Saturday, March 05, 2011

The federal government's Dietary Guidelines discuss plant-based and vegetarian diets

Although USDA is frequently described as being in the pocket of the meat and dairy industries, the new Dietary Guidelines for Americans, 2010, include a highly favorable discussion of plant-based and vegetarian diets.

The new edition has a chapter on eating patterns, focusing on real foods and not just nutrients.  This chapter on eating patterns provides a nice counterpoint to the reductionism -- what Michael Pollan calls "nutritionism" -- of scientific discussion of diet and health.  Some of the guidelines' healthy eating patterns include meat, while others do not.  For example, the USDA Food Patterns and the DASH diet each include moderate amounts of meat and plenty of lowfat dairy.  At the same time, the guidelines explain clearly that meat is not essential, and near-vegetarian and vegetarian diets are adequate and even "have been associated with improved health outcomes."

In reviewing the scientific evidence on vegetarian diets, the guidelines say:
In prospective studies of adults, compared to non-vegetarian eating patterns, vegetarian-style eating patterns have been associated with improved health outcomes—lower levels of obesity, a reduced risk of cardiovascular disease, and lower total mortality. Several clinical trials have documented that vegetarian eating patterns lower blood pressure. On average, vegetarians consume a lower proportion of calories from fat (particularly saturated fatty acids); fewer overall calories; and more fiber, potassium, and vitamin C than do non-vegetarians. Vegetarians generally have a lower body mass index. These characteristics and other lifestyle factors associated with a vegetarian diet may contribute to the positive health outcomes that have been identified among vegetarians.
The new guidelines have adapted to many of the scientific criticisms of earlier editions.  For example, Harvard scientist Walter Willett has long argued that replacing saturated fats with unsaturated fats saves lives, and that refined grains and simple carbohydrates are no healthier than unsaturated fats.  The new Dietary Guidelines seem to me in complete agreement.

However, not everybody likes the new guidelines.  The most vociferous low-carb advocates say it is wrong to encourage reduction of saturated fats.  Just this week on the Civil Eats blog, which I usually like, Kristin Wartman tore into the conventional wisdom on saturated fats: "The notion that saturated fats are detrimental to our health is deeply embedded in our Zeitgeist—but shockingly, the opposite just might be true."  I generally agree with the way low-carb advocates criticize sugar and simple carbohydrates, but this corollary view of saturated fats worries me.  Wartman's view seems threatening to the guidelines' favorable perspective on near-vegetarian and vegetarian diets, which are typically lower in saturated fats.

Because the saturated fat corollary to the low-carb criticism of the Dietary Guidelines is widely believed, I need to spend a couple more paragraphs on why a low-saturated-fat diet might be okay for your health.

First, there is the scientific evidence that lowering saturated fat reduces risk of heart disease.  If you are primed to disbelieve any reductionist arguments, you can ignore this paragraph.  But what strikes me as a lay reader of the scientific literature is how systematic and transparent the Dietary Guidelines' evidence reviews now are.  For saturated fats, you can see exactly what protocol was used to select studies for review, and what criteria were used to evaluate them.  You can confirm that the external Dietary Guidelines Advisory Committee report correctly reflected the evidence review, and that the actual Dietary Guidelines correctly reflected the Advisory Committee report.  If you read any of the main low-carb advocates, such as Gary Taubes, see if you can discern in a similar way what his or her implicit protocol was for selecting studies to consider.  If you find fault with the Dietary Guidelines' evidence review, please send a link in the comments to any systematic and transparent evidence review that you find superior.

Second, there are the foods themselves that contain saturated fats.  Part of the concern about "nutritionism" is that USDA lacks the courage to criticize specific foods.  The guidelines may criticize saturated fats, but they won't mention specific foods, the critics say.  I have my own slightly gentler expectation for the federal government's frankness in reporting.  If the guidelines use scientific jargon for a food component such as "saturated fats," I think they should have the gumption to say what foods typically contain that component.  This is exactly what the new edition does very well.  Here is the new report's pie chart to accompany the discussion of saturated fat (click if needed for a bigger image).



So, the main implication of the Dietary Guidelines' continued criticism of saturated fat is to recommend reducing some combination of: regular cheese, pizza, grain-based desserts, dairy desserts, chicken and mixed dishes, sausage, franks, bacon, ribs, burgers, and so forth.  As long as we all agree that occasional treats in reasonable portions are harmless, this advice sounds just fine to me.  I don't see why anybody would complain about the Dietary Guidelines' advice about saturated fat.

From exactly the opposite direction as the low-carb critics, the new guidelines have also been attacked by the Physicians Committee for Responsible Medicine, a health research and advocacy organization that is favorable to vegetarian and vegan diets.  PCRM is suing USDA for not criticizing meat and dairy strongly enough in the new Dietary Guidelines.  In a recent telephone interview, I asked PCRM president Neal Barnard whether it is wise to rely on the courts to decide dietary guidance, rather than the external scientific review process of the Dietary Guidelines Advisory Committee.  He responded that even the Advisory Committee had understated the risks of animal foods: "We’re not asking them to deviate from being science based.”

