carbs or protein
November 29, 2009 2:03 PM   Subscribe

Dean Ornish vs Gary Taubes. Ornish claims through clinical trials to have reversed the effects of heart disease. Taubes argues that there's evidence high carbohydrate, low fat diets increase the risk of heart disease, obesity, cancer, etc. Surely they can't both be right. Is there a unbiased opinion out there? What do the plurality of clinical trials seem to indicate?
posted by leotrotsky to Health & Fitness (17 answers total) 9 users marked this as a favorite
Something to consider: Dean Ornish has an M.D. from Baylor College of Medicine, and his study was published in the peer-reviewed The Lancet. Gary Taubes has a Masters in Journalism, and his article was published in the NYT Magazine.
posted by Houstonian at 2:32 PM on November 29, 2009 [3 favorites]

Taubes wasn't doing the studies himself, Houstonian. He was reporting on clinical studies done by medical researchers like this one and the ongoing work of James J. Kenney.

The jury really is still out on this. There are some studies that show one thing, and some studies that show the other.
posted by Sidhedevil at 3:14 PM on November 29, 2009

Also, it's going to depend on which risk factors someone has on a more granular level. High-carbohydrate, low-fat diets tend to lead to a reduction in overall cholesterol levels, but a rise in triglycerides; high-fat, low-carbohydrate diets tend to lead to a rise in overall cholesterol levels, but a reduction in C-reactive protein.
posted by Sidhedevil at 3:20 PM on November 29, 2009

What do the plurality of clinical trials indicate about what? Do you mean the impact of diet on heart disease? If so, here's one review on the subject which sounds like it does represent the conventional take on the available literature.

Personally, I take any nutrition research with a huge grain of salt. Have you noticed how there's so much room for debate on this subject, and every week a "study" has recommended something new or bucked conventional wisdom? This is a bit of a red flag about the quality of research on diet. Dietary interventions are very difficult to incorporate into prospective randomized trials. In fact, if you peruse the literature yourself, you will find that virtually no prospective clinical trial has evaluated the impact of diet on a hard outcome like heart attacks or death for reasons of impracticality. Consequently, we have come to have an over-reliance on retrospective and epidemiological research that is very susceptible to bias, confounding, and an inability to establish causality. Alternatively, you'll find smaller prospective clinical trials but they will virtually always be using "surrogate" markers of health as their measured outcome. Cardiology research in particular (for example, recent controversies over ezetimibe, torcetrapib, and multiple triglyceride lowering agents) has suffered a great deal from incorporating some rather poor reasoning into their clinical trial designs. This is not to say that the massive amount of published literature on diet is unreliable, but it needs to be interpreted with caution.
posted by drpynchon at 3:39 PM on November 29, 2009 [8 favorites]

Surely they can't both be right.

That's far from clear. Lots of people think that a rigid dietary regimen -- no matter which one -- tends to produce better health, just by forcing people to pay careful attention to what they're eating.
posted by escabeche at 3:58 PM on November 29, 2009

There are many factors at play, which is why such studies tend to be very confusing to the layman. For example, there is individual variability - one kind of diet works for one person, but not another, and incidentally this factor accounts for a huge amount of wrong conclusions drawn from research, particularly research on murines that is then heedlessly applied to humans. You'll have some study or another on a mouse strain, and that strain might be genetically crippled with horrible hormonal problems, then they feed the mouse some food X which compensates for the hormonal problems, and journos proclaim - "if you want to be healthy, eat X!!". Of course, unless you have the same genetic makeup as the crippled mouse, you'll derive zero benefit from such a diet, but may in fact be harmed.

There are many other factors at play too. For example calories - which affect diets differentially. Bottom line, there are too many factors to disentangle, and a lot of the studies are very sloppy indeed. I wouldn't jump to change my diet with every study that comes out.
posted by VikingSword at 4:20 PM on November 29, 2009 [1 favorite]

Maybe neither is right.

Maybe it's calorie restriction that makes the difference.

Maybe it's whole foods in general and reducing it to numbers is idiotic, which is Michael Pollan's opinion in In Defense of Food.

Either way, you can exclude several foods that all demonize: white bread, sweets, soda, candy, processed meat, processed grains. Ornish's high carb does not include lollipops!

And there is a diet that kind of follows both (low saturated fat, low carb) , the Paleo Diet a la Loren Cordain, who is also a distinguished scientist: lean meat (turkey, wild boar, venison, bison), fish, sea vegetables, tubers, leafy greens, roots, berries, which is probably what humans ate for most of their evolution.
posted by melissam at 4:46 PM on November 29, 2009 [1 favorite]

I would also note that it's not carbs vs. protein. Taubes is pro high-fat, not high protein. Excessive amounts of protein can cause problems. People in the very far north who eat only meat tend to be otherwise healthy (disputing Ornish's claim that meat = heart disease and cancer) except for the low bone density.
posted by melissam at 4:49 PM on November 29, 2009

The clinical trial that Dean Ornish ran is very uninteresting, I've uploaded the full text here.

The experimental group of 28 people
-stopped smoking
-started exercising
-received stress management training
-ate a low fat vegetarian diet

Lots of confounding factors, was it the smoking, lifestyle or dietary habits?
Also note, the control group had much higher HDL which may or may not be a better predictor of heart disease mortality than coronary stenosis.

Contrary to what a couple people have commented, there are some good trials looking at this issue with sample sizes far greater than Ornish has worked with.
-Womens Health Initiative Dietary Modification Trial
-Lyon Diet Heart Study
-Multiple Risk Factor Intervention Trial

While there is probably no unbiased opinion out there, I would recommend ignoring what Ornish has to say.

