I think I promised six meta-analysis there, but I guess we got eight in the end. But anyway, we see that actually the meta-analysis, despite having slightly different inclusion criteria and being done at different times have pretty consistent results. There's a pretty consistent finding of about a 1 kilogram greater weight loss in the low carb condition compared to the low fat diets. However, the one meta-analysis that did look at periods of time found that this difference

went away when considering follow-ups above 12 months. If we ignore that and just consider the 1-kilogram difference at face value, we should ask how significant is that? Now, there's a few factors to consider here. One is that if we use that BMI to estimate how overweight the average subject is, we get a figure of around 25 kilograms overweight for the average subject. So ideally, the average subject would be wanting to lose about 25 kilograms, a bit less for women, a bit more for men on these studies. Usually, in many of these studies, obviously it varies, the total weight loss after six to 12 months is usually about 6 to 7 kilograms. So most participants don't come close to achieving their weight loss objective, at least if you measure that by healthy BMI.

So comparing a one kilogram difference over six months to a year for people who are 25 kilograms overweight, this is not a very large difference. It's a small fraction of the total weight that ideally they would be wanting to lose. Especially when we consider, as I noted, that the difference appears to be largely attributable to the earlier periods of time in the studies and that when you look at later follow-up periods, this difference actually goes away. And the second point, the second reason why this small difference is pretty suspect is, as I mentioned earlier, these studies all rely on self-reports of adherence to the diet. And there's evidence to indicate that people on low-carb diets may be slightly more accurate in reporting their caloric intake, and conversely people on the low-fat diets may be slightly

slightly more under reporting their caloric intake, which essentially means that this one kilogram difference could easily be explained by difference in caloric intake of the two groups. Now, to an extent, this is potentially consistent with what the carbohydrate insulin model says, right? Because they say, look, if you reduce your glycemic load, then that will result in lowered energy intake. But we can't actually tell whether that's what's happening here.

All we know is that there's a very small, given the time span and how overweight these people are, a very small difference in weight loss between these two diets, we don't actually know if that's driven by the mechanism that the carbohydrate insulin model proposes, or whether it's just driven by the fact that people on low-carb diets are very slightly, find it slightly easier to adhere to their diets and restrict their caloric intake a little bit more than the people on the low-fat diets. From these studies, it's very hard to tell. Another point that I would make is that if you look at the individual studies that are reported in these meteringalysis.

Often the effect is the small difference between the studies is driven by one or two studies that report quite a large effect. And most of the studies don't report much of any effect. So, for example, in the Mansour et al, the first matter analysis that I mentioned, one of the studies reports an almost six kilogram loss in favor of the low carb diet,

and another reports a four kilogram difference, whereas most of the other studies don't report any significant difference at all. And this is the same with Huatal, which is a slightly more recent mineral analysis. Again, there are only two studies of the 20 or so that they examined, which show a statistically significant benefit for the low-carb diets. But those two report quite a large benefit, and so that drives much of the effect. So what we have overall is a situation where the overall effect size is very small

compared to how much weight that these people would ideally be losing. The difference is very small. It's readily explained by known differences in the reporting accuracy and memory recall accuracy of intake on low-carb versus low-fat diets. And the effect is largely driven by about two studies or maybe three studies which show very large effects favoring low-carb diets, whereas most of the individual studies don't show those effects. And finally, The evidence also indicates that when you look at longer-term follow-ups, the effect goes away. So my overall personal evaluation of this evidence is that it is not conclusive either way. I would say that this is evidence is consistent with there being no significant difference between low-carb versus low-fat diets.

So nominally, it does show a small benefit to low-carb diets, but for all the reasons I've given, I personally don't think that that is a sort of genuine benefit, or there's not sufficient evidence that the carbohydrate-insulin model is sort of correct. So so far what we've got from the physiological studies, the rodent studies and the dietary studies is not much conclusive evidence either way. Possibly a small benefit from some rodent studies and from the dietary studies to low carb diets, but it's pretty small, equivocal, and there are alternative explanations that don't rely on the carbohydrate insulin model, which may better explain the results. What I think is the most interesting evidence are the inpatient studies. So these are very different to the dietary studies. Instead of just telling people to adopt one or the other diet and then following up periodically and asking what they've eaten and weighing them and so forth,

these studies actually involve taking a group of people and then putting them in a hospital environment where they're monitored full-time, usually for two weeks. And this allows for a wide range of measurements to be taken about blood glucose, insulin levels and so forth, as well as changes in weight. it allows for full observation of everything that they're eating. So we don't have to rely on self-reports. And this is a crucial difference is that we're able to control what they're eating or at least know what they're eating. So some studies will strictly control the caloric intake.

