You’ve seen the news sites blowing up with variations of this headline “New Scientific Study Proves Paleo Diet Causes Weight Gain and is Dangerous for Diabetics”. And you’ve seen the many impassioned responses in the Paleo community. Both of which have me hulking out. Yes, I find myself in the very awkward position of needing to rebut both how this new scientific study and its implications are misrepresented in the media as well as some of the knee-jerk anti-science reactions from the Paleo community. Just in case you haven’t seen the articles, let me bring you up to speed first.
The Media Misrepresents Paleo to Critique It…. Again
A new scientific study was published online on February 15, 2016 in the journal Nutrition & Diabetes. The researchers are based at the University of Melbourne in Australia.
The scientific article is titled “A low-carbohydrate high-fat diet increases weight gain and does not improve glucose tolerance, insulin secretion or β-cell mass in NZO mice”.
I didn’t read the word Paleo there, did you? As far as I can tell, the application of this study to the Paleo diet originated from the University of Melbourne press release to the Australian media, picked up on February 18th. The stories have since percolated internationally, and can now be found on mainstream media websites in many countries. It should be noted that Australia has tabloid-style media with a history of very aggressively attacking the Paleo diet while completely misrepresenting it (here’s my rebuttal to a previous Australian media-led critique of the Paleo diet full of inaccurate description of what Paleo entails).
Actually, the study that spurred this anti-Paleo media frenzy in no way evaluated the Paleo diet, but instead a ketogenic diet. And while, there’s certainly people who opt to combine these two dietary approaches, the standard Paleo diet in no way resembles keto in its macronutrient breakdown. Not only that, but the low-carbohydrate high-fat diet fed to the mice in the study includes ingredients that are unequivocally not Paleo foods.
It seems as though the senior author of the study himself drew the erroneous conclusion that the diet his lab studied replicated the Paleo diet and he is quoted as calling out celebrity Paleo personalities like Chef Pete Evans for promoting “a fad diet despite lack of scientific evidence”. I will get into details about the study and how it in no way evaluates the merits of the Paleo diet, but there’s something else I need to address first.
The responses from Paleo leaders so far have focused on a different aspect of the study: the fact that it was performed in mice. And before I go into the details of the study, what cool information it actually tells us (despite the author mistaking any relevance to Paleo) and where the media completely has it wrong, let’s address this response.
Common Critique: “But The Study was done in Mice!”
Leaders in the Paleo movement are already being quoted as dismissing the relevance of this scientific study because it was performed in mice rather than humans. The validity of the study is further being called into question due to the senior author’s affiliation with the Australian Diabetes Society (he’s the president) and insinuation that there may be influence from pharmaceutical companies or other lobby groups with statements like “who is funding this study?” .
This response is actually very typical across alternative health communities whenever a new animal study is published that doesn’t provides any sort of contrasting evidence. And it’s one that irks me every single time: Randomized Controlled Clinical Trials in large, varied human populations are not the only scientific studies of merit. Furthermore, scrunching our eyes closed while holding our hands over our ears and screaming “nah nah nah nah nah” every time a study is published that doesn’t conform to our current conclusions or beliefs hinders our ability to increase our understanding of nutrition and human health. If there was a study that showed some kind of health detriment from following a Paleo diet, that wouldn’t mean that we’d completely abandon the Paleo template—there’s a growing body of scientific literature supporting this approach to also consider. But, if the goal is to frame guidelines for foods choices to promote optimal human health, then no evidence should be discounted. Instead, we can try to understand where conflicting reports in the scientific literature originate, and identify areas that need more research.
Let me be clear: People who dismiss scientific studies that provide evidence in opposition of their views citing that they were performed in animals rather than humans prove their ignorance of science.
It’s true that not every study performed in animals directly translates to humans, however the similarity between animals (yes, even mice) and humans has long been recognized in scientific research and used to our advantage to further our knowledge of biology, physiology, biochemistry, pharmacology, genomics, and every other science related to human health. Especially when a study is mechanistic in nature, as the study in question is, great insight can come from research performed in animals.
As for conflicts of interest, a simple read of the “Conflict of Interest” section of the paper itself provides the answer: two authors receive salary support from the National Health and Medical Research Council, the Australian version of the NIH, and no other conflicts of interest were declared. I mean, is our only rebuttal to a scientific paper that opposes our views to insinuate that the authors perjured themselves? Is this how little respect we can give the academic community?
Leaders in the Paleo movement come from a variety of backgrounds and contribute varied expertise to our community. And I want to make a special mention here of the amazing Chef Pete Evans who was targeted directly by the Australian media and senior study author via the original press release, and who bares with grace similar attacks on a way-too-frequent basis. It is absolutely not my expectation that Paleo leaders who do not have scientific or medical backgrounds be equipped to perform the same scientific analysis and response to a paper that I, for example, can do with my academic background. So, shame on the media for even asking anyone without a scientific or medical background for a quote. I’ve got your back, Pete. But for those Paleo leaders who do have acronyms at the end of their names, might I suggest reading the paper and thinking about the information it provides objectively? When you do, you find a far more valid reason why this paper doesn’t support the media claims that a Paleo diet is dangerous.
