“The Food Pyramid? It’s just wrong.”

“A diet based on food that turns to sugar once it’s in your body was never going to end well.” The last 35 years of dietary advice summed up in one tweet-worthy sentence, and the whole embarrassing public health nutrition fiasco explained in less than 3 minutes.

Hooboy, do I have a treat for you.

Despite the recent nobody-is-listening-so-we’d-better-yell-louder American Heart Association report that encourages everyone to chug soybean oil like it’s cheap beer, butter is baaaaack–and starring in its own video.

My friend, Jenni Calihan at EatTheButter is the mastermind behind this, but I’ll humbly take credit for some “editorial input” here.

Watch first, then we’ll talk:

First of all:  I’m still snickering over the “not based on rigorous science” frame:

Next:  Maybe you’re a newbie to the world of non-mainstream, it’s-actually-okay-to-eat-animal-fat nutrition, or maybe you are just nutrition-skeptic curious, or maybe you’ve been around all of the various dietary dogma blocks.  For whatever reason, if you get a little uncomfortable with all the hyperbole, oversimplification, and finger-pointing about who Made America Fat–the cows! the Snackwells! the stupid lazy gluttonous Americans!–this little video offers a straightforward, easy-to-understand explanation for our current nutrition quandary, without resorting to distortions, exaggerations, and the blame game.

I’m happy to take some credit for that. For me, when you’re talking physiology and biochemistry (crackers turn into sugar, fat doesn’t “make you fat”), things are pretty straightforward. But it gets trickier talking about “good” or “bad” science or what the Food Pyramid did or didn’t do, or cause.

This is one my favorite things about this video: Although the relationship between dietary guidance and outcomes is noted, cause and effect regarding the specifics of the diet is not explicitly stated (although I suspect that various parties will infer what they will).

For example, when the voice-over says “The Food Pyramid? It’s just wrong,” does that mean, “The science behind the Food Pyramid is just wrong” or that “Pyramid-shaped food advice is just wrong”? The first is an argument waiting to happen, but the second just says that the advice we gave out–whether pyramid- or plate-shaped–was, for some reason, the wrong advice for most Americans.

Was it all those starches turning to sugar after few minutes in our stomachs?  Did eating less nutrient-dense food, like red meat, mean we ended up eating more food overall in order to be adequately nourished? Did the division of food in to “good” and “bad” categories set up cycles of “good” and “bad” eating? Or, to get all academic on you, did the application of the halo of “healthy” to cheap, convenient, tasty, but nutritionally lacking industrialized food, coupled with a neoliberalist imperative to make health the responsibility of the individual and “solutions” to health a matter of the marketplace, make many Americans–particularly those caught in a widening income gap and increasing economic pressures–more susceptible to eating patterns that exacerbated all of the other toxicities of modern life?

You don’t have to answer that.

My point is that we (really) don’t know what exactly about our dietary advice is the problem, and—as far as I am concerned at this point—it doesn’t matter.

For 20th (and now 21st) century Americans, this kind of dietary advice, provided in this particular political, sociocultural, industrial, and economic context, simply did not—and still does not—“work” to keep Americans healthy. That’s it.  We can argue until the grass-fed cows come home about why it didn’t work, but it didn’t work.  Acknowledge and move on.

And in the meantime, you can share this video with friends and family and know that it got the “hyperbole-free” stamp of approval from me.





24 thoughts on ““The Food Pyramid? It’s just wrong.”

  1. Hi Adele! Thanks for mentioning the book: William Rothstein’s Public Health and the Risk Factor: A history of an uneven medical revolution

    You are such a good source of obscure stuff that happens to interest me, and this book really fits the bill. Once I started digging into the whole prevention paradigm and subsequent development of risk factors, it hit me that risk factors don’t really make sense. You can’t take a population incidence and extrapolate that to any individual’s probability of getting that condition. It’s just invalid. Epidemiology is a statistical POPULATION based model, right? You can’t apply population incidence as individual risk, for so many reasons, ugh.

