Do Carbohydrates Control Body Fat?


Do Carbs Control Body Fat?

Carbohydrates have gotten a “bad rap” because they elicit an insulin response. And as everyone knows, insulin causes us to become fat (or does it?). In fact, the carbohydrate-insulin-obesity hypothesis has been blamed as the major cause of the obesity epidemic. Unfortunately, when we look closer at the science this idea doesn’t quite hold up

Does Insulin Increase Body Fatness?

 Common reading literature such as Good Calories, Bad Calorieshave put the hormonal regulation of body fat into limelight. In books such as GCBC, insulin is painted as the primary cause of common obesity. While it is true that hormones play a critical part in the regulation of body-fat, it is naïve and myopic to single out insulin as the sole contributor. Our endocrine system is extremely robust with almost countless signaling molecules. While insulin plays a critical part in substrate partitioning and metabolism (including the fate of fat and carbohydrates), it plays only one part in the signaling cascade.  To ferret out whether insulin is in fact the main regulator in body fatness and if increased insulin levels due to carbohydrates cause obesity we need to look at the literature.

Insulin Action: Muscle, Liver, and Adipose Tissue

Insulin is an endocrine hormone, meaning it affects multiple tissues in the body. Specific to our focus is skeletal muscle, the liver, and our adipose tissue (fat). However before we get there we need to note that insulin first acts in the liver to reduce hepatic glucose output.  

When skeletal muscle is “exposed” to insulin several things happen within the muscle cell. First, glucose transporters (e.g. Glut4) move to the cell surface and begin transporting glucose into the muscle cell. Secondly, insulin signals the muscle cell to shift to carbohydrate-based metabolism. Now it is important to note that the energy status of the cell plays a role in exactly what happens in the muscle cell. If the muscle cell is low on glucose and/or muscle glycogen the insulin signal will instruct the cell to utilize the incoming glucose for fuel and to begin creating muscle glycogen from any spare glucose that is coming in. 

Additionally, insulin will cause an uptake of fatty acids into muscle cells (Figure 1). If the muscle cell is already full of glucose and glycogen and topped off with intramuscular triglycerides then insulin signaling will do all of the things it does in a low energy state but will also turn excess glucose into fat through de novo lipogenesis (Figure 1). de novo lipogenesis from carbohydrates is actually quite “hard” to induce in humans (here are several papers on the topic ref #1, ref #2, ref #3, ref #4)
figure 1

What about the liver? The liver is one of the main sources of regulating blood sugar in the body. When our body sugar gets too low a hormone named glucagon signals the release of sugar from the liver (hepatic glucose output). Insulin, which is release in cases of elevated blood sugar due to carbohydrate (or protein) ingestion, signals the liver to stores its glucose, convert extra to liver glycogen, and then when that is full to synthesis fatty acids from extra glucose. Again, we do see that insulin can have a “lipogenic” effect. However, it is important to note the lipogenic effect of insulin on the liver is only seen in the case of saturated glucose and glycogen stores, which in essence is a state of excess energy.

Now to discuss the star of the “insulin-obesity” hypothesis, the role of insulin and our adipose tissue. Insulin does in fact decrease the rate of lipolysis in adipose tissue and does stimulate fatty acid and triacylglycerol synthesis (Figure 2). The science is crystal clear on this.    

Figure 2

Now from this brief overview it would indeed appear that insulin promotes fat storage… well unfortunately that is the wrong wording for it. Insulin is “fat-sparing”, not fat promoting. Yes, insulin does reduce fat oxidation and promote a lipogenic environment. However, for insulin to have a profound effect on body fatness and fat gain, we would have to have a constant high signal of insulin, something that is not typically seen. Our bodies default metabolic state is one of fat oxidation and insulin is simply the switch that turns it to carbohydrate metabolism (Figure 3).  The net effect of insulin is simply a switch in substrate metabolism, it does not produce a large “obesegenic affect” due to dietary composition alone. FOR INSULIN TO HAVE A SIGNIFICANT OBESEGENIC EFFECT IN A HEALTHY INDIVIDUAL IT MUST OCCUR IN THE PRESENCE OF EXCESS ENERGY.

Figure 3

Is there evidence of this? Put simply yes. Our body functions without (or very low levels) of insulin most of the time.  In this state we are burning fat for fuel. In extreme cases, such as Type 1 Diabetes where there is ZERO insulin they are burning almost exclusively fat. Only when individuals with T1DM administer exogenous insulin do they switch to carbohydrate metabolism. This is evidence that insulin is essentially the conductor of our metabolic train. It tells us when we should burn sugar because we have enough of it to burn, but it does not tell our body that we should become fat. 

