David Katz, M.D.Director, Yale Prevention Research Center
One recent study demonstrated that the Mediterranean diet is better than a low-fat diet at improving a wide array of cardiac risk factors. Another showed that the Portfolio Diet is superior to a low-fat diet at lowering serum lipids. But then again, very low-fat, plant-based diets have been shown to cause regression of atherosclerotic plaque, to prevent cardiac events and to modify gene expression in a manner that should lower cancer risk.
So, do we know what diet is "best"?
No, we do not. And anyone who claims we know decisively what specific diet is best for health is either misguided, selling something or both.
But what in my opinion we do know decisively is the basic theme of "best" eating -- for our health and the health of the planet. And we know a number of variations on that theme as well, opening up opportunities for the indulgence of personal preference -- or in other words, for loving food that loves us back.
For those inclined to eat only plants, a consistent body of evidence supports health benefits of a vegetarian or vegan diet. Eating only plant foods does not guarantee a healthful, balanced diet. Sugar -- among the more reviled of modern dietary excesses -- comes from plants, after all. So the potential health benefits of veganism require that it be well practiced. I recommend the work of Davis and Melina to those who need help getting there from here. A vegan diet, when well-devised, offers benefits to one's personal health and the health of the planet. Judicious supplementation will be required; intake of Vitamin B12 will be negligible, and Vitamin D levels will depend either on supplementation or routine sun exposure.
What may surprise the staunchest proponents of a vegan diet -- such as Dr. T. Colin Campbell and his many followers -- is the lack of evidence that veganism is better for our health than well-practiced omnivorousness. There are three important considerations related to this point.
First, it stands to reason that human physiology is well-adapted to certain animal foods. While paleoanthropologists may continue to debate the exact degree to which our Stone Age ancestors were gatherers versus hunters, there is consensus they were both. Animal foods -- meat, fish, eggs -- are a native part of the human diet.
Second, dietary details matter in both the animal and plant kingdoms. The category of animal foods is home to everything from wild salmon, to a wide array of salamis. Some choices are likely to promote health, others to threaten it. A one-size-fits-all assessment of the health effects of eating "animal foods" is almost certain to be erroneous.
And third, what we have here is an absence of evidence, rather than evidence of present or absent health effects. Consider the study it would require to prove that a vegan diet, or optimized omnivorous diet, were superior to the other: thousands of people, randomized to one of several dietary patterns, and then followed for decades. Even if there were people willing to enroll in such a trial (doubtful!), costs would be prohibitive. For now, we may simply acknowledge that no such trial has been conducted.
So we are left with evidence of clear health benefits from some mixed diets, without any means to compare them directly to the benefits attached to plant-only diets. The Mediterranean diet -- more a class of diets than one per se -- stands out. Studies suggest benefits across a spectrum of health outcomes, from weight to cancer, cardiovascular disease to life expectancy. The benefits are not discernibly less, and could even be greater, than those chronicled for vegetarianism.
A balanced, mixed diet of plant and animal foods was used in the Diabetes Prevention Program, and produced a 58 percent reduction in the incidence of Type 2 diabetes in high-risk adults. In the various DASH diet studies, a diet including low- and non-fat dairy was more effective than a plant-based diet without dairy, at lowering blood pressure.
That said, the widespread support for the DASH diet may owe much to the rich getting richer. The DASH trials were supported by the National Heart, Lung and Blood Institute, and therefore the promising outcomes of these studies have the might of the NIH behind them. There are legitimate concerns about potential adverse effects of dairy intake that DASH-related literature tends to ignore. Nor do we have head-to-head comparisons of a DASH-style diet to a vegan diet, or Mediterranean diet, to show which produces the best long-term health effects. DASH is a good diet, to be sure -- and clearly far better than the typical American diet. But contentions that it is the "best" diet derive from good PR, not good data.