    This quote from a review of the book sums up my thinking quite well:

    Then he deals with the extrapolation of risk factors in populations to risk factors in individuals. Quoting Claude Bernard, he points out one problem: “‘an average description [of the risk factor for a group] will never be matched in nature,’ inasmuch as each case differs from the average. . . .’ ‘In my opinion statistics [risk factor data] can never yield scientific truth'” (p. 19).

    All of this just reminds me what a monumental tragedy we have with our current state of nutrition science. So much unnecessary suffering and death from closed minds in power.

    I like the video. It’s all not just wrong, but dangerous, harmful, and killing people every f**king day. Nutrition science and nutrition epidemiology are just layer upon layer of bad science and misinformation that’s dangerous. I wonder if some enlightened anthropologists or sociologists in the future will look back and see what dark ages we live in with regards to nutrition science and health guidance. It’s usually too sad for me to think about.

    1. I don’t read every scholarly book that I have cover-to-cover. Rothstein’s book is one of the exceptions. I was interested in his take on this because he’s a sociologist, not an MD or an epidemiologist. From what I can tell, he’s got no dog in this fight, and he brings a historical perspective that is missing from a lot of other discussions.

  2. Adele: Yes, and I suspect the decline in smoking is the largest factor in the decline in CVD deaths. I wonder about China. They smoke like crazy. And with the terrible air pollution, they must be dropping like flies. Recently finished an essential book, for all those who are interested in good health: “The Salt Fix,” by Dr. James DiNicolantonio. Another piece of stupid, dangerous medical advice bites the dust. I’ve noticed since I’ve been gorging on the stuff, I recover much faster during all aspects of my workouts, and even minor injuries are a thing of the past. Salt is the new health food.

    1. Thanks for reminding me about The Salt Fix. It’s on my list of post-dissertation reads. Salt was unmistakably valued in earlier eras as essential to health and central to safe food preservation. To treat it as if it is some kind of poison is not only to ignore physiology, but to ignore history.

      1. Since moving to a low(but not no)-carb, higher-fat diet a few years ago, I’ve found my taste for salt changed. I add more salt now to home cooked food and am left wondering whether lower salt guidelines may be associated with a Standard HCLF dietary approach? Particularly considering that over 60% of a US/UK diet falls in the ultra-processed category now.

        1. Are lower salt guidelines associated with the standard HCLF dietary approach? A good question, and I would say the answer is yes. Although people like Marion Nestle are always going on about how our dietary guidance is a mess because of industry influence, the truth is that our dietary guidance for the prevention of chronic disease was a mess from the beginning.

          The 1977 Dietary Goals brought together a number of strands of thinking that, in terms of nutrition science, were quite distinct. At the time, a survey indicated that sodium/salt reduction was relatively low on the list of priorities for dietary changes that scientists thought would help prevent heart disease. And yes, it is bizarre to conduct science through surveys–but this is not exactly science. Rather, it is that very grey area between science and policy. The 1979 AJCN review of the advice on sodium/salt given in the 1977 Goals indicates, after outlining the advantages of sodium/salt reduction, that the policy implications of this are not straightforward: “However, the advantages of preventing hypertension by reduction of salt intake in the entire population as compared with early detection and treatment of new cases have not been clarified” (Tobian 1979, 2662). In other words, sodium/salt reduction for particular individuals is clinically indicated; for populations, not so much. This illustrates the problems with having specialists in one field (nutrition science) make calls about what should happen in another area (public policy), an issue further complicated by the fact that there were other concerns at work.

          The 1977 Goals pursued two conflicting ideological commitments: 1) to have Americans eat a more “natural” diet, and 2) to have them eat less meat and butter and fewer eggs. Both of these commitments go back to counter-culture thinking (see Diet for a Small Planet, etc.), but they are in contradiction with each other because eating fewer animal products means eating more manufactured fats and egg/meat substitutes. I think sodium/salt got caught up in this because it addresses both commitments. Sodium/salt is a defining characteristic of both “industrially processed” foods (along with refined grains, sugars and oils) and “traditionally processed” foods (sausages, hams, cheeses, etc.) So even if it wasn’t a priority in terms of heart disease prevention, it helped to push forward the agenda at hand.