One additional note is that T1DM also tells us that insulin can’t be an “evil fat promoting hormone” because without insulin we can die… the key to insulin and optimal health is having proper insulin signaling.

If insulin were the key hormone responsible for fat gain than we would expect people who consume more carbohydrates as a percent of their diet to have a higher percentage of body fat than those who consume less. Well, we don’t see this either anecdotally or in the research. Anecdotally, we can look at populations such as the Kitavans or Okinawans whose diet consists or a large portion of carbohydrates yet are a rather lean population.

To prove this point scientifically, lets look a study that addresses the following question: do isocaloric diets that result in higher insulin levels increase body fatness?

One study had 8 participants consume a high-carbohydrate diet (60% CHO) for 7 days and a high-fat diet (60% fat) for 7 days and measured total energy expenditure and nutrient oxidation. The study showed that the composition of the diet did not effect total energy expenditure (a key component to long-term weight gain) but DID “rapidly shift substrate oxidation to closely reflect the composition of the diet” (1). Recently, a study exploring diets that contained 70% carbohydrates showed that it was possible to lose weight and improve insulin sensitivity.  Another study conducted by Dr. Kevin Hall showed that calorie for calorie, fat restriction leads to greater weight loss than carbohydrate restriction over short periods (2).  This suggests that insulin signaling plays a major role in determining whether fat or carbohydrate is used as a primary fuel source, but not a major role in overall energy expenditure. 

Additionally, if insulin were the driving factor in body fatness we would expect virtually all overweight or obese individuals to have elevated insulin levels to cause weight gain. Such is not always the case. In fact, there is a large portion of overweight individuals with normal, “healthy” insulin signaling (3).

We should also point to one critical piece of evidence: protein also elicits an insulin response. In fact, some high-protein foods elicit a greater insulin response than high-carbohydrate foods (has your brain exploded yet? I know mine did when I first read this article). Why is this important? Well the most consistent finding from nutritional research is that diets higher in protein are effective in reducing body-fatness. Whey protein specifically has been shown to elicit large insulin responses yet it has been repeatedly shown that whey protein can help reduce body fat in wide ranges of populations.

So if insulin were the main culprit in body-fatness than consuming protein would also encourage fat accumulation and not fat loss.

When we take all this information together we can draw the following conclusion: insulin is a substrate regulator. It switches your body from burning mostly fat to mostly carbohydrate, promotes glycogen synthesis, and protein synthesis if necessary. In a metabolically healthy individual, the metabolism is flexible and robust enough to handle fluctuations in substrate and maintain leanness even in the presence of a high-carbohydrate intake that would, on the surface, seem to promote a “fat-accumulating” environment.  Only in the presence of excess energy does insulin truly promote a fat-accumulating environment, even then our body has a nifty way of attempting to attenuate that environment.

Insulin and Obesity in Humans

I hate beating the dead horse (the poor thing is already a goner), but this is an important concept I want people to understand. It will be the basis for our approach to nutrition and fueling your performance in the gym. As a coach, I need us to be on the same page on the central idea of the last article and this one: carbohydrates are not the sole reason for body fat and in some cases increasing carbohydrate intake can be beneficial.

figure 1

Image c/o

As we discussed earlier there is substantial common-reading literature that suggests that hyperinsulinemia is the driving cause of obesity in America. Specifically that increased sugar consumption leads to hyperinsulinemia which then leads to obesity. I would argue that increased sugar consumption has significantly contributed to the rise in obesity and metabolic diseases over the past few decades; however, from the data we will discuss in this post the increased insulin response leading to lipogenesis is not likely the mechanistic cause.*

Insulin Levels and Body Fat

Insulin and obesity

Figure 4

Insulin levels are typically increased in obese individuals (figure 4); however it is not requisite. As mentioned last week, there are cases (more than you would think) where obese individuals do not show increased fasting insulin. For a reference, Hivert and colleauges reviewed the literature and showed the wide ranges of conflicting data on this topic back in 2007

This suggests that insulin is not the direct cause of common obesity. If it were, we would observe increased insulin in all cases of obesity (barring genetic disorders, extreme cases, or late stage T2DM where beta-cell dysfunction drops insulin levels). 

For something to be a direct “cause” to “effect”, like stating that insulin causes obesity you would have to demonstrate (within measurable experimental error) that rises in insulin always precede obesity and are present in the obese state, which has not been shown.

