Google News had a link just now to a response by Dr. Ornish to the study. Following it I ended up at:
http://www.msnbc.msn.com/id/17490143/site/newsweek/
The headline:
Why I Disagree With this Study
A new comparison of four diets—including mine—is misleading and riddled with problems.
The headline says a great deal about Dr. Ornish. He's functioning as an advocate, not as a scientist. First of all, scientists don't generally "disagree with studies." They often disagree with conclusions draw from studies. He doesn't like the conclusions that are rather obvious to draw, so he attacks the study. But he does not disagree with *all* of the study. For example, he cites the study as noting, as it does, that the difference in weight loss between the Ornish and Atkins groups was not statistically significant. Those who say that the study showed greater weight loss on Atkins than on Ornish are correct. It did. But Ornish is correct to note that the difference wasn't statistically significant.
But covered over in this confusion is the basic fact: low fat proponents have been claiming for years that low carb diets were ineffective and unsafe. Yet this has never been proven in any study, and indications from the studies that exist are that it may not be true. Including this study.
Is the study "riddled with problems". Well, that depends. If you think that the study showed that Atkins, followed according to recommendations, is better than Ornish, likewise followed according to recommendations, then, indeed, there is a problem, since the study was not designed to compare such groups.
Rather, the study strikes to a crucial point: what should public health authorities *recommend*? This study showed the effect of recommending (and educating) peope with regard to a particular diet. In that respect, recommending Atkins is shown by this study to be as good a recommendation as any of the other three, or better. And this is actually revolutionary. The conclusion of the study was not that Atkins is better than Ornish, so for Ornish to attack the study on the basis that it didn't prove that is entirely beside the point. The conclusion was that it was reasonable for physicians to recommend Atkins.
Here is a key paragraph from Ornish's rebuttal:
Quote:
The authors concluded, “Women assigned to follow the Atkins diet, which had the lowest carbohydrate intake, lost more weight and experienced more favorable overall metabolic effects at 12 months than those assigned to follow the Zone, Ornish, or LEARN diets.” This is simply not true. If you read the study carefully, you will find that the authors found that there was no significant difference in weight loss between the Atkins and Ornish or LEARN diets after one year! (There was significantly more weight loss on the Atkins diet after one year only when compared with the Zone diet.) This directly contradicts the primary conclusion of their study.
|
Note that Ornish twists the meaning of what he quoted. He contrasts, not the weight loss shown, but *statistical significance," which isn't what the authors claimed. (They found statistical significance only between Atkins and the Zone.) But that Atkins dieters *did* show greater weight loss is still interesting, and it merely emphasizes that the hypothesis that Atkins is *worse* than Ornish in this respect becomes relatively unlikely to be true. What was the "primary conclusion of the study"?
Quote:
In this study ... overweight ... women assigned to follow the Atkins diet, which had the lowest carbohydrate intake, lost more weight and experienced more favorable overall metabolic effects at 12 months than women assigned to follow the Zone, Ornish, or LEARN diets. While questions remain about the long-term effects and mechanisms, a low-carbohydrate, high-protein, high-fat may be considered a feasible alternative recommendation for weight loss."
|
That Ornish disagrees with this accurate and sober statement simply shows bad science. Taubes got fired up when he noticed the ubiquity of bad science among the low-fat advocates. This is just one more example.
The point is that conclusion the authors drew in the study is exactly appropriate, given what they were studying. They did not extend their results to claim, for example, that there were no possible harmful effects in areas not studied from, say, the high saturated fat content in a typical Atkins diet. They did not claim that participants actually followed the Ornish diet, another point that Ornish attacks. Ornish is defending his diet, not considering whether or not *recommending* the other diets is also reasonable.
What the public needs is better science, and better analysis of science, and too many diet advocates have become far more interested in being right, that is, in trying to prove that they were right, than in expanding our knowledge.
For years, the low-fat people have taken weak studies with conclusions that clearly did not follow from what was actually studied, and have reported them as proving their ideas. But if a study comes out that even can be taken as hinting that perhaps these ideas are limited or incomplete or maybe even mistaken, they attack it on far weaker grounds. They strain at a gnat and swallow a camel.
Absolutely, those of us who have come to generally trust the Atkins diet could be wrong. There could be some long-term effect that we don't know about. Avoiding carbs is, in some sense, unnatural. We know that carbs are not necessary to human nutrition (unlike what is commonly assumed by opponents of low-carb diet, as if carbs were a necessary nutrient), but quite clearly we are adapted to eat them and to find them tasty.
But, given the state of our knowledge and what we know happens when it is recommended that people follow low-carb diets, particularly Atkins, and particularly with caution being recommended about, for example, fiber and vitamins, about not taking the diet as meaning one should just eat eggs and butter and steak and forget about vegetables -- which is not what Atkins recommends! --, it is quite reasonable to recommend Atkins as a diet, and there seems little reason to strongly discourage people from doing so.
There is a population for whom Atkins works spectacularly. This won't show, necessarily, in general studies of the kind in question, for compliance reasons, among others. Perhaps there are people for whom it does *not* work, quite simply. All this is extremely difficult to study, but the fact is that the status quo is that if you went on Atkins, your doctor was quite likely to tell you that (1) it won't work and if you lose some weight, it is just water, (2) you may drop dead from a heart attack. Quite simply, these ideas, which have been repeated over and over again in consulting rooms and in the media, are wrong. And it's about time that they get tossed.
I still read, in articles critical of low carb diets, that these diets are "high in cholesterol," as if the idea that cholesterol in the diet was harmful was still a viable one. It was a mistake, the mistake has been known as suchh for a long time, but it still persists. It's bad enough that saturated fat is considered unhealthy, per se (rather than unnatural trans fats, the confounding factor that caused the early studies to conclude that saturated fat was harmful), but the cholesterol claim, I think, isn't even particularly controversial any more. It's been rejected. You can eat eggs.
Unless, of course, you are allergic to them. In the end, diet is an individual thing, one size does not fit all.
But Atkins fits a whole lot more people than the low fat crowd has been willing to admit.