The Real Rebuttal: The Diet in the Study is in Not Paleo
In the Supplemental Material provided by the study authors, the ingredients and macronutrient ratio of the low-carbohydrate high-fat diet fed to the mice in their study were shared. Here they are:
So, let’s talk about the ingredients. Whiplash warning: try not to shake your head too vigorously as you read the following three ingredients that together make up 70% of this diet that the media has dubbed “Paleo”.
The diet is 40% by mass cocoa butter, a fat particularly rich in the saturated fat palmitic acid (also richly found in cheese and grain-fed industrially-produced beef). Fun fact about palmitic acid: it is known to impair hypothalamic insulin and leptin signaling (see here) thereby increasing appetite and possibly contributing to the obesity epidemic. While very dark chocolate is considered an acceptable Paleo treat, I’m fairly confident that no one actually following a Paleo diet would think that it’s a good idea to get a half of their daily calories from dark chocolate. Interestingly though, the mice were given free access to food in this study and those fed the low-carbohydrate high-fat diet tended to eat slightly less calories than those fed the regular chow.
The diet is 20% by mass, casein. Casein is the dairy protein that has been linked to inflammation and cancer, and is one of the chief physiologic and biochemical arguments against dairy consumption on the Paleo diet. Note that whey, another protein in milk has beneficial health effects, and while dairy is not considered part of the standard Paleo diet, its overall effect on human health is far from cut and dry (see this blog post).
The diet is 10% by weight canola oil, a an oil extracted from rape seed using high heat and pressure and which is then degummed, bleached, separated and deodorized before being packaged for human consumption. Canola oil is absolutely not endorsed on any version of the Paleo diet and has been linked to various health detriments in the scientific literature (see this article by Prof. Loren Cordain and this article by Mark Sisson).
And, of course, let’s talk about the macronutrient ratio: 81% calories from fat, 6% calories from carbohydrates, and 13% calories from protein. That doesn’t look anything like Paleo. Not even the traditional diet of the Inuit follows that kind of macronutrient breakdown (see here and here). What it does look like is a ketogenic diet that has a focus on refined and inflammatory ingredients.
The Paleo Diet is NOT a Ketogenic Diet
The paper specifically evaluates the effects of a low-carbohydrate high-fat diet on weight and body composition, blood lipids, glucose tolerance, insulin secretion, and β-cell mass in mice following this diet for eight weeks. And, while the authors don’t use the word Paleo, they also never use the word “ketogenic”, although the macronutrient profile used follows ketogenic diet guidelines to a tee (81% calories from fat, 6% from carbohydrates, 13% from protein).
I have been a vocal critic of ketogenic diets (see my talk at last year’s Paleo F(x) and this blog post) and more recently of too-high-fat implementation of the Paleo diet (≥60% calories from fat) because of the collection of scientific studies proving health detriment from these approaches. I am, in fact, most of the way through writing a thorough summary of valid health concerns from high saturated fat intake and high total fat intake for an upcoming article (mainly detriment to the gut microbiome). Not that I think low-fat is the way to go, but rather than I support diets in the 30-50% calories from fat range, comparable to what we see in hunter-gatherer populations (there’s more science-based articles on this coming in the near future, so please hang tight on this one).
The media, especially in Australia where this story originated, frequently confuse ketogenic diets with Paleo, or misrepresent Paleo as an all-meat diet. I have seen multiple instances of the Australian media describing the Paleo diet to the effect of “a zero carb diet that only includes meat, vegetables, fruit and nuts”. Yeah, I know you’re shaking your head along with me (I guess this should be another whiplash warning). Last I checked, vegetables, fruits and nuts all contain carbohydrates.
Studies of hunter-gatherers show that macronutrient ratios for nearly all populations fall in these ranges:
- 19 – 35% protein
- 22 – 40% carbohydrate
- 28 – 58% fat
And, if you read any of the mainstream How To books on Paleo, you’ll find recommendations for how to compose our plates that fall within these ranges. The Paleo diet includes high quality meats, seafood, eggs, vegetables of all kinds, fruit, nuts and seeds. Plenty of carbohydrate sources in there, lots of veggies (Paleo is NOT an all-meat diet; see my post here), and while healthy fats are endorsed, the standard Paleo diet recommendations fall short of the cusp for health detriments proven in the scientific literature of 60% calories from fat.
Certainly, there are people within the Paleo community that choose to combine Paleo and ketogenic diet approaches; however in my experience, they are always clear that they are following a ketogenic version of the Paleo diet (or a Paleo version of the ketogenic diet). And, while I can’t speak for them, I don’t think any of these people would intentionally imply that the Paleo diet is inherently low-carbohydrate, high-fat or ketogenic. And they’d be right, it’s not.
So, let’s be clear: the study quoted by the media as proving that the Paleo diet causes rapid weight gain and is dangerous for diabetics does not actually evaluate the Paleo diet.