Insulin Levels Rise and Fall with Body Weight

Ok, so now it may be plausible that increased fasting insulin is an effect of obesity, not a cause. Well, there is evidence to suggest this is likely the case. For example, track insulin as body weight moves up and down the scale; we also observe changes in fasting insulin with weight change. In studies where participants gain weight there is a concordant rise in insulin (Figure 5). Conversely, weight-loss leads to a decrease in insulin (Figure 6).

Separately it is difficult to tease out whether it is a cause or effect of weight gain but when taken together it is more likely that insulin levels are affect by weight, specifically adipose tissue.

Insulin and weight gain

Figure 5


Figure 6

What is going on?

One of insulin’s main roles is to maintain glucose homeostasis. Briefly, increased insulin promotes glucose uptake into tissues. Therefore, we would expect that increase levels present in obese individuals would result in lower blood glucose levels; however, what we observe is an increase in glucose (figure 7).

Figure 7

This suggests that while there is increased insulin, it is not functioning properly. While there is increased insulin in obese people (in most cases), it appears its action is reduced. This would explain the concurrent rise in both insulin and blood glucose

Additionally, in adipose tissue, insulin promotes lipogenesis and reduces lipolysis.

Therefore, we would observe decreased plasma fatty acids (FA) because insulin would promote fat uptake into the adipose tissue and suppress fat release. 

Figure 7

Figure 8

However, we see the opposite; increased plasma FA in obese and diabetics, suggesting insulin action on adipose tissue is also decreased (figure 8) . If insulin action were increased in obese individuals due to the elevated insulin than we would observe decreased plasma FA as insulin action promotes lipogenesis and decreases lipolysis in adipose tissue; however we see the opposite in obese individuals.

This information suggests that in obese individuals, insulin is elevated, but the data suggests insulin action is actually decreased.  This “jives” with the literature that obese individuals develop insulin resistance, which may eventually leads to type II diabetes. When we take all of this together it appears that elevated insulin is not the main cause of common obesity or fat gain.

Conversely, obesity is likely the cause of insulin resistance.  The main take home here is that insulin is not the culprit behind common obesity and substantial increases in adipose tissue.
Now lets consider the topic covered in the last post “Carbohydrates and Insulin: Do Carbs Make You Fat”. In that post we clearly indicate that insulin has specific actions on substrate metabolism. Specifically, insulin promotes carbohydrate metabolism and reduces fat metabolism. Additionally, while insulin doesn’t cause obesity, it can have short-term effects on how nutrients are stored in the body. This suggests that manipulating insulin through dietary modification may help improve recovery (i.e. muscle glycogen synthesis and muscle protein synthesis) and can be used as a tool to improve body composition in certain cases. 

Hopefully the last two posts have driven home the following points about insulin: 1) it is a substrate conductor that tells your cells what to use for fuel, 2) is is most likely a result of obesity than it is a cause, 3) we can manipulate it to optimize fuel portioning to promote fat oxidation and recovery from workouts. 

*As a disclaimer to this article  Dr. Stephan Guyenet presented a substantial amount of the data presented here at AHS13. I am utilizing this information to convey my message here and expand upon it to set the stage for the implications of dietary intake.


Follow Us on Facebook


21 thoughts on “Do Carbohydrates Control Body Fat?

  • I find this very interesting being a T1D myself and being concerned about having to increase my insulin intake due to increased carbohydrates to support my activity level. T1D do tend to take more insulin than a non-diabetic in general though, so would you say that this would make it harder for a T1D to lose weight than a normal individual? I have slightly gained (body fat % went up) since increasing my carbohydrates/insulin to support them (according to my Wave #’s). Perhaps T1D’s still do need slightly less carbohydrates than their equivalent non-T1’s because they already take more insulin? Would be interested to know.

    • T1D does present an interesting case as the insulin is not endogenously regulated, but rather exogenously. It is possible to stay lean while using insulin with control over calorie intake and insulin use. One approach is to eat higher protein and fat and then use enough carbohydrates to fuel your training and minimize the use of exogenous insulin 🙂

      As the normal disclaimer, I am a PhD not an MD so for medical advice make sure to consult your physician 🙂

  • The increase in fasting glucose is likely a signal of initial phases of insulin resistance. It can be reversed by losing weight and exercise. It is unlikely that medication is needed at this point.