Other studies, depending on what the question is, will allow participants to eat whatever they like, but we'll measure exactly what they do eat. So one hypothesis of the carbohydrate insulin model or prediction of the carbide insulin model, is that low glycemic load diets, or low carb diets, would help weight loss by increasing the basal metabolic rate relative to the high glycemic in next diets, which reduce the basal metabolic rate. And again, the mechanism here is supposed to be that the insulin sensitivity resulting from the spikes of glucose and insulin in the bloodstream leads to fat storage, which leads to a shift of body mass away from metabolically accuracy. to metabolically less active tissue, and also results in a decline in circulating fuels because you're

depositing all of that energy in the out of the adipose tissue, and it's not available in the bloodstream, so that's interpreted by the brain as a semi-starvation state, so it's trying to conserve energy, it reduces the metabolic rate, the base metabolic rate, and also reduces spontaneous energy expenditures, like from activity and so forth. So this is the hypothesis, and that's testable, right? So this can be tested by inpatient studies. So one metan analysis looked at this. It considered 28 controlled feeding study. So controlled feeding study is when you supply the food to people in an inpatient setting and measure their weight loss and you know a variety of other factors as well. And so this study looked at differences between low fat versus low carb diets, controlled feeding.

So they know exactly what people are eating. The main downside from this is that the treatment is very expensive of course. So you have smaller sample sizes and shorter timeframes. Also the environment is not very realistic. You have people in a a hospital rather than in their actual daily life. And so it's less clear whether the results can be generalized to a real world environment when you're not controlling exactly what people eat. However, it's still very useful for investigating fundamental metabolic mechanisms and processes, because, again, the carbohydrate incident model predicts that the resting energy expenditure should be higher in the low carb diets than in the low fat diets because of this promotion of metabolically active,

as inactive tissues. And because again, in the high-carb load diet, the idea is that the body enters this sort of semi-starvation tissue starvation state where it has low circulating fuel level, so it thinks it needs to conserve energy. So what this study found is that across the 28 studies, overall, there was a 26 kilocalorie per day, which is about 100 kilojoules per day, difference in favor of the low-fat diets. However, this difference was so small. that it was judged to be clinically insignificant. So remember that the sort of average caloric intake for men is about 10,000 kilojoules per day. And for women, it's like 7,500 or 8,000 or something like that.

So when you're looking at 100 kilojoules, that's on the order of 1%, which is an incredibly small difference. And so almost certainly not clinically significant. And if you look at also the individual studies in the manner analysis, So there are two studies that show sort of moderately high effect sizes in favor of the low-factite, and there's two that show sort of moderately high effects sizes in favor of the low-carb diet, and most show no real difference. So here the effect is not driven by any one or two studies like plausibly it was in some of the other manner analyses.

So this, I think, is fairly strong evidence that there is no clinically significant differences in resting energy expenditure, as a result of the short-term alteration, so usually about two weeks, in macronutrient dietary balance. More recently, there has been a single study this time that was conducted with very rigorous controls, which actually conducted a within participant study design.

