No Big Surprise ... A Compendium of Errata Etc. from The Big Fat Surprise by Nina Teicholz

After reading the new book  The Big Fat Surprise: Why Butter, Meat and Cheese Belong in a Healthy Diet by Ninary Taueicholz more throroughly, I became rather frustrated.  How does a book that contains so many glaring errors even make it into print?  This book is being promoted as discussing how scientists got everything all wrong, one in particular.  It discusses nutritional science and recommendations gone wrong.  It firmly places the blame for obesity on advice to reduce dietary fat.  In doing so, however, Teicholz commits far more egregious errors than any she would like you to believe have occurred.  So I'll do a revolving post, adding them in here as the mood strikes, rather than clutter the blog with tons of posts.   My hope is that this post doesn't end up to be of epic length, and that this book turns out to be a flash in the pan that fades quickly from memory.    Each new entry will bump the post and be at the top with previous entries to follow, most recent first.  They will be separated by a horizontal line and each section dated with a *** preceding it that you may search on that string in your browser to scan for others.  

---

*** 2/27/2015

Americans Have Reduced Fat Intake From 43% to 33%

In the Introduction, Teicholz makes the following claim:

Unaware of the flimsy scientific scaffolding upon which their dietary guidelines rest, Americans have dutifully attempted to follow them.  Since the 1970s, we have successfully increased our fruits and vegetables by 17 percent, our grains by 29 percent, and reduced the amount of fat we eat from 43 percent to 33 percent of calories or less.

When I first read this book, I was taken aback by these numbers which conflicted with even the statistics most commonly cited by low carb advocates (NHANES).   I was further taken aback that such a crucial premise to the entire book was unreferenced.  That was, until, I became aware of Teicholz's use of Taubesian obfuscatory notes.   What makes both her and Gary Taubes' use of these notes all the more curious and maddening that, in seemingly arbitrary fashion, both do at times utilize a numbered referencing system.  In order to know whether a statement is actually referenced, one would have to flip back and forth to the Notes as they read each paragraph.  It's little solace that at least on Kindle one can click on the note and be linked back to the text.  This makes fact checking these two abominable works very difficult and time consuming, a fact that is evidently deliberate on both of their parts.  

I'm not sure why I didn't bother until now, but the ridiculous tooth gnashing going on in LLVLCluelandia of late over the release of the 2015 DGAC Report has spurred a new spate of "low fat made us fat" hysteria and ever the more wild claims about American compliance, dutiful adherence, obedience and whatnot with what their government told them to eat.

So, here is the note (formatted as it appears in my Kindle version of BFS)

“33 percent of calories or less” . . . “share of those fats that are saturated has also declined”: Centers for Disease Control and Prevention (CDC), ‘“Trends in Intake of Energy and Macronutrients— United States, 1971–2000,” Morbidity and Mortality Weekly Report 53, no. 4 (2004): 80– 82.

So, first things first.  The 43% figure remains unreferenced, because, as you can see, this notation style uses snippets of phrases and links those to the source.  However, since this same source dates from 1971, one would expect to see the 43% figure in this report.   For a change, this report was incredibly easy to find online.  Googled the title, and first link clicked and voila!  If you prefer some further analysis in PDF form, there's also this.

As it turns out, this is the exact same source as for these plots that are ubiquitous around the IHC thanks to renowned thermodenialist Richard Feinman.

What the report says is as follows, direct quotes but bullet-pointed for ease of reading, (all p-values removed were p less than 0.01):

• During 1971--2000, a statistically significant increase in average energy intake occurred (Table). 
• For men, average energy intake increased from 2,450 kcals to 2,618 kcals
  
• for women, from 1,542 kcals to 1,877 kcals 


• For men, the percentage of kcals from carbohydrate increased between 1971--1974 and 1999--2000, from 42.4% to 49.0% , and for women, from 45.4% to 51.6%  (Table). 
• The percentage of kcals from total fat decreased from 36.9% to 32.8% for men and from 36.1% to 32.8%  for women. 
• In addition, the percentage of kcals from saturated fat decreased from 13.5% to 10.9% for men and from 13.0% to 11.0%  for women. 
• A slight decrease was observed in the percentage of kcals from protein, from 16.5% to 15.5% for men and from 16.9% to 15.1% for women.

The decrease in the percentage of kcals from fat during 1971--1991 is attributed to an increase in total kcals consumed; absolute fat intake in grams increased (5).

USDA food consumption survey data from 1989--1991 and 1994--1996 indicated that the increased energy intake was caused primarily by higher carbohydrate intake (6).