As the discussion continues, I will be interested to hear what people think of (a) the saturated fat corollary to the low-carb critique, (b) the PCRM argument that meat should have been more fiercely criticized, and (c) the Dietary Guidelines themselves.

12 comments:

Nate said...

"If you find fault with the Dietary Guidelines' evidence review, please send a link in the comments to any systematic and transparent evidence review that you find superior."

Here you go:

http://wholehealthsource.blogspot.com/2010/12/diet-heart-controlled-trials-new.html

Dr. Guyenet's blog post discusses and links to an excellent British Journal of Nutrition meta-analysis of controlled trials measuring the cardiovascular effects of replacing saturated fats with seed oils.

Both the blog post and the paper are worth reading.

Anonymous said...

It would be wise to update your information before criticizing someone's highly researched article.
In fact, it is now widely accepted by medical professionals that total serum cholesterol is not a valid predictor of cardiovascular disease or risk. Medical experts in cardiovascular health are quick to point out that HDL is far more important than LDL.

usfoodpolicy said...

Thanks for the comments!

The article mentioned by Nate has to do with recommendations specifically to increase n-6 PUFA (poly-unsaturated fatty acids). I have no view on this issue, and don't see how it reflects on the evidence review under discussion. Does this literature review on the distinction between n-3 and n-6 policy-unsaturated fatty acids cast doubt on the evidence review about saturated fatty acids? If so, what is a better -- more transparent and complete -- source of evidence on saturated fatty acids?

I am not sure which highly researched article Anonymous is concerned with: perhaps Taubes (2002) or Wartman (I'll let readers decide for themselves if it seems "highly researched")? I have no particular view on total serum cholesterol as a predictor for heart disease. I don't see how that question should affect a layperson's judgment of the evidence review for the Dietary Guidelines, which seemed to address all sorts of cholesterol markers.

Nate said...

Hi Professor Wilde,

Thank you for your reply. N-6 PUFAs are relevant because the USDA recommends substituting saturated fats with PUFAs and MUFAs and the overwhelming majority of PUFAs in the American diet are n-6 PUFAs. The Ramsden et al. review found that n-6 PUFA interventions increase CHD risk and increase risk of death from all causes.

It did find that replacing saturated fat AND trans fat with n-3/n-6 PUFAs reduces CHD risk, but trans fat is a major confounding factor.

I'll have to post about MUFAs another time, but note that the only meta-analysis in the USDA evidence summary found an increase in coronary events when saturated fat was replaced by MUFA.

"Does this literature review on the distinction between n-3 and n-6 policy-unsaturated fatty acids cast doubt on the evidence review about saturated fatty acids? If so, what is a better -- more transparent and complete -- source of evidence on saturated fatty acids?"

Yes and I'd point to last year's Krauss et al. meta-analysis as a definitive source on saturated fatty acids and cardiovascular disease. His group found no association between saturated fat intake and CHD or CVD. Krauss is one of the most prominent lipid researchers in the world and has completely changed his views on saturated fat.

The review can be found here: http://www.ajcn.org/content/early/2010/01/13/ajcn.2009.27725.full.pdf+html

usfoodpolicy said...

Hi Nate. That is indeed exactly what I was looking for, a systematic literature review about saturated fat in particular. I will read that, but also I will read the flurry of criticism this article seems to have generated in the pages of the AJCN. The upshot for me as a layperson -- and I know this may not satisfy you -- is that I will probably wait until the scientific community has a chance to digest this debate before changing my own eating habits.

Meanwhile, just to gauge how close or far apart we are, do you think near-vegetarian diets are unhealthy because of a shortage of saturated fats? This seems to be your point, but flips the statement direction and burden of proof.

Thanks for the comments.

usfoodpolicy said...

Parke,

The study you have referred me to is outdated. It is widely known that total serum cholesterol is not a valid predictor of cardiovascular disease or risk. If you look carefully at what the study has found you will see that while the diet they fed participants (which was lower in saturated fats than the Standard American Diet) reduced LDL cholesterol -- it also reduced HDL cholesterol and raised triglycerides. It reads: “Compared to the average American diet, the Step I and Step II diets also decreased HDL-C and raised TG levels in the blood.”

Experts in cardiovascular health now know that HDL is far more important than LDL. HDL is protective as an anti-inflammatory. In fact, total serum cholesterol numbers are quickly being ruled as inaccurate markers for predicting disease. HDL is such a potent negative risk factor it may negate the presence of high LDL.

The study also says, “Furthermore, these authors showed that subjects who were insulin resistant responded less favorably to the Step II diet than did those with normal insulin sensitivity.” This is another important factor since the reduction in saturated fats appears to be “less favorable” to those with an increased risk for diabetes.