So they alternated each participant in an inpatient setting between low carb and low fat diet for two consecutive weeks. So two weeks on the low carb, two weeks on the low fat, randomized the order of those. So this is an extremely rigorous study because they compared the differences within the same participants. So that eliminates a large amount of noise due to individual differences

if you assign different people to the two groups. This assigned the same people to both groups and looked at the difference in response on one versus yellow. So this is probably the best study that's addressed this question about resting energy expenditure and also energy intake. So it looked at both of those because this was not a controlled feeding study. This was a study that allowed participants to eat as much as they liked, but the foods that they could hit were restricted, according to the low carb or low fat diet,

and the experiment monitored how much weight they lost, as well as how hungry they felt. It measured blood glucose levels, insulin levels on a regular basis. It measured food intake and caloric intake, and it measured how their subjective sense of hunger and fullness and things like that. So very rigorous study, and as I said, I think by far the best inpatient study that I've seen. And for those of you who are interested, this is Hull etal, effective a plant-based low-fat diet versus an animal-based ketogenic diet on ablabitum energy intake published in Nature Medicine in 2021. So let's talk a bit about what this study found. So first of all, they found that

each of the participants had a lower caloric intake on the low fat compared to the low-calib diet and quite a significant difference actually about 700 kilojoules per day difference. This differs from what the metta analysis found, which is no significant difference. although again this study I tend to have higher confidence in because it was a within participant study and they show a graph of this but the effect is quite profound it's not just that the average was different between the low carb versus the low fat diets it's actually every single individual participant had a lower caloric intake on the low fat versus the low carb diet and the difference was quite a lot about 700 kilojoules per day and again compared to the average intake recommended intake for men of about 10,000 that's a noticeable difference also they found no

significant difference in self-rated, so subjective hunger, satisfaction or fullness resulting from the meals. There was also no rated difference in the pleasantness or familiarity of the foods that they were eating. So there's no observed or measured difference in how much participants enjoyed the food or felt satisfied by it, but the caloric intake was significantly higher in the low carb compared to the low-fat condition. Now when we turn to changes in body weight, there was an interesting dichotomy here. So despite consuming more calories, in the low carb condition, participants actually lost more weight. So they lost about two kilograms, a little less than two kilograms on average over the two weeks, compared to about one and a half on the low fat condition.

So about half a kilogram difference, greater loss on the low carb condition. But, and this is one advantage of this inpatient study, which is able to take very accurate measurements, When you actually look at the type of weight that they're losing, on the low fat diet, there was no change in the fat-free mass. So this is, you can divide the body mass into fat and fat-free mass. So fat-free mass is everything that's not fat. On the low-carb diet, in the first week, about one and a half kilograms of fat-free mass was lost. And not much change in the second week. If you then look at the actual fat mass, so on the low-carb diet, there was effectively no change in the fat mass.

whereas on the low-fat diet, there was a steady reduction in the low-fat mass. So a bit less than half a kilogram per week, or maybe 0.3 kilograms per week. So the data only goes for two weeks, so it's hard to be very clear about it. But what appears to have happened is that when participants were on the low-carb diet, they lost a lot of mass in the first week, relatively speaking. But essentially none of that was actually fat mass. potentially it was water or maybe changes in protein composition in the body or something else, right? And I presume that that is an adaption of the body to a substantial change in the macronutrient composition of the diet.

There's metabolic changes resulting in changes in the fluid retention and maybe protein mass in certain parts of the body, in the liver or in the skeletal muscles or wherever else. I don't know exactly. But there was no actual fat loss or essentially no fat loss in the low carb diet. Whereas the low fat diet, because they were consuming fewer calories, they actually lost, lost fat mass at each successive week. When it comes to changes in blood glucose and blood insulin, we see exactly what's expected. So with the low carb diet, we see a gradual rise, both glucose and insulin levels, up to about two hours after consuming the food, and then a gradual decline in the hours

following that. Whereas the low fat diet, which had relatively high carb loading, there was a rapid spike in glucose levels within about 30 minutes after meals, and a more rapid diet. decline and likewise there was a rapid spike in insulin and then a relatively rapid decline. Although in both cases actually insulin levels, even up to six hours after meal, the glucose levels and insulin levels still stayed higher in the low fat compared to the low carb diet. Now this is not consistent with the prediction of the carbohydrate insulin model. The carbohydrate insulin model predicts that when you have a low carb diet, the circulating

nutrient levels should stay at a relatively high level hours after food extinction, whereas on the high calm diet, the high glycemic index diet, the nutrient level should decline quite rapidly below the level of the low carb diet, which is what triggers the cellular semi-starvation and the increase in hunger, the body trying to conserve its energy and things like that. But this is not what was observed in this study, that there was a spike of insulin and blood glucose, but it didn't decline. more than to a level below what the low-carb diet did. So this is pretty strong evidence, at least in this short-term study,