    • Which references would you like to see? There are a lot of concepts in this article that can’t be attributed to a single study. In those cases they are not cited. In cases where specific data points are mentioned there are references embedded in the text. Just click on the reference number 🙂

  • Just curious if you have read any of Dr Ray Peat’s theories on this? Seems as though he is particularly keen on PUFAs being a major player in this dysregulation. I mean is it simply just a.matter if the aging process that we keep gaining weight ? Just seems that resistance has roots to something we are all doing? Great post for sure, appreciate the effort.

    • I’d say Ray Peat would be entirely wrong as there are studies showing that PUFA’s aid insulin sensitivity. In particular, one study showed that replacing saturated fats in the diet with PUFA improved insulin sensitivity for muscle uptake of glucose for people with insulin resistance of the form of muscle glucose uptake (though not for those with livers that have developed insulin resistance).

      • The literature regarding PUFA vs SFA and insulin resistance is pretty all over the map. It depends on the cell type, assays used, populations used (when looking at humans), diet history, etc. There aren’t any good data I am aware of that convince me about the superiority of SFA over PUFAs. I’d be happy to read anything you send my way on the topic 🙂

  • Insulin levels rise in Pima Indians before they get fat, not the other way around. You get fat because you’re insulin resistant.

    • Dr. Garry Lee,

      Thank you for the interest in our week. I would be interested in reading the paper that presents the data for which you are drawing that conclusion? Virtually all other data suggests the opposite; that insulin levels rise as a result of skeletal muscle dysfunction and obesity, even in the Pima Indians (I am actually working directly with that data set as we speak). It is indeed possible to become insulin resistant without becoming obese; this typically happens as a result of cellular overload, altered cell metabolism (some nod to muscle triglyceride abnormalities), and a few other things.

      Relevant References

  • Great article. It’s nice when someone breaks it down. My husband is currently on Testosterone shots, no estrogen blocker, all blood work, etc., is fine. But he has greatly gained weight, not in strength, and even in a caloric deficit, keeps gaining weight.

  • If you set up the straw-man that it is all due to insulin, you will be able to show that the man is wrong. There is a big difference between “it is always due to” and “is a major controlling variable.” I always describe metabolism by analogy with American football. You will get very far following the quarterback but any particular play may be due to the size of the defensive center or, more likely, everything.

    If you really want to know about carbohydrate and insulin, check out the FB page called TYPEONEGRIT. You can see how much better glucose control is at low dietary carbohydrate. The major reason that people with T1 don’t take their insulin is because of fear of getting fat. Some of that may be due to not excreting glucose but I suspect, careful examination of that phenomenon would tell you a lot.

    And of course, the real value of low-carbohydrate diets can be seen in the pathetic, obsessive campaign of the medical establishment to deny its efficacy and raise “concerns,” none of which ever pan out. (And therefore we hardly need another speculation on what’s wrong with them). It might be a good time to figure out how to make them work. IMHO

    • Thank you for the thoughtful insights and comments. I am a bit confused regarding your last two points. Do you have data to support the idea that taking insulin to control blood sugar has any substantial influence on common obesity among people with type 1 diabetes? I am unaware of any such data.

      Secondly, the idea that low-carb diets can be helpful in the management of type 2 diabetes is not in the same ballpark as asking the question of whether the carbohydrate-insulin hypothesis is a large contributor to common obesity. That is equivalent to saying “our body gets obese by storing fat, therefore if we never eat fat we also can’t get obese”. So it seems to me that the data supporting low carb diets as a potential therapeutic strategy for losing weight and reducing risk factors still doesn’t get us to the carbohydrate-insulin hypothesis.

      The last point I think is also a bit of a straw-man. The sentence,”pathetic, obsessive campaign of the medical establishment to deny its efficacy and raise ‘concerns,'” I think is missing a few key pieces. The medical community has published several studies clearly stating that hypocaloric low-carb diets are a potentially useful tool, but couches them with the valid point that there can be some complications. Additionally, several studies show similar outcomes with varying degrees of carbohydrate intake. I have read your critical review and find it very comprehensive and rich in data; however, the conclusions from the paper don’t strike me as strong enough to support the idea that low-carb should be the only first line approach for several reasons, with the primary reason being pragmatic. I whole-heartedly agree that they should be considered as an option as they can be as effective as other approaches. Low-carb diets are indeed one tool, just not the only tool.



  • This is a fantastic summary of the data on insulin and its effects on the body. But as a female who has been diagnosed with PCOS, I still wonder what role insulin might be playing and whether there is evidence that low carbohydrate diets are beneficial for these types of scenarios. I would LOVE to know that low carb is not required in such cases, but there is very little data to convince me one way or the other.

Comments are closed.