There’s a theory – one with which you are probably familiar – that carbohydrates drive obesity, primarily through elevating the hormone insulin.
It’s often suggested that it’s not excessive calories that drive fat accumulation, but rather excessive carbohydrate.
This theory is often called “the carbohydrate insulin hypothesis”.
There are plenty of anecdotes of folks losing weight and improving their health with low carb diets who swear by this hypothesis.
Does this theory and do these anecdotes hold up against what we know about how the body works and what we see in studies?
Is it true that insulin drives obesity and that the key to losing weight is to restrict carbohydrates?
Fortunately, there have been a variety of studies that have looked into this question.
While a few of the studies we’ll be looking at are comparisons of low carbohydrate diets and low fat diets, that comparison is not the intent of this article.
We’re only looking at one question.
Does carbohydrate consumption independently drive weight gain?
If so, we should see that diets with low carbohydrate content result in weight loss while diets with high carbohydrate content result in weight gain.
Alternatively, we should see less weight loss and/or more weight gain with high carbohydrate diets relative to low carbohydrate diets.
Any weight loss at all on a high carbohydrate diet, however, would arguably suffice to demonstrate that carbohydrate intake does not exclusively drive fat accumulation.
Let’s see what these studies have got to say.
A Randomized Trial of a Low-Carbohydrate Diet for Obesity
The first study we’ll look at was published in The New England Journal of Medicine (NEJM) in 2003.
63 obese participants were randomly assigned to one of two groups.
The first group was instructed to initially restrict carbohydrate intake to 20 g per day and then gradually increase carbohydrate intake until their desired weight is achieved.
The second group met with a dietitian to learn about calorie counting and how to implement a high-carbohydrate, low fat, low calorie diet (1200-1500 kcal/day for women, 1500-1800 kcal/day for men, 60%/25%/15% carbohydrate/fat/protein).
A disadvantage to this study is that it does not set out to keep all variables controlled except for carbohydrate intake.
An advantage to this study is that it offers an opportunity to see how two different strategies – focusing on carbohydrates versus focusing on calories – affect weight in real life, as opposed to having food intake controlled.
Both groups lost weight.
While the low-carbohydrate group lost a bit more weight than the high-carbohydrate group, the difference is statistically insignificant.
That the high-carbohydrate group lost weight on par with the low-carbohydrate group – or at all – suggests that carbohydrates alone don’t drive weight gain.
Is it possible that the high carbohydrate group lost weight because they were eating less carbohydrate than at baseline?
Perhaps, although baseline dietary data is unavailable.
“Comparison of the Atkins, Ornish, Weight Watchers, and Zone Diets for Weight Loss and Heart Disease Risk Reduction”
The next study we’ll look at was published in the Journal of the American Medical Association (JAMA) in 2005.
Researchers randomly assigned 160 participants to one of four dietary strategies – Atkins (carbohydrate restriction), Ornish (fat restriction), Weight Watchers (calorie restriction), or Zone (balanced macronutrients).
After one year, the participants in each group lost the following amount of weight:
- Atkins (carbohydrate restriction): 4.6 lbs. (2.1 kg)
- Ornish (fat restriction): 7.3 lbs. (3.3 kg)
- Weight Watchers (calorie restriction): 6.6 lbs. (3.0 kg)
- Zone (macronutrient balance): 7.1 lbs. (3.2 kg)
Adherence was low, although the Atkins group did consume less carbohydrate than the other groups according to self-reported dietary intake (for what that’s worth).
The Atkins group ate approximately 68 g, 137 g, 190 g, and 190 g carbohydrate at 1 month, 2 month, 6 months, and 12 months.
The Ornish group ate approximately 220-230 g carbs through the entirety of the experiment.
Most relevant is that the Ornish group’s carbohydrate intake did not change from baseline, yet they still lost weight over the course of the study.
If carbohydrates are the cause of weight gain, how did this group lose weight with no change to their carbohydrate intake?
Not only that, but the Ornish, Weight Watchers, and Zone groups all lost more weight over the course of the year than the Atkins group, even though they ate more carbs.
If carbohydrates are what drive fat storage, why would the carbohydrate restricted group see less weight loss than the other groups?