[continued in next comment due to character limit in Blogger ...]

usfoodpolicy said...

[... continued from previous comment ...]

As for your blog response, when discussing the benefits of vegetarian diets versus the Standard American Diet there are many factors to consider in getting a clear picture of why vegetarians may in fact be healthier. One, vegetarians tend to be more health-conscious overall. Two, compared to the SAD vegetarian diets often contain less processed foods, fast foods, and sugary foods.

Nowhere in my article do I criticize vegetarian diets -- which in fact can quite easily contain high amounts of saturated fats in the form of butter, cheese, milk, milk products, and eggs. In fact, I believe personally that a plant-based diet that includes good, clean sources of these foods is one of the healthiest diets we can eat. Some people benefit from a decent amount of pasture-based, grass-fed meat products but I believe you can thrive on a vegetarian diet, like the one described above.

Also, I am concerned that studies often lump saturated fats in with trans-fats, as has been the case in the past. This was the case with coconut oil, which was deemed an evil fat after studies were done using hydrogenated versions of it. More recent studies, using unprocessed coconut oil have found it to be a very healthy saturated fat.

My basic premise, as I mention in my article, is to eat foods in their real, whole state. This eliminates several foods that you might not realize: low-fat, or fat-free dairy products, hydrogenated and processed oils or fats, and commercially-raised grain fed meat, poultry, eggs, dairy products. Fortunately, there are recent studies that take into account these important differentiations. Grass-fed milk has been found to have a healthier fat profile than its conventional counterpart, as have grass-fed beef and pasture-raised eggs.

And I have to take issue with this statement: “The most vociferous low-carb advocates say it is wrong to encourage reduction of saturated fats. Just this week on the Civil Eats blog, which I usually like, Kristin Wartman tore into the conventional wisdom on saturated fats.” Despite your implication here, I am not a low-carb advocate. Rather, I am concerned that the long-time recommendations to reduce saturated fat (and all fats for that matter) from the American diet has resulted in an increase in carbohydrate consumption which has in fact caused more harm than good.

Thank you for your interest,

Kristin Wartman

usfoodpolicy said...

Hi Kristin,

Thanks for sending your comment and rebuttal!

The point of difference between us is not about the inflammatory properties of HDL cholesterol, or insulin resistance, or grass-fed beef.

The point of difference is whether low-saturated-fat diets are unhealthy. You say yes. I say, "I doubt it." Strong claims require strong evidence.

Your article quotes Walter Willett, but my own copy of Willett's "Eat, Drink, and Be Healthy" recommends replacing saturated fat with unsaturated fat. Your article quotes Gary Taubes, the well-known low-carb science writer, but I find the evidence review for the Dietary Guidelines more convincing, because of its newer data, systematic protocol for selecting articles to review, and transparent evaluation criteria.

Describing low-saturated-fat diets as unhealthy raises concern for near-vegetarian and vegetarian diets, because these diets on average include lower amounts of saturated fat. Notwithstanding your concern about confounding variables, if it were true that low amounts of saturated fat were unhealthy, it sure would be surprising that vegetarians frequently enjoy such good health.

I'll leave the rest of this comments thread open and will read it with interest (but will resist sending any more of my own comments).

And, thanks for sharing your thoughts in this forum.

Parke

kristin wartman said...

Hi Parke,

I must clarify one thing in your response. You write, “The point of difference is whether low-saturated-fat diets are unhealthy. You say yes. I say, "I doubt it." Strong claims require strong evidence.”

Actually, I never said that diets low in saturated fats are unhealthy, rather I’m saying that diets high in carbohydrates (particularly refined grains and sugars) are unhealthy. There’s a big difference here -- one could easily eat a diet low in saturated fats that's also low in refined sugars and grains and potentially be very healthy.

The diet that I argue is unhealthy, is one based on industrial, highly processed foods loaded with refined grains, sugars, trans-fats, and other artificial ingredients. Unfortunately these were the foods that came to replace the whole, saturated fats once a major part of the American diet. I’m aware that some people do well on diets low in saturated fats that are also low in highly refined, processed foods. There are also those people who eat greater amounts of unprocessed, saturated fats in their natural state and are equally as healthy.

It’s important that the public understand the nuances and complexities of nutrition, health, and diet. No one diet will work for everyone. To simplify my argument into one inaccurate sentence that says I think "low-saturated-fat diets are unhealthy" is not only untrue, but does a disservice to people trying to make sense of the issues.

Thanks,
Kristin

Nate said...
This comment has been removed by the author.
Nate said...

Professor Wilde,

I hope you read the replies to the critiques too:

http://www.ajcn.org/content/92/2/460.citation
http://www.ajcn.org/content/92/2/459.1.citation

I'm arguing that saturated fats are not unhealthy, not that they are required for a healthy diet.

Perconsultant said...

the question we have to ask ourselves is whether low-saturated-fat diets are healthy. You say yes. I say, "I doubt it." Strong claims require strong evidence.