A in the Journal Nutrition and Diabetes has raised some interesting questions about the effects of a high-fat diet on glucose regulation in New Zealand obese mice 1. While only a single study, it has set off a media storm questioning the efficacy of a Paleo-style diet.
Which is interesting because what may be most important about the study is not the questions it raises, but the fundamental and all too pervasive misunderstanding it demonstrates about the Paleo Diet.
The study explores the effects of a very high-fat/low-carbohydrate diet on diabetes prone mice. But as I explained in an article I wrote last spring, there is a danger in talking about the Paleo Diet purely in terms of macronutrient ratios. Hunter gatherers didn’t know what a carbohydrate or fat was. Instead, they ate particular foods while avoiding others. This is one of the most important and progressive aspects of the Paleo diet. It is about what whole foods we should eat (lean natural meats, vegetables, fruits, and nuts) and which we should avoid (grains, legumes and dairy.) All based on how we evolved.
Within the context of these healthy foods, there is a wide range of possible healthy ratios.2
Yet, many in the nutrition community remain locked to trying to determine the “optimal” nutrient ratio while lumping all fats and all carbohydrates together. That is the fundamental mistake of this recent study led by Dr Andrikopoulos and his team in Australia. In their attempt to criticize the Paleo Diet based on these ratios they nearly gloss over the fact that their diet is based on non-Paleo foods like cocoa butter and casein. Foods that, as I’ll explain momentarily, have been proven to have a direct impact on glucose regulation. As a result, the study provides no clarity on the Paleo Diet whatsoever.
But as Dr Cordain says, the devil is in the details. So let’s look more closely at the low-carbohydrate/high-fat diet fed to the mice in the study. The diet broke down as follows:
- Carbohydrates came purely from sucrose (simple sugar)
- The fats were 55% saturated fat, 37% monounsaturated fat, and 8% polyunsaturated fat
While the nutrients ratios can vary in a Paleo diet, an early paper analyzing a sample Paleo Diet had very different ratios3.
- The carbohydrates in this sample menu were primarily complex carbohydrates
- Of the 39% fat in the diet, 7% was saturated, 17.2% was monounsaturated and 10.4% was polyunsaturated
Consequently, even from a macronutrient ratio perspective, the diet fed to the mice did not resemble a Paleo diet.
But the question remains if the ratio in Andrikopoulos’ study could be consumed based on Paleo-friendly foods. We may not know because the study did not attempt to use Paleo foods. In a separate , Andrikopoulos detailed the high-fat/low-carbohydrate diet fed to the mice. The table below shows the breakdown of the diet:
|Primary||g/kg – %|
|Cocoa Butter||400 – 40.0%|
|Casein||200 – 20.0%|
|Sucrose||106 – 10.6%|
|Canola oil||100 – 10.0%|
|Clarified butter fat (Ghee)||100 – 10.0%|
|Cellulose||50 – 5.0%|
|Calcium carbonate||13.1 – 1.31%|
|AIN93G vitamins||10 – 1.0%|
|Potassium dihydrogen phosphate||6.9 – 0.69%|
|DL-Methionine||3 – 0.3%|
|Choline chloride (75%)||0.25%|
|AIN93G trace minerals||0.14%|
So, in a study that’s being used to criticize the Paleo diet in the media, virtually none of the foods consumed are part of a Paleo Diet. In fact, the “healthier” control diet is in some ways closer to the Paleo Diet with foods like fish meal, only 2% of carbohydrates coming from simple sugars, and a much lower saturated fat content.
More importantly, 60% of the mouse diet (by weight) came from cocoa butter and casein. Yet Andrikopoulos and his team failed to reference many studies exploring the effects of these specific foods on diabetes and glucose regulation.
Among these, a 1991 study tested the effects of saturated vs polyunsaturated vs monounsaturated fats on insulin sensitivity in rats4. The saturated fats which came from cocoa butter hurt peripheral glucose utilization. And even though this earlier study showed better glucose control with polyunsaturated fats, the control diet in Andrikopoulos’ study had a much higher ratio of polyunsaturated fats than the high-fat diet.
Likewise, a study in Diabetologia looked at the effects of casein in two rodent models and found “milk caseins are unlikely to be exclusive promoters of type I diabetes, but could enhance the outcome of diabetes”5. But it is important to point out that this study explored type I diabetes.
A more recent 2006 study in Nutrition Research explored the effects of casein on the insulin/glucose response in normal rats. After an overnight fast, a glucose challenge was found to raise both insulin and glucose levels in the casein-fed rats6. Interestingly, the primary finding of the Andrikopoulos study (which is the basis for the media’s criticism of the Paleo Diet) was virtually identical. A glucose challenge raised the insulin and glucose levels of mice fed the high-fat diet. The results from both studies are shown in the figures below (figure 4 from the recent Andrikopoulos study and figure 3 from the 2006 study on rats):
Andrikopoulos’ group attributed these results to increased insulin resistance. The 2006 study drew a different conclusion: “these findings are consistent with the insulinotropic action of amino acids and with the role of some of them as gluconeogenic precursors.” Again it’s interesting that Andrikopoulos did not reference this study despite having virtually identical results with the same food item.
So can we attribute Andrikopoulos‘ results to casein? No, but that’s the point. Correlation is not cause and Andrikopoulos provides no evidence to prove his results are due to the high-fat/low-carbohydrate ratio of the diet, to casein in the diet, or any of a number of other variables not even addressed. To then go to the media and claim it is evidence of the negative health effects of a Paleo Diet is simply irresponsible.
While Andrikopoulos’ study may raise questions about potential dangers of a very particular high-fat diet, it represents itself an even greater danger. That of drawing any conclusions about an overall diet strategy based on a macro-nutrient ratio while ignoring the type of foods themselves. Until Andrikopoulos can address these very glaring holes and test a diet based on evolutionary foods, I don’t personally plan on changing my Paleo-based diet anytime soon.
Nutrition & diabetes. 2016;6:e194.
 Cordain L, Miller JB, Eaton SB, Mann N, Holt SH, Speth JD. Plant-animal subsistence ratios and macronutrient energy estimations in worldwide hunter-gatherer diets. Am J Clin Nutr. 2000;71(3):682-692.
 Cordain L. The nutritional characteristics of a contemporary diet based upon Paleolithic food groups. Journal of the American Nutraceutical Association. 2002;5(5):15-24.
 Lardinois CK, Starich GH. POLYUNSATURATED FATS ENHANCE PERIPHERAL GLUCOSE-UTILIZATION IN RATS. J. Am. Coll. Nutr. 1991;10(4):340-345.
 Beales PE, Elliott RB, Flohe S, et al. A multi-centre, blinded international trial of the effect of A(1) and A(2) beta-casein variants on diabetes incidence in two rodent models of spontaneous Type I diabetes. Diabetologia. 2002;45(9):1240-1246.