Actually, the Study Evaluates the Health Detriments of Ketogenic Diets
This study was actually a pretty neato study, and certainly rigorously designed. And, as much as it’s been misrepresented in the media and has the Paleo world in a tizzy, the results represents some pretty important nutritional science.
The strain of mice used are called New Zealand Obese mice, a strain of mice with a strong genetic predisposition to become obese, insulin resistant, leptin resistant and glucose intolerant even when fed regular mouse chow (as opposed to other rodent models of obesity or diabetes that require a high-fat diet to stimulate the health conditions) and very exaggeratedly so on high-fat chow. They were fed either the low-carbohydrate high-fat diet described above or standard mouse chow for 8 weeks after weaning and various reliable and quantifiable markers of metabolic health were made.
The mice fed the low-carbohydrate high-fat diet were heavier, with much higher percentage body fat, had higher insulin levels (fasting and fed; previous studies have no consensus on the effect of ketogenic diets on insulin levels with results from different studies varying greatly), higher fasted blood glucose (although lower fed blood glucose, which is consistent with previous studies), and no benefit to the pancreas function (contrary to two studies in rats with a chemically-induced form of diabetes).
But, it wasn’t all bad news for the low-carbohydrate high-fat diet mice: besides the lower blood glucose levels after eating, the blood lipid profile in the low-carbohydrate high-fat diet mice was improved, with lower triglycerides and higher HDL with no significant change in LDL (note that studies are all over the board in terms of how ketogenic diets affect blood lipids).
The authors conclude that their study shows greater impairment than benefit from low-carbohydrate high-fat diets and draws into question their recommendation for prediabetics.
The media portrayal is that this study is the nail in the coffin for the Paleo diet. Okay, so we can all agree the diet studied wasn’t Paleo. So, is this study the nail in the coffin for ketogenic diets? Nope, it’s not–especially given how refined and potentially inflammatory the ingredients used for the low-carbohydrate high-fat diet are. My concerns about ketogenic diets stem from reports of adverse reactions, potential negative side effects (especially to immune health, gut microbiome health, and female hormone health), and conflicting results in the scientific literature. I have never said that there are no situations in which ketogenic diets are not therapeutic, nor have I ever claimed that they can’t be beneficial in some circumstances (I just realized how many negatives this sentence has, so I’m hoping you can parse it out!). Instead, I encourage people to fully educate themselves about how much remains unknown about the longterm effects of ketogenic diet-style macronutrient intake and understand that the science is far, far, far from reaching a consensus on the purported benefits. As for this study, it simply adds to the already muddy waters.
Paleo Diets Do Improve Diabetes
Despite the media portrayal, this study can not make any statements whatsoever about the safety or efficacy of the Paleo diet for diabetics. Fortunately for us, a couple of studies have already been performed evaluating the Paleo diet for diabetics!
This study by Prof. Frassetto’s lab showed that the Paleo diet (a moderate fat intake version consistent with Prof. Cordain’s recommendations that focus more on lean meats and natural unrefined plant-derived fats like olive oil and avocado) outperform the American Diabetes Association’s dietary recommendations in terms of glucose control and lipid profiles over two weeks. This study by Prof. Lindeburg’s lab showed that the Paleo diet (again, a moderate fat version) improved glycemic control and several cardiovascular risk factors compared to the American Diabetes Association’s dietary recommendations in patients with type 2 diabetes over a 3-month trial period. This study showed that Australian aboriginals that became diabetic after adopting a Western diet and lifestyle lost weight and had improved carbohydrate and lipid metabolism after returning to a hunter-gatherer lifestyle for 7 weeks. And finally, this study done in pigs showed that the Paleo diet resulted in higher insulin sensitivity, lower inflammation and lower blood pressure compared to traditional cereal-based feed.
As an aside, Paleo diets have also been shown to promote weight loss, lower cardiovascular disease risk factors, reduce colon cancer risk, and be completely safe over study periods as long as two years.
The study that has thoroughly excited the anti-Paleo media in no way evaluates the impact of a Paleo diet since the diet used in the study does not reflect Paleo guidelines (even the small percentage of people who combine Paleo with ketogenic diets are not getting 10% of their calories from canola oil or the most inflammatory dairy protein casein!). But, just because the study is being grossly misrepresented in the media does not make it a bad study.
But here’s the thing: if the study had used a Paleo diet and had shown some kind of negative health impact, it’s still not an appropriate response to reflexively dismiss the study because it doesn’t conform to our ideals. Instead, it would have been a piece of information to weigh and consider, to ruminate on and to potentially act on. The Paleo diet is not a historical reenactment, but instead a nutrient-dense whole foods diet based on our current understanding of the nutritional merits of individual foods weighed against any possible health detriment of those same foods. It is not fixed, but instead continues to evolve as more relevant science is performed. And it’s not a firm set of rules, but rather a template from which an individual can experiment to identify what is optimal for them. And, here is my promise to you: I will never dismiss a study because it conflicts with recommendations I’ve made in the past. Instead, I will maintain an open mind and continue to do my best to communicate scientific advances to you in a balanced and objective way, avoiding hype and sensationalism, and being completely up front about the current limits of human knowledge.