against what the carbohydrate insulin model is saying, as well as the fact that the people on the low-carb diet did not consume less energy, which is the prediction that they would feel less hungry, more full, that their metabolically active tissues would, that there would be a relative increase in the body composition towards metabolically active tissues, their metabolic rate would be higher and said that they would lose weight. But there was no observation of this in this study. So this single study, I think, provides the strongest evidence against the carbohydrate insulin model. The main drawback is that it is only two weeks in duration, and that is a

point that's being raised by proponents of low carb diets, we'll get to in a moment. So let's conclude and put together all the pieces that we've talked about today, about the relative benefits of low carb versus low-fat diets. So first of all, let's talk about the underlying metabolic mechanism. The mechanism, according to the carbohydrate insulin model, is supposed to be that increased insulin following a high glycemic index diet. Meals leads to deposition of fat and an antipose tissue, which then leads to a shift away from metabolic active towards metabolically, less active tissues.

The crash in glucose and insulin levels following the initial spike is then supposed to lead to a reduction in nutrient availability, leading to cellular semi-starvation, leading to a reduction in basal metabolic rate, and increased hunger and increased food intake. And those are supposed to then lead to the obesity. The problem with this is that the mechanism by which all of this is supposed to happen, so specifically how insulin is supposed to increase hunger and slow metabolic rate is not specified. And the evidence that we have is that insulin doesn't do this.

Insulin does not have a clear relationship, as we talked about, between ghrelin or leptin, and does not have any clear relationship in general to feelings of hunger. Low blood glucose does, but insulin does not seem to have this clear relationship. And hunger indeed is, and satiety is mediated by so many factors that it's, I don't think, very plausible based on the evidence that we have, that insulin plays this key role. So the mechanism is not really very clear, and based on what we do understand about the metabolic processes, I don't find the carbohydrate insulin models will be very plausible. When we turn to rodent studies, these are somewhat mixed. So free feeding studies that involve high fat diets have been

in rodents have been shown to cause adiposity, whereas high sugar diets don't. Defenders of the carbohydrate insulin model respond to that by saying that, well, it's the excessive saturated fat intake which causes the adiposity in these rodents, whereas if you substituted it with unsaturated fat, it would not have this effect. I don't find this a very convincing argument, because the carbohydrate insulin model predicts that low glycemic index carbohydrates should be the main driver of obesity, not saturated fats. So it's unclear then why the difference in saturated versus unsaturated fats is supposed to be so important in the case of high fat diets in rodents. There was the other type of feeding of rodent feeding diets we talked about which compared

not high fat versus high sugar but rather compared high glycemic index versus low glycemic index and those studies found or at least some of those studies have found that the high glycemic index rodents even controlling for caloric intake had great adiposity and worse outcomes. But as we saw that it's not clear what that's actually driven by because there's there is not sufficient controls for fiber intake gut bacteria and other things like that. So overall, I think that the rodent studies are equivocal and not really a strong evidence in favor of one study or the other, though I do find that the free-feeding high-fat diets leading to adiposity is not really what would be expected under the carbohydrate insulin model and the response to that I don't find very convincing.

Something I haven't talked about but is mentioned in some of the reviews on this issue is genetic evidence. So I'll quote from one study, quote, every known monogenetic disorder that causes human obesity involves a gene intimately involved in the control of energy intake by the herphthalmous, end quote. So this is contrary to the idea that it's actually adipose tissues that regulate and their response to insulin levels that are the main driver of a food intake. This is what the carbohydrate insulin model predicts, which is that it's really the adipose tissues themselves that are driving increased intake in response to insulin. Whereas, as I quoted here, the evidence that we have at least from disorders caused by single genes, that all of the cases that of these that lead to obesity involve genes that control for energy intake by the hypothalamus. So it seems to be driven by the brain, not by the adipost tissues. This is more consistent with the standard view, which is that it's excess caloric intake,

which drives obesity and not the carbohydrate insulin mechanisms that we talked about. However, there's always the caveat that, look, this is monogenetic disorders, and maybe the genetic disorders which operate under carbohydrate insulin mechanisms are just not monogenetic, and so they don't appear. So that's possible. Seems a little ad hoc to me, but you can't rule that out. We talked about long-term dietary studies, so those are the outpatient ones. And what we found is that there was a small about one kilogram difference over periods of less than 12 months.