        2. Supposedly, low carb causes the kidneys to excrete more salt, so you need to take in more. I add salt to everything now. Currently eating beef heart and added salt multiple times to it.

          1. So, yes, there may be some physiological changes that come with increasing/decreasing carbs. Yancy 2010 demonstrated that weight reduction through reducing carbohydrate foods can produce greater improvements in blood pressure than weight reduction by other means,
            even when weight loss is similar.

            1. It’s the insulin (again!) which blockades sodium processing by the kidneys. A high insulin diet such as we are all supposed to eat mandates lower sodium, an appropriate insulin diet not so much – except for a relatively small percentage of the population who genuinely do requre sodium restriction.

              Conversely potassium has more effect on most, probably why fruit and vegetables may reduce BP.

      2. The stream-of-conciousness thought process provoked by smoking and disease led me to something which drives me crazy. It seems that “correlation does not show causation,” like “vaccines don’t cause autism,” is an essential caveat in any public statement by scientists, at least in the biological sciences. But, though true in general, it is not true as a blanket statement, and smoking and lung cancer is the best example of this. As far as I know, there never has been an RCT for smoking, and there never will be. As long as an association meets a sufficient number of the Bradford Hill criteria we can have great confidence that that it shows causality.

        1. You’ve just opened up a giant can of worms. I happen to be working on a section for my dissertation on nutritional epidemiology and causality. The problem with comparing smoking (or vaccines) as an “exposure” with eating is that even close relatives of the person whose behavior is under observation can give relatively accurate accounts of the “exposure.” Kids and spouses know that mom/dad smoked a pack a day of unfiltered Lucky Strikes; doctors have vaccine records. Even the individuals who did the eating have trouble accurately recalling what eating they did. So that’s the first issue. The second issue is weak vs.strong associations; strength of association was Hill’s first criterion. In diet-chronic disease associations, we have never gotten anywhere near the 10-fold increase in risk that smokers have over non-smokers for developing lung cancer; the National Cancer Institute has indicated that relative risks less than 2 should not be used as the basis for public policy (Milkowski 2010). The size of associations in nutritional epidemiology studies is small–relative risks of 1.5 to 1.6–accompanied by considerable opportunities for random or systematic error and bias (Byers 2001).

          Hill’s criteria were never meant to be a “checklist” (Hoffler 2005), but rather considerations that should be taken into account when making judgments about associations in epidemiology. But even taking the “checklist” approach, nutritional epidemiological studies of chronic disease (NECD) fail that test. In addition to weak associations, there is lack of consistency (especially when non-white and/or non-American populations are observed). There is lack of specificity (saturated fat “causes” heart disease, diabetes, stroke, etc. etc.). With cross-sectional studies, there is lack of temporality. Frequently there is no consistent biological gradient (i.e. more exposure does not equal more disease), which is why many NECD studies compare the “highest” and the “lowest” quartiles or quintiles–because the intervening levels of consumption demonstrate a non-linear relationship. The one that bugs me the most is lack of plausibility; far too many publications simply skip the whole proposed biological mechanism discussion altogether. It goes without saying that there is lack of coherence in NECD studies; lack of proposed mechanisms makes it impossible to estimate how well observations fit them. Experimental confirmation? Um, no. Not at all. The “analogy” criterion is probably the most problematic because it begins with the assumption that what happens with food and diet is analogous to what happens with deficiencies and toxicities, but this was the question under consideration from the beginning.

          1. I would think that only scratches the surface. For instance, why does everyone believe natural processes need to be linear? Many are bathtub curves (high near “zero”, then low, then high again far away from “zero”), such as total “cholesterol” versus overall mortality (for some studies, on men, anyway — high “cholesterol” for women appears protective). Some may be J curves, which is what salt might be: too low salt = high total death/disease, but over a threshold, you get gradually rising rate of death/disease.