Data from NHANES for 1971--2000 indicate similar trends. The increase in energy intake is attributable primarily to an increase in carbohydrate intake, with a 62.4-gram increase among women and a 67.7-gram increase among men. Total fat intake in grams increased among women by 6.5 g and decreased among men by 5.3 g.

This would amount to a net 1.2 gram INCREASE for the population as a whole.  In other words, as a population, we did NOT decrease our fat intake.   This amount, either way, amounts to roughly one pat of butter.  A teaspoon of oil.  In other words, it is MEANINGLESS no matter if statistically significant.

As far as I'm concerned, peddling increases or decreases in percentages as meaningful increases or reductions in intake is rank dishonesty.  Further compounding that deception by representing changes as percentages of percents is unconscionable.  

But by percentages, fat intake fell from 36.5% (averaged men & women) to 32.8% (same for both).  This is a FAR cry from the 43% claimed in the text as a starting point in the 1970's.  And this is Teicholz's source for the 33% "or less" flair.  A figure of 0.2% less is hardly worthy of displaying the decimal place.   Rather than a reduction of 10% of total calories, we're talking slightly more than one-third of that:  3.7% of total calorie difference.   But this is meaningless anyway since the change was -- in the words of the report Teicholz cites -- due to an increase in total caloric intake that was predominantly in the form of carbohydrate.  A mathematical anomaly.  Not, NEVER WAS, due to a reduction in fat intake.  

We have not successfully cut a darned thing.  The real question in obesity and nutrition these days should be:

WHY DO THESE PEOPLE 

KEEP LYING TO YOU???? 

---

*** 5/21/2014

Excess Protein, Rabbit Starvation, 

The Inuit & Vilhjalmur Steffanson

In Chapter 10, entitled Why Saturated Fat is Good For You, Teicholz bemoans Volek's frustration with the mainstream nutrition established for not taking a "more unbiased, balanced" approach.  (Ahem)  She discusses how a long term trial of two or more years would be needed to see the effects of so much protein and fat in the diet.  This includes a link to end-of-chapter note XIX.

XIX . The effects of too much protein was one concern, and this is justified— but is problematic only when a diet lacks fat. When protein is eaten, the kidneys and liver remove the nitrogen and excrete it through the urine. Dietary fat is essential to this process.(1)  When overly lean meat is eaten, nitrogen cannot be properly processed and builds up to potentially toxic levels. This condition is a common danger to dieters nowadays, eager to cut back on carbohydrates but, given longtime biases, reluctant to eat more fat. (2)    The Inuit considered overly lean meat to be an inadequate source of nourishment. Stefansson dubbed the problem “rabbit-starvation” and suffered from the condition himself when he went through a period of eating lean meat but not enough fat during his yearlong meat-only experiment in 1928 (Stefansson 1956, 31) (3).   {Kindle Locations 5768-5771}

(1) The idea that dietary fat is involved in, let alone essential for protein metabolism is simply not to be found in any reputable text on the topic. We're in Nora-land here on this one.  Here's a quick look at protein metabolism, etc., probably unnecessary here, but more to drive this point home.  The problem with protein is that while it can be broken down to feed into various energy generating pathways, when the "amino" part is removed, we get ammonia and this is indeed toxic if levels get too high.  Luckily, most of us have lovely livers that make sure this doesn't happen by converting it to urea that is excreted in urine.   As with alcohol, the liver can only process so much and the kidneys can only clear so much.    Too much protein is a concern whether or not carbohydrate or fat accompanies it in the diet.  

(2) The issue with "too much protein" is based on the limits of metabolism, not the leanness of the protein.  The theoretical maximal range of protein intake is quite varied in reports -- I've seen anywhere from 250g up to around 400g.  That's anywhere from 1000 to 1600 calories.  Thus protein cannot sustain a human for very long on its own.  To repeat, this has nothing to do with carbs or fat in the diet, indeed your basic PSMF is generally considered quite safe and beneficial to many.  Dieters need not fear eating lean protein.  However, clearly lean protein cannot sustain a human being.  This brings us to ...

(3)  "Rabbit Starvation" - It's truly sickening how this term has been thrown around in the community and misrepresented.  All the more strange when this is a moot point to a dieter trying to lose weight who has plenty of fat to burn to sustain them.  But check out Wikipedia on this topic:

Rabbit starvation, also referred to as protein poisoning or mal de caribou, is a form of acute malnutrition caused by excess consumption of any lean meat (e.g., rabbit) coupled with a lack of other sources of nutrients usually in combination with other stressors, such as severe cold or dry environment. Symptoms include diarrhea, headache, fatigue, low blood pressure and heart rate, and a vague discomfort and hunger (very similar to a food craving) that can only be satisfied by consumption of fat or carbohydrates.