When we looked at periods of longer than 12 months, the difference went away. most of that effect was driven by maybe two or so studies that had quite large effects in favor of low carb diets whereas most studies found no difference one of the challenges here is just that there's a lack of accuracy in reporting of energy between in outpatient studies and some evidence that indeed people may be a bit more accurate in their reports and thereby able to monitor their intake of calories a bit better when on a low carb compared to a low fat diet because low carb diets are more responsible to a low-fat diet because low-carb diets are more restrictive and involve more monitoring of what you're eating and fewer options. So overall, I don't find those long-term dietary studies to be very strong evidence, except in so far as what those studies show is that there isn't a large benefit in favor of low-carb diets.

If there is any benefit, it's small. And this is not what the carbohydrate into the model claims. The carbohydrate-insome model claims that there should be a very large effect of switching from low-fat to low-carb diet, not a small equivocal effect. And that's not a small, equivocal effect. that's maybe driven by these effects of monitoring your food intake. That's not the prediction, as I mentioned before. The response given by proponents of low-carb diets is that participants in these studies find adherence difficult, which means it's difficult to interpret results,

which, I mean, I agree with, but the problem is that that is not what the low-carb, the carbohydrate insulin model states. The carbohydrate insulin model rejects the idea that conscious restriction of intake is very important, at least for the majority of people. switching from a high carb to low carb diet should automatically reduce feelings of hunger, increase satiety, and also increase your rates of metabolic activity because of the

difference ways in which carbohydrates versus fats and proteins are processed by the body and the effects they have on insulin and deposition in fat tissues and so on. So this should happen sort of automatically. It shouldn't require a strong conscious restriction of caloric intake. And so the fact that there's difficulty in adhering to these diets is really not consistent with what carbohydrate insulin model predicts. Then there is the inpatient studies, the short-term studies that we talked about. So these controlled studies, particularly the one that I talked about, the whole Laetal study, these do not support the carbohydrate insulin model. Most of these

studies don't find any difference in basal metabolic rate depending on the low fat versus the low carb diets. And in fact, the whole Laetal one actually found that the intake of food was actually, that the caloric intake was greater in the low carb compared to the low fat diet, and this resulted in more loss of fat mass in the low fat compared to the low carb diet, which is contrary to the prediction of the carbohydrate insulin model. Most of the other studies find little difference in caloric intake and a little difference in basal metabolic rate. So again, these are contrary to what's predicted by the carbohydrate insulin model, which predicts that there should be a significant difference in satiety and in the metabolic rate driven by insulin, which turns to you.

driven by the glycemic loading of the diet. Proponents of the carbohydrate insulin model respond that two weeks is a too short period to have any observable effects, and therefore you need to have longer studies to really test this properly. I find this claim to be suspect because at least many of these proponents of low-carb diets offer health advice in the form of popular books and articles and things, in which they say that you should expect to have effects of the diet almost immediately on the first day or two. Now you could say, well, there's a difference between I mean, they're published scientific work versus their popular work, but again, I find the idea that it takes longer than two weeks for your body to adjust to this change, and therefore you don't observe any effects within the two weeks to be not very convincing. There's one final line of evidence, which I haven't mentioned before, but I'll just, the carbohydrate insulin model postulates that a major reason why insulin cause weight gain is due to the fact that insulin inhibits the lipolysis of fatty acids in atopor.

tissue. So it's essentially keeping the fat there rather than allowing it to be metabolized. Whereas the traditional mainstream view is that, well, it will be metabolized depending on whether that energy is required or not, like by the brain, for example, it says insufficient glucose. There is actually a way to test that, which is to artificially inhibit adipose lipolysis using a drug called axiomox, I'm probably mispronouncing that, but people who are treated with this, this treatment which inhibits the lipolysis of adipose, so it effectively prevents the use of fats that are stored in adipose tissues for energy purposes did not actually result in any significant differences in energy intake, resting energy expenditure or body composition.

