The recent attention to dietary cholesterol, and by extension eggs, has stirred up the customary responses in a world of dietary divisiveness, if not outright dogma. There are rumblings of protest among my many vegan and vegetarian friends and colleagues, who are convinced the Dietary Guidelines Advisory Committee ran amok, whether for reasons of well-intentioned error or something more nefarious. There are comparable rumblings of celebratory validation among advocates of Paleo diets, convinced this committee finally recognized a vintage truth others had neglected. Enough of both protests find their way to my inbox, that it has been a week of unusual seismic activity for me.
I prefer solid ground, frankly, if only to avoid motion sickness. So the fervor for the poles invites a question in the middle: can we say what diet is best for health?
My devotion to the question, and to the extent humanly possible, an unbiased answer, has involved efforts both lesser, and greater. The greater effort, a textbook running to over 750 pages and nearly 10,000 citations, was as excruciating as one would presume.
On the basis of such self-flagellation in the service of truth, I can say with considerable, if not perfect, confidence that the answer is: absolutely yes, and definitely no.
Yes, absolutely, we can say what diet is best for health if by "diet" we mean a basic pattern, or general theme. Overwhelmingly, evidence derived from a stunning array of populations, settings, and methodologies supports a diet of real foods close to nature, with an emphasis on plants. Michael Pollan famously encapsulated this as "food, not too much, mostly plants." Frank Hu and I landed on, "wholesome foods, in sensible combinations," which might allow for the observed variations in animal food intake more facilely, and which deemphasizes quantity because getting quality right seems to be the best way of addressing that. Still, we wound up rather well aligned with Pollan, arriving there via totally independent work, mine published in Annual Review of Public Health, and Frank's in the Lancet.
But no, we most definitely cannot say which diet is best for health if by "diet" we mean a very specific, prescriptive pattern of foods. Why not? The relevant studies have not been done, and are exceedingly unlikely. Imagine the trial necessary to say decisively, for instance, that an optimally constructed vegan diet is superior to an optimally constructed Paleo diet, or vice versa. Since "best diet" should mean superior effects on both years in life and life in years, this would need to be a study of not only vitality, but also longevity -- meaning it would have to last an entire lifespan. We would, presumably, need to randomize neonates at weaning from their mothers' milk (by which time our study would already be slightly contaminated by differing exposures to nutrients via that milk, and in utero, based on variations in the maternal diets -- but we could soldier on and not make perfect the enemy of good) to those two distinct dietary patterns and then follow thousands (or tens of thousands) of them for decades and decades to see if, all other things being equal, one of the diets produced better outcomes than the other.
Let's not hold our breath until that one comes out in the New England Journal of Medicine.
In the mean time, like it or not, we are all obligated to make do with lesser varieties of evidence than that -- and a bit of sense.
While many do, no one can say that a Paleo diet is best for health on the basis of truly robust evidence. But no one can say it isn't, either (I ask my vegan friends to hold their ethically rescued horses as I reason through this). Every other species on the planet does best when eating its native diet. We take it for granted that lions should eat meat and giant pandas should eat bamboo, for no more evidence-based reason than: that's what they eat. That's what they are adapted to eat. There are, to my knowledge, no randomized trials to determine who gets what when it's feeding time at the zoo.
Homo sapiens are constitutional omnivores. That is not controversial, representing the consensus view of every qualified biologist, physiologist, and anthropologist to opine on the matter. Some go further, such as E.O. Wilson in The Social Conquest of Earth, to note, for instance, human adaptations not only to meat -- but to cooked meat. The hunt, and the tribal fire around which its success was celebrated, are of common, ancient lineage.
The former -- eating meat -- more so, as it goes back even further than our 6 million year old evolutionary branch point from our closest cousins, the chimpanzees. They, too, are omnivores, although I trust their occasional penchant for grisly cannibalism is something we can comfortably renounce across the full spectrum of dietary inclinations.
What of specific evidence about health effects of the Paleo diet? My impression, based on extensive correspondence with colleagues wearing different stripes, is that everyone thinks there is much more out there than there really is.
I searched PubMed for peer-reviewed studies, using the terms "Paleo diet" or "Paleolithic diet," and no other restrictions to generate the broadest possible yield. The first search yielded 124 results; the second, 108. To give you some sense of how shockingly sparse this makes the relevant evidence, I searched for "extrapulmonary tuberculosis," and that yielded 2,893 citations. The term "ankylosing spondylitis" yielded 14,718.
Amazing, right? All this debate and diatribe -- but it's all smoke from that age-old tribal fire, and almost no light. (For those wondering, the search term "vegan diet" retrieved 3,260 papers, so the Paleo researchers have a lot of catching up to do -- whether in the service of showing benefits or harms.)
My vegan colleagues routinely direct me to at least one study showing metabolic harms from a Paleo diet assignment. They may be right, but this is slim evidence. The paper was the work of an undergraduate student; was not published in the peer-reviewed literature; and is deficient in a number of details required to make sense of it. I can assure you, I looked for confirmatory studies in the peer-reviewed literature, and all I could find were a few publications suggesting benefits, not harms, of the Paleo diet related to both cancer and cardiovascular risk.