The idea that keeping your dietary fat under 10% is healthy or beneficial seems circumspect to say the least, Dr. B G has quite a bit of good information on the subject in these ten blog posts
posted by zentrification at 4:51 PM on November 29, 2009 [1 favorite]


Any references to back up the idea that excessive amounts of protein can cause problems?

The low bone density link to protein intake is very weak at best, much more likely to be due to vitamin D fluctuation among northern people.
posted by zentrification at 4:55 PM on November 29, 2009

Here is some spitballing on the topic. I tend to agree in general with drpynchon. As someone that has followed trends in health and fitness research pretty closely for the last 5+ years, it is pretty clear that there is no consensus and no obvious plan that works for everyone. But having done all that reading, here are my thoughts:
The problem with the Ornish diet may not be that the diet per se leads to problems, after all, it is a very puritanical regimen and I suspect that those that actually manage to truly adhere to it simply eat very little period and probably do have pretty good heart health. The problem with the Ornish diet is probably that it is incredibly hard, especially in this society, to truly adhere to the tenets of the diet. Corn syrup is not a part of the Ornish diet, low-fat cookies play little role in that diet, and yet I know from personal experience half-assing the Ornish diet (or any low-fat diet) probably means eating a lot of sugar/"white" carbs instead of fat, which is no good to anyone.
I think the real question to ask is, how likely are people to adhere to the diet and how are they going to "cheat"? What is the heart health of the average person that claims to be following such a regimen, at least in spirit (low carb vs low fat) if not to the letter? If fat is the ultimate bad thing, you may be tempted to cheat by eating more sugar. If carbs are the ultimate bad thing, you may cheat by eating peanut butter straight from the jar instead of a cookie. Which of those cheats adds up to a worse outcome?
Low-fat adds up to obesity and diabetes not because it is bad by itself, but because the average person's interpretation of the diet has an incredibly bad profile. Whether the average person's interpretation of low-carb has as bad a profile is less clear, but it is much harder for big ag to manufacture cheap fake food that tastes good and is truly low-carb, so I think we get a small leg-up from that if nothing else.
posted by ch1x0r at 6:09 PM on November 29, 2009

I'm not a doctor, nor do I pretend to be one. I'm following the Taubes-esque suggestions (i.e. very low carb, pretty high fat), though I've had a few instances of falling off the wagon. I've lost around 18 pounds in three months, which is perfectly acceptable to me.

My wife is starting Eat To Live, the Joel Fuhrman regime (basically vegan super-low-fat). This can also work if followed strictly; a blogger/coder I know has lost 75 pounds since starting this regime.

Yes, we're doing the complete opposite programs. Our home is a short-order cook bonanza.

Honestly, I prefer Taubes' review of the studies, purely on the basis that he's not he doesn't have any particular axe to grind (apart perhaps from the "ornery controversial science journalist" one). Doctors who "prescribe" a particular diet are likely to find all the supporting research and not spend too much time on research that doesn't back them up.

If Taubes is any indication, there's a lot of research out there that says "this study [at best] shows no efficacy for the conventional nutritional guidelines. However, we still think they should be followed, and we humbly request more funding so we can try to support them next time."
posted by lowlife at 6:21 PM on November 29, 2009

Seconding melissam that it's more useful to focus on what *everyone* agrees is bad for you. I have read a lot of these books -- Taubes, the China Study, one of the calorie restriction books, and a few others -- and in my case it's mainly a procrastination device. It's more fun to discuss the controversy than to make obvious-but-difficult changes in my diet :)

My brief list of what everyone seems to think is bad for you -- or at least many do, and I haven't seen credible research that recommends any of these things: sugar, dairy for adults, starches, excessive salt, any high-density processed carbs. would be curious if anyone has corrections/additions.
posted by pete_22 at 4:21 AM on November 30, 2009

If you want to read more arguments for the Ornish side (though they all have different focuses and don't agree on everything with each other), check out Dr Fuhrman (mentioned by lowlife, but I should mention that he is not super low fat, he is no-oil, but fat from nuts and avocado's are encouraged), Dr McDougall, Dr Esselstyn. I'm always impressed by Mc Dougall's stars, even though I don't actually think his diet is the ideal diet (too much grains). I do think it shows that even with a not-perfect diet, you can achieve astonishing health improvements and I think the most important thing is cutting out junk food and eating more vegetables.
posted by davar at 5:11 AM on November 30, 2009

Another theory is that excess consumption omega-6 fatty acids is a major factor in heart disease. Both diets substantially reduce their consumption, Ornish through prescribing very little fat in general and Taube through proscribing high-omega 6 vegetable oils.

I'd add more consuming more than 10% of fat in the form of omega-6 fatty acids to pete_22's list of no nos.
posted by melissam at 8:36 AM on November 30, 2009

Taubes' book, "Good Calories, Bad Calories," lays out what I thought was an irrefutable case. The high-carb, low-fat diet has become a religion among nutritionists and a medical community that has based its recommendations on bad science.

I'm not a conspiracy theorist, but if you read "Good Calories, Bad Calories," you'll see some egregious errors that have been foisted upon an unsuspecting public.

I would also like to recommend the work of Brian Peskin. The guy comes off as a used car salesman sometimes, but his conclusions are all based on the science from medical textbooks, not flawed studies. His book, "The Hidden Story of Cancer" is worth reading, if for nothing else the section on how most studies released to the public nowadays are essentially worthless.

Great discussion. I find this topic fascinating and would like to hear more.
posted by billybee at 4:32 PM on December 5, 2009 [1 favorite]

I found this Mefi thread from 2004 interesting reading.
posted by cashman at 12:46 PM on March 10, 2010

« Older Out of the Flickr group loop   |   Sending mail...psyche! Newer »
This thread is closed to new comments.