Now this is contrary to what would be expected by the carbohydrate insulin model, because remember if it's the adipose tissue that is driving the hunger, driving the food intake and also driving the changes in metabolic rate, inhibiting lipolysis from adipose tissue should have a similar effect as a high glycemic index diet, which by increasing blood insulin levels also results in the inhibition of vatoposalopolis. But doing this artificially with this drug does not have that effect. Proponents of the carbohydrate insulin model respond to that the drug also affects other biochemical pathways, and so that makes the results hard to assess, which could be true, but also does feel like a bit of an ad hoc response to me. Now, if you're

interested in reading more about this and looking at the arguments for and against the carbohydrate insulin model and low carb diets. Two of the well-known authors who have published about this and whose articles I relied on are Kevin Hall, so he is a proponent of the mainstream low-fat diet and is an opponent of low-carb diets or a critic of low-carb diets. David Ludwig is a proponent of the low-carb diets and the carbohydrate insulin model has written a number of articles defending it. So if you want to read up on more about this, you can Google both of those names. One final comment that I'll make is in addition to all of the evidence that we've discussed, which, as I've argued, overall, I personally find is contrary to the claims of the carbohydrate

insulin model, and I don't find the evidence for low carb diets to be convincing. There is also the fact that low carb diets are typically quite high in meat and other animal products, which could lead to other negative health effects, depending on how well those are offset. So it could potentially lead to low intake of fiber, particularly for the kegogenic, the very low-carb diets is a potential issue and could also have effects for micronutrients, depending on how well those are supplemented or whether there's sufficient vitamins and minerals in whatever it is that you are consuming. There's also a potential ethical and environmental

implications of consuming a diet which is very high in meat or animal products. And so even if there is a potential dietary or health advantage, that must be considered in relation to those potential downsides. That's a bit of a separate question, but I did just want to mention it there because I think it is relevant when assessing the efficacy of these diets. So to conclude, as I've argued, I personally don't find the evidence for the carbohydrate insulin model or for the benefits of low-carb diets to be very convincing. I find that the postulated mechanisms are not well-specified enough. The core principle of it, that insulin, that increased insulin, results in fat deposition, which causes increased hunger, increased energy intake, and reduced metabolic.

rate is not well supported by evidence and the mechanisms by which that is supposed to happen are quite unclear. The long-term dietary studies show quite a small effect and I think are driven by a couple of studies and if there is any effect, it's probably due to differences in caloric intake rather than the actual mechanisms postulated by the low-carb diets and the carbohydrate insulin model itself. And the more controlled inpatient studies consistently do not find much of a difference in energy intake or metabolic changes like basal metabolic rate across the two studies. And in fact, the best controlled highest quality study that I'm aware of, the whole study actually found that there was a benefit to the low-fat diet relative to the low-calb diet

when it comes to fat loss and a lower energy intake on that diet. These reasons why the health bodies throughout the world generally still recommend a low-fat diet relative to a low-carb diet, because the evidence for a low-carb diet being superior really just isn't there. And low-carb diets are much more restrictive, as I mentioned before, and have potential downsides of it being more difficult to eat as many whole grains and vegetables and fruits, which are known to have, as we talked about in previous, but it's known to be associated with beneficial health outcomes, and it's particularly quite hard to get enough fibre on a low-carb diet as well. In addition to the potential higher cost,

and environmental impact of diets that are high in meat and animal products. This combined with, as I said, the lack of evidence and plausibility in favor of the carbohydrate insulin model is why most mainstream medical institutions and national dietary boards recommend low fat versus low carb diets and why I also tend to find that the evidence does not support the superiority of the low carb diet, the carbohydrate insulin model. So hopefully you found this interesting. that this episode may be controversial to some, so if you disagree or think that I miss something, feel free to send me an email. My address is Fods12.gm.com. That's F-O-D-S-1-2 at gmail.com.

If you'd like to support the podcast, you can make a donation via PayPal or become a Patreon supporter, and I'm very grateful to all of my current and past supporters. Another way to help the show is to leave a favorable review on Spotify or whichever other aggregator you use. Thanks very much for listening, and I'll talk to you next time.