This really shouldn't surprise anyone. What health changes a diet engenders depend on what the diet is changing. Any reasonable rendition of a Paleo diet is a vast improvement over the typical American diet which not only already includes meat, but often highly processed meat from inappropriately fed, captive animals. A true "Paleo" diet and a good vegan diet are more like one another than either is like the diet that currently prevails in America, and much of the modern world.
But I am not writing to advocate for the Paleo diet. I am merely trying for dialogue by a bit more light of the fire, and a whole lot less smoke.
There are two fundamental problems with the Paleo diet, regardless of the intervention trial evidence, and no matter our adaptations to it. We may term them: (1) the pastrami problem; and (2) the population problem.
(1) The pastrami problem is that many people fly the "Paleo diet" banner as an excuse to eat -- well, pastrami. And cheese. They merely eschew the bread. But of course, there was no Paleolithic pastrami (to say nothing of cheese), nor anything like it. The flesh of wild animals is compositionally nothing like that of fattened, domesticated animals used in the production of pastrami, sausage, bacon, and so on. The conflation of hot dogs for mammoth or antelope would be bad enough, but the situation is clearly far more thematically adulterated than that. I had correspondence just this week from a Paleo dieter, delighted with the news about eggs, and referring to a "Paleo" breakfast of eggs, bacon, and all fried up in butter. Paleolithic butter? I don't think so.
So while there are devoted experts who practice a Paleo diet with real fidelity, and might be benefiting from it, their far more numerous disciples seem inclined to stick a feather in their butter-fried Chorizo and call it Paleolithic.
(2) The population problem is far more serious. I would have been inclined to address this anyway, but I note that one comment left in response to a recent, prior column scoffed at the notion that human population density is relevant to dietary choice -- since all of the humans on earth could stand on a land mass the size of Texas. Gee, that sounds like fun. Anyway...
There are over 7 billion of us Homo sapiens on the planet. The best estimate I could find for average land area required for sustainable hunter-gatherer subsistence is 32 square kilometers per clan of 100 humans.
The surface area of the earth is 510 million square kilometers. But, of course, we live on a watery planet -- yet are terrestrial hunter-gatherers for the most part. So, the relevant surface area -- the land -- is just 149 million square kilometers.
Assuming that the current population of over 7 billion of us needed approximately that same 32 square kilometers per tribe of 100, what land mass would be required for us all to make a living as traditional hunter gatherers? Roughly 2 billion 240 million square kilometers, or 15 times the entire land surface of earth.
That means even if we hunted and gathered our way across every square inch of the Sahara, Antarctica, and the Himalayas and actually found food there -- only 6.7 percent of us would avoid starvation. My interlocutor from the prior column -- the "let's all stand in Texas" guy -- is welcome to challenge this, but I need to see the math.
This, then, is where modern humans should come together in discussing the Paleo diet. The relevant question is not, really, whether an optimal Paleo diet is better or worse for human health than an optimal vegan diet, or Mediterranean diet. The relevant question is: which is really an option for a growing population on a shrinking planet?
Mammoth, and virtually all of its contemporary plants and animals, are irrevocably off the menu for reasons that trump debate; they are extinct. That has been the inevitable result of human population growth and dispersion, and the process has only accelerated into the modern era. There is a line we cannot cross where the mass extinction of plants and animals can only portend our own.
Paleo diet enthusiasts could potentially live alongside vegetarians and others on the survivable side of the line if we embraced our commonalities. A plant-based diet is about humans eating plants. A true Paleo diet is about humans eating plants, and eating free-ranging animals that in turn ate plants. Neither allows for factory farmed meat, abuse and incarceration of animals, or adulterations of "native" feed. To the extent that Paleo is practicable for a population of 7 billion, it replaces the flesh of often cruelly treated, unnaturally fed animals with the far more nutritious flesh of humanely treated, natively fed, free ranging animals -- and takes cruelty, hormones, and antibiotics off the menu into the bargain. Both diets, and everything in between, renounce the hyper-processed, glow-in-the-dark junk that currently predominates.
In the end, though, if we could re-leash the dogmas of perennial and unproductive dietary warfare, we might collectively turn our attention to the practicalities: variations on the theme of optimal eating for human and planetary health alike. Dietary practices that produce mass extinctions, or cause 93.3 percent of us all to starve, make poor contestants in this never-ending pageant.
Paleo enthusiasts can legitimately eat their bison, or venison, or antelope, although not their butter-fried sausage, and might -- for all we know -- derive real health benefit from doing so. But we don't live in an age of isolated, scattered human tribes; we live amidst a modern population of 7 billion, and growing fast. A mostly plant-based diet is the common interest of any of us who want to stick around.
To put it bluntly, Dorothy, we are not in the Stone Age anymore. As the litany of extinctions indelibly conveys, we just can't have our mammoth, and eat it, too.
David L. Katz, MD, MPH, FACPM, FACP does recommend we all stand together, just not necessarily in Texas.
Director, Yale University Prevention Research Center; Griffin Hospital
President, American College of Lifestyle Medicine
Editor-in-Chief, Childhood Obesity
Author: Disease Proof