“Comparison of Weight-Loss Diets with Different Compositions of Fat, Protein, and Carbohydrates”
A study similar to the one just discussed was published in NEJM in 2009 and didn’t look solely at different carbohydrate levels, but also varying protein and fat levels.
In this study the researchers assigned 811 participants to one of four diets of varying macronutrient composition:
- 20% fat, 15% protein, 65% carbohydrate
- 20% fat, 25% protein, 55% carbohydrate
- 40% fat, 15% protein, 45% carbohydrate
- 40% fat, 25% protein, 35% carbohydrate
All four diets were designed for similar fiber intake and calculated caloric deficit per the participants’ reported resting energy expenditure and activity levels (750 kcal).
Over two years the participants lost an average of 8.8 lbs. (4 kg).
All groups – even the high carbohydrate group – lost weight.
While the researchers concluded that weight loss was not statistically significantly impacted by high vs. low protein, high vs. low fat, or high vs. low carbohydrate, let’s look at the numbers in more detail.
Those who ate the least carbohydrate lost 1.5 lbs. (0.7 kg) more weight than those who ate the most carbohydrate over the course of those two years.
Those who ate the most protein lost 2 lbs. (0.9 kg) more weight than those who ate the least protein.
These results suggest that – at least in this study – protein intake had a bigger influence on weight than carbohydrate intake.
“Calorie for calorie, dietary fat restriction results in more body fat loss than carbohydrate restriction in people with obesity”
Next, let’s look at a study published in Cell Metabolism in 2015, during which participants were brought in to a metabolic ward where their energy expenditure could be monitored and their food intake controlled.
This study was specifically designed to evaluate whether or not there is any kind of “metabolic advantage” to a low carb diet, as suggested by many advocates.
Nineteen participants would spend 5 days consuming a baseline diet designed for eight maintenance, then be split into two groups.
One group would spend the next six days consuming a carbohydrate-restricted diet, while the other group would spend the next six days consuming a fat-restricted diet.
Both the fat restricted diet and the carbohydrate restricted diet were designed to be 30% lower in calories than the baseline diet.
The participants would then be let out for a two to four week “washout” period, then readmitted to repeat the process.
The second time through, however, the group that had previously implemented the carbohydrate-restricted diet would implement the fat-restricted diet and vice versa.
Protein content was held constant (100-105 g) throughout the baseline and intervention diets, and total energy (1918 kcal) was identical for the two intervention diets.
What the researchers found was that net daily fat loss was greater with fat restriction than with carbohydrate restriction (89 g/day vs. 53 g/day), even though carbohydrate restriction resulted in a 22.3% reduction in insulin secretion.
This greater rate of fat loss resulted in 463 g total fat loss during six days of fat restriction and 245 g fat loss during six days of carbohydrate restriction.
If carbohydrates make us fat, why would the low fat high carbohydrate diet produce any weight loss at all?
Not only that, but why would fat restriction result in more fat loss than carbohydrate restriction?
The authors acknowledge in their discussion that the short duration of this study is a limitation.
“Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men”
The next study we’ll look at was published in The American Journal of Clinical Nutrition in 2016 by several of the same researchers as the study we just covered.
This was another metabolic ward study in which the participants’ energy expenditure could be monitored and dietary intake could be controlled.
The intent of the study was to assess whether a ketogenic diet was associated with changes in energy expenditure, amount of fat vs. glucose burned, or body composition.
The participants spent four weeks consuming a high-carbohydrate baseline diet before switching to an isocaloric (identical energy content) ketogenic diet for four weeks.
Both diets provided 2398 kcal/day and 91 g protein.
The baseline diet offered 300 g carbohydrate and 93 g fat, while the ketogenic diet offered 31 g carbohydrate and 212 g fat.
Energy expenditure during the ketogenic diet phase was calculated to be approximately 57 kcal/day greater when measured by metabolic chamber and 151 kcal/day greater when measured by doubly labeled water than energy expenditure during the baseline diet.
Even though fat loss was not intended during the baseline phase, it was observed throughout both phases of the study and actually slowed down during the ketogenic diet phase.
Thus, fat loss was observed even with a carbohydrate intake of 300 g/day.
