Overview: Carbs, Insulin, Sugar and Obesity
This is a topic that clogs up every comment section about weight loss, calories, tracking, IIFYM, flexible dieting, "clean" foods, calorie deficits, energy balance, insulin resistance, metabolisms, and just about anything else nutrition related. Everyone has an opinion.
Sadly, Science doesn't care much for opinions. This is a look at the current body of scientific evidence on the topic of insulin, carbohydrates and body weight. This article hopes to highlight:
1: What is the carbohydrate-insulin model of obesity?
2: The evidence that falsifies this model
3: Why low carb diets may still work well for you, and the studies that support them as a useful method for achieving a calorie deficit.
The carbohydrate insulin model of obesity makes the following assertions:
1: Carbohydrates raise insulin levels
2: Insulin up regulates fat storage and down regulates fat metabolism.
3: This causes increased body fat
This is a widely held belief made popular in books such as:
The Case Against Sugar - Taubes
Why We Get Fat - Taubes
The Diet Delusion - Taubes
Fat Chance, the hidden truth about sugar, obesity and disease - Lustig
The Obesity Code - Fung
It is also the basis for arguments along the lines of:
1: Sugar and carbohydrates are the main drivers of weight gain.
2: People with insulin resistance, PCOS or hormonal imbalances can’t lose weight
3: Eating “clean” unprocessed foods are necessary for weight loss
4: Sugar and white bread are inherently bad for us
But is it true?
1: It is scientifically demonstrable that carbohydrates do raise insulin levels
2: It is scientifically demonstrable that insulin is a driver of fat storage and down regulate fat metabolism
3: The notion that increases in insulin directly cause weight gain, and weight gain is correlated to percentage of diet coming from carbohydrates have both been scientifically falsified.
So what are the studies that falsify the carbohydrate-insulin model?
Dr Kevin Hall conducted two studies, 2015 and 2016, both in metabolic wards. This means the amount of energy going in (food was provided to subjects) and the amount of energy going out (kcal expenditure measured in ward) was accurately measured.
Hall 2015 study found that:
Isocaloric diets conducted in a metabolic ward showed statistically significant greater weight loss in the low fat group than the low carb group across a 4 week period. This happened despite insulin levels being higher in the high carb group as expected. So not only did the low carb diet not show greater weight loss, it actually showed lower fat loss than the low fat diet.
Hall 2016 study found that:
Isocaloric diets conducted in a metabolic ward showed no difference in weight loss between a Ketogenic and balanced diet. The ketogenic diet was associated with small increases of energy expenditure that were “near the limits of detection with the use of state-of-the-art technology”. Again, insulin levels were shown to be higher in the high carb group, but this had no effect on the total weight loss.
Low carb advocates have previously suggested energy expenditure may be increased by 350kcal+ in ketogenic diets versus balanced diets. This was also falsified by this study that demonstrated an expected increase of energy expenditure to around 100kcal per day in the first week when switching to the ketogenic diet, but no increase was detected by the fourth week of the diet.
Systematic Review: Celeste E, et al (2014)
This review looked at all previous low carb vs low fat studies. The criteria for the review were as follows:
- Used isoenergetic diet comparisons (calories are the same in both diets)
- Used explicit cut-off ranges for macronutrients for treatment and control diets. (equal protein)
- Used interventions were of a diet component alone, or combined interventions were similar (i.e either both or neither groups took part in exercise)
- Used RCT’s only
19 studies met these standards and the results stated:
‘Trials show weight loss in the short-term irrespective of whether the diet is low CHO or balanced in terms of its macronutrient composition. There is probably little or no difference in weight loss and changes in cardiovascular risk factors up to two years of follow-up when overweight and obese adults, with or without type 2 diabetes, are randomised to low CHO diets and isoenergetic balanced weight loss diets.’
Big Data: Johnston BC, et al (2014)
Meta-analysis of ‘named diet plans’ across 7286 individuals concluded:
‘Significant weight loss was observed with any low-carbohydrate or low-fat diet. Weight loss differences between individual named diets were small. This supports the practice of recommending any diet that a patient will adhere to in order to lose weight.’
But what about all the studies I’ve seen that say low carb is better for weight loss?
There are plenty of studies that have shown low carb diets to be advantageous for weight loss. See my previous article for 23 of these studies. The issue with these studies, is they do not isolate the carbohydrate reduction as a single variable.
Common study limitations include:
> calories were not matched in both diets
> groups were told to simply reduce carb or fat intake rather than given specific targets
> intakes were self reported and open to under/over reporting
> protein was not matched in both groups
These studies serve as good evidence that reducing carbohydrate may be a successful tool to reduce overall calorie intake, and adhere to a weight loss diet. However, there are too many potential variables to be able to use these studies for evidence as a physiological role in the mechanism of weight loss/gain.
A great example was the Morenga et al, 2013 systematic review and meta-analysis on the subject that found the following results:
In adult trials without strict control on overall food intake, reducing dietary sugar intake was associated with decreases in body weight. Likewise, increases in sugar intake was associated with increases in body weight.
However, under isoenergetic conditions (same calories), replacing sugars with calorie equivalent macronutrients showed no change in body weight. This suggests the reduction in sugar has a positive effect on weight loss, due to the reduction in calorie intake, not because it is sugar.
Hall’s 2015 and 2016 studies falsify the carbohydrate-insulin model of obesity. Current research points towards energy balance as the mechanism of weight loss and weight gain. Low carb, low fat, portion control, intermittent fasting, or any other weight loss method may find success when trying to achieve a negative energy balance.
Reasons you might reduce carbs:
> If a lot of your calories come from carbohydrate sources
> If it encourages broadening the diet to include more nutrient dense foods rather than processed foods
> If you find moderating sugar or carb heavy snacks difficult and would prefer to remove them entirely than try to manage them sensibly.
> If you have a condition such as insulin resistance or PCOS. These conditions are very complex. While raised insulin may not have demonstrated causation of weight gain, it certainly can manipulate other hormones and their role in appetite. When satiety hormones are out of whack, it is not simply a case of “try to eat less”. The brain responds directly to these signals and your “will power” alone will rarely be enough to prevent you consuming additional calories.
> In a metabolic ward patients have no choice but to adhere to the calories they are given. In the real world, you may find that the practical implications of eating lots of refined sugars within your calorie target will inevitably lead to over consumption and therefore, exceeding your calorie target and gaining weight.
Reasons you might not reduce carbs:
> If you are able to sensibly manage the foods you eat to fit within your energy requirements, protein requirements, and micronutrient targets.
> If you are trying to lose weight and find any other method of achieving a calorie deficit more manageable.
Further Reading and Credits:
> There is a great podcast from Sigma Nutrition in which Kevin Hall speaks in more detail about his two studies above. He also addresses Gary Taubes’ defamatory remarks at a recent low carb conference and Taubes' subsequent apology in relation to these studies.
> Martin Macdonald is always a valuable source for up to date studies and excellent content. Here is his far more direct view on the insulin hypothesis
UPDATE AS OF FEBRUARY 2018
> A systematic review and meta-analysis from Sartorius et al (2018) adds further support to the above. "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 obesity".