            Also, genetics must play a role. I don’t do well with any carbs, but there are others I know who do not have nearly the problem I do. Were we to eat the same carbs, my blood sugar would be sky-high, while theirs would barely budge.

            Add to that poverty/rich, insulin resistant/not insulin resistant, the complexity of the human biome and its possible effects (which may change depending on your diet), the crappy way most studies are done, etc., and I’ve reached the conclusion we probably don’t know what’s going on.

            1. It does just scratch the surface 🙂 Which is why the idea of taking an intervention that benefits a few at-risk folks and applying it to an entire population is a really questionable endeavor. I have a chapter in a book coming out soon on the ethics of this sort of approach to public health.

            2. Yes I see biochemistry as a stacked series of intersecting J curves and U curves, where responses to drugs are about the only linear thing.

            3. Have you read Roger Williams’s Biochemical Individuality? I think you would enjoy his (too often overlooked) perspective on the complexity of trying to understand how those intersecting and various responses contribute to an individual’s own highly idiosyncratic internal milieu.

            4. Oh PS

              the only pyramidal thing is your body shape when you eat the Food Pyramid

  3. The AHA can always say that the advice lowered CHD mortality and therefore extended life, whatever the other consequences.
    However this is not supported by their own evidence. For example, in Finland, really the poster child for this sort of advice, saturated fat decreased significantly between 1982 and 2007, but there was only a small increase in PUFA – about 1%. It was NOT a “substitute PUFA for saturated fat” intervention of any strength at all.
    It was a “substitute carbohydrate for fat and SFA” intervention. Like WHI. Like something that can’t be supported at all by referring to the substitution meta-analyses of Jakobsen or Farvid et al or any meta-analysis of RCTs.
    The only macronutrient factors that could have affected CHD risk in this population, according to metas of RCTs and associational studies are 1) a small increase in ALA, maybe. 2) an improvement in carbohydrate quality, with all that implies about micronutrients. And neither of those things requires a decrease in SFA per se. It merely requires a more varied diet than “spreading butter on your cheese”, if people really did that, in a context where the carbohydrate foods are sugar, flour, and potatoes, and there are no fresh vegetables or fruit.

    Public Health Nutrition: 13(6A), 932–938 doi:10.1017/S1368980010001126
    Explaining the 25-year decline of serum cholesterol by dietary changes and use of lipid-lowering medication in Finland
    Liisa M Valsta*, Heli Tapanainen, Jouko Sundvall, Tiina Laatikainen, Satu Ma¨nnisto¨,
    Pirjo Pietinen and Erkki Vartiainen

    1. “The AHA can always say that the advice lowered CHD mortality and therefore extended life, whatever the other consequences.” The implication then would be that Americans actually followed this advice, making it much harder to weasel out of a cause/effect relationship regarding diabetes.

      As for the decline in morbidity and mortality rates in the U.S., I wouldn’t be the first person to point out that the decline in deaths from heart disease began more than a decade before the Dietary Goals/Guidelines were created. Others have noted that there is a distinct disconnect between declines in serum cholesterol levels, which began only after 1976-1980 surveys, and the decline in heart disease mortality, which began, as I said, long before. If avoiding sat fat leads to lower cholesterol levels, which in turn leads to reduced risk of heart disease, that’s not what seems to have taken place. See William Rothstein’s Public Health and the Risk Factor: A history of an uneven medical revolution for more on this.

      It’s worth noting that in both Finland and the U.S. rates of smoking declined dramatically coincident with the decreases in heart disease morbidity and mortality.

      1. Sacrilege! String her up!

        Anybody got any string? (for those old enough to remember the Firesign Theater)

        1. My husband has one of their comedy albums–for those of us who remember albums 🙂

Leave a Reply to Gary Ogden Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s