Changes in respiratory quotient showed a shift away from carbohydrate metabolism and towards fat metabolism during the ketogenic diet phase.
However, the rate of fat loss slowed upon a transition to a ketogenic diet, even with a small increase in energy expenditure and proportion of energy derived from fat.
The authors mention in their discussion that they suspect the increased energy expenditure may have been overestimated due to water weight loss, which is typical upon transitioning to a carbohydrate-restricted diet.
“Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults and the Association With Genotype Pattern or Insulin Secretion”
The next study we’ll look at was published in JAMA in February 2018 and is one of my favorite studies.
609 participants were randomly assigned to restrict either fat or carbohydrate, and then after two months gradually add incremental amounts of the restricted macronutrient back into their diets until reaching a level they could maintain indefinitely.
The participants were screened at the beginning of the experiment for genetic patterns and insulin profiles that might suggest better suitability for either a low carbohydrate or low fat approach.
Both groups were eating 240-250 g/day carbohydrate at baseline.
The low fat group ate 200-215 g/day carbohydrate throughout the study, while the low carbohydrate group ate 100-130 g/day carbohydrate.
Again, both groups lost weight, even with a mean difference of 75-100 g/day of carbohydrate over the course of a year.
The low fat group lost 11.7 lbs. (5.3 kg) body weight and the low carb group lost 13.2 lbs. (6.0 kg) body weight.
While there may be a slight edge – 1.5 lbs. over the course of a year – in weight loss with a low carbohydrate approach, the researchers concluded that the difference in weight outcomes were statistically similar.
The researchers also saw that weight loss and improvement in health risk factors were not affected by how the assigned diet aligned with the baseline genetic and metabolic profiles.
Participants’ success did not appear to be influenced by how well their genetics of insulin levels aligned with their assigned diet.
One thing I really like about this study is that the researchers emphasized food quality.
I’ll touch more on this point towards the end of this post.
Does high-carbohydrate intake lead to increased risk of obesity? A systematic review and meta-analysis.
Let’s shift gears a bit and, rather than look at single studies, look at meta-analyses (studies of multiple studies).
The first meta-analysis we will look at was published in BMJ Open in 2018.
The authors looked through hundreds of studies and identified 22 observational studies of disease-free individuals to evaluate the relationship between carbohydrate intake and risk obesity.
The authors looked out outcomes through two different lenses:
- Risk of obesity with high-carbohydrate intake versus low carbohydrate intake
- Risk of obesity as percentage of total dietary carbohydrate intake increased
The authors found a weak increased risk of obesity with high-carbohydrate versus low carbohydrate and a weak decreased risk of obesity with increasing dietary carbohydrate intake.
The authors concluded, “based on our findings it cannot be concluded that a high-carbohydrate diet, or increased percentage of total energy intake in the form of carbohydrates, increases the odds of being obese.”
Carbohydrate quantity in the dietary management of type 2 diabetes: A systematic review and meta-analysis
The second meta-analysis we’ll look at was published in 2018 in Diabetes, Obesity Metabolism, and explored the effects of low carbohydrate diets on body weight and blood markers in adults with type 2 diabetes.
Rather than look at observational studies, as the meta-analysis we discussed above examined, this meta-analysis looked at randomized control trials.
The authors found 23 studies including 2178 participants, and concluded that low carbohydrate diets were not associated with greater weight loss than high carb diets in long-term or short-term studies.
If you’ve found it easier to look, feel, and perform your best on a low carb diet, have at it!
If, however, you’ve found it easier to look, feel, and perform your best on a higher carb diet, have at it!
These studies demonstrate that weight loss is comparable across a wide variety of carbohydrate intakes.
While carbohydrate restriction is an effective tool for weight loss, carbohydrate restriction is not required for weight loss.
There may be plenty of reasons to go low carb.
The carbohydrate insulin hypothesis, however, doesn’t appear to be one of these reasons.
Be careful not to miss the forest through the trees.
I’ve written before about how consistently eating quality food – more than relative carb or fat intake – is the most overlooked factor in physical transformation.
Work on eating more real food and less processed junk – consistently – and see how things go.
You’ve got this.