Starvation, as in malnutrition ... note that there is zero fear of malnutrition if carbohydrates are supplied in the diet.  Like in, oh, say, those Native Americans from the Southwest.  Rabbit Starvation was a very real issue for those like the Inuit when BOTH carbohydrate and fat sources were scarce, particularly in the late winter and early spring.  The Inuit didn't arbitrarily restrict carbohydrate, remember.  

Everyone with a basic knowledge of nutrition and human metabolism would consider lean meats to be an inadequate source of (total) nutrition.  That's why we eat other stuff when available, like the tubers, turnips, corn and berries in addition to fat.  

The next claim is the biggest whopper of them all:  "Stefansson ... suffered from the condition [rabbit starvation] himself when he went through a period of eating lean meat but not enough fat during his yearlong meat-only experiment in 1928"   FALSE

A pattern is emerging where one can only presume Teicholz has never even seen many of the primary manuscripts to which she attributes her facts and quotations.  Here is one paper describing Stefannson's Year 'o' Meat that has been so erroneously reported and abused to drive an agenda.

PROLONGED MEAT DIETS WITH A STUDY OF KIDNEY FUNCTION AND KETOSIS.

“During the first 2 days his diet approximated that of the Eskimos, as reported by Krogh and Krogh (3), except that he took only one-third as much carbohydrate. The protein accounted for 45 per cent of his food calories. The intestinal disturbance began on the 3rd day of this diet. During the next 2 days he took much less protein and more fat so that he received about 20 per cent of his calories from protein and 80 per cent from fat. In these two days his intestinal condition became normal without medication. Thereafter the protein calories did not exceed 25 per cent of the total for more than 1 day at a time. The high percentage of calories from protein may have been a factor in the production of the diarrhea. “

You see, Stefannson attempted to emulate the meat-heavy Inuit diet as reported, not as his various musings have been interpreted.  This is similar to that in Heinbecker.   You see for the first two days he tried to eat 45% protein, 55% fat which amounted to roughly 275 grams protein per day.  He got sick.  In TWO days.  After that his intake dropped by around 4-500 calories and protein was dropped to 20%, thus around 100 grams per day of protein.  He got better.  The Inuit eat a lot of rotted "high" seal flesh.  They are also adapted from rearing and no doubt genetically to high protein consumption.  Stefannson was not.  He was experiencing some degree of protein (ammonia) toxicity ... but starvation?  Hardly.  At 55% fat he  was taking in roughly 150 grams of fat.  In summary:

• 2 days - no true "starvation" occurs in this time frame
• 275 g protein - likely exceeded his threshold of metabolism and some degree of ammonia toxicity occurred.
• 150 g fat does NOT equal starvation, rabbit or otherwise

I'd be remiss if I didn't point out one more flaw in including the Inuit in any discussion of saturated fats.  The fat in their diet is largely marine mammal derived, very heavy on the unsaturated fats, light on the saturated.   It's time to end the "me too" copycatting of using this culture and romantic notions of Stefansson to drive the agenda.  Please!

---

***  5/19/2014


UPDATE 2/27/2015:  There is an update to this error that I've blogged separately on here:  Nina Teicholz "Corrects" The Big Fat Surprise ~ Digs Hole Deeper.  Essentially, in the paperback released in January 2015, Teicholz corrected some errors.  In the case of this error, she altered the wording slightly to imply that the elders had residual health from being raised on meat-centric, mostly buffalo, diets, and attributes the buffalo eating to a different book and set of tribes.  This time to the Southern Plains Indians of Texas and a 1961 book by W.W. Newcomb, Jr.    /UPDATE

Hrdlička and the diet of Southwestern Native Americas

In my recent (partial) review, I highlighted a smattering of issues with the book but focused mainly on her misrepresentation of Aleš Hrdlička's Physiological and Medical Observations among the Indians of Southwestern United States and Northern Mexico.  In Chapter 1, entitled The Fat Paradox, Good Health on a High Fat Diet, Teicholz spans the globe misrepresenting the diets of culture after culture, from the Inuit to the Masai, then to the Hunza of India and the Native Americans of the Southwest US, Teicholz presents examples of cultures we are to presume consumed a high fat diet.  While the former two eat more animal fat, I'm hard pressed to find that the Hunza did, and as discussed at length in my last post, the SWNA (gotta coin an acronym here, South West Native Americans) most certainly did not eat anything of the sort.  Here is the quote I take issue with:

Meanwhile