The Portfolio Diet, devised by my friend and colleague Dr. David Jenkins at the University of Toronto, was not originally intended to be a serious contender for best diet laurels. Dr. Jenkins and his team designed and tested this diet to make a point: food is powerful medicine. Specifically, they wanted to show that a well-chosen "portfolio" of foods for lipid-lowering could do the job as well as statin drugs. With a diet incorporating a hefty dose of soluble fiber and plant sterols, they showed exactly that.
But Dr. Jenkins is quick to note that 75g of soluble fiber per day is apt to be unpalatable for most people, so the Portfolio Diet was more about proof of principle, rather than guide to practice. But gentler variations on the theme of "Portfolio" eating, and the new study showing its ostensible superiority to a low-fat diet, have it in the mix.
Other work by Dr. Jenkins' team does a nice job of backing the "low-carb" diets into the appropriate corner. In general, such diets are not admissible participants in the best-diet pageant.
A truly low-carb diet is low in vegetables, fruits, whole grains, and to a lesser extent, beans and legumes as well. Proponents are quick to protest that their low-carb diet does not limit vegetables, or beans. Well then, it's not a low-carb diet, but a selective-carb diet. And every "good" diet, every reasonable contender for best-diet laurels, is already a selective-carb diet, while also preserving the good sense to include fruits. So again, truly low-carb diets are not a consideration.
But one variation on the theme of low-carb eating is a mostly plant-based diet, relatively high in protein. Dr. Jenkins and his team termed this the "Eco-Atkins" diet, highlighting a way to eat more protein without imposing the environmental harms attached to eating more meat. Perhaps Eco-Atkins deserves a spot in the diet pageant, but frankly, that's debatable. There is research to show that given a choice between achieving a low dietary glycemic load by eating less carbohydrate, or by eating better carbohydrates, eating better carbohydrate is better.
I hasten to note, since low-carb advocates will bring it up if I don't, that low-carb diets perform at least as well, and sometimes better, than the competition for weight loss over the short term. While this tends to be veiled in mystique, propagated by the likes of Gary Taubes, implying that the sources of calories matter, while calories do not. I believe that the truth is far simpler and far more compatible with the laws of thermodynamics. Carbohydrate is the largest nutrient class, and greatest source of calories, for all omnivores and herbivores alike. When omnivores "cut carbs," they tend to cut calories -- a lot. I have closely examined the calorie intake levels in some of the studies pitting the Atkins' diet against others and Atkins' was the most calorie-restricted.
More germane to today's topic, however, are the many potential, adverse health effects of an animal-food-based, low-carb diet, and the lack of any evidence linking such a diet to health benefits over time.
We might do well to consider sustainability as a feature of any diet competing to be "best." Not to put too fine a point on it, but in our voracious multitudes -- now roughly 7 billion of us -- we can quite literally destroy the planet by the injudicious indulgence of our appetites -- in particular by placing an emphasis on meat. We are fast becoming to the whole planet what locusts are to a field of wheat. And as ostensibly sentient creatures with options, we should find that had to justify.
There is an ethical case against eating animals as well; one I hear about routinely from a friend and colleague who heads up an animal ethics group at Yale. But the arguments in that area are more subtle than first meets the eye and warrant dedicated attention -- so I will return to that theme in my next post and just leave it dangling for the moment.
Limiting our focus to the direct effects of diet on our own health, we have a strong case for eating mostly plants. Adding in some concern for the fate of the planet, the case only strengthens.
The theme of healthful eating consistently emphasizes the same foods: vegetables, fruits, beans, legumes, nuts, seeds and whole grains. Some variations include fish and seafood, others do not. Some include low- and non-fat dairy, others do not. Some include lean meats, others do not. All banish highly processed foods delivering concentrated doses of refined starch, sugar, trans fat, certain saturated fats and/or salt to the realm of rare indulgence.
The contest to determine "the best" diet has simply not been run, possibly can't be and probably never will be. The theme of healthful eating, however, is very well established. Adopt your preferred variation on that theme, but stick to the theme and let the food you love ... love you back.
Dr. David L. Katz; www.davidkatzmd.com
Follow David Katz, M.D. on Twitter: www.twitter.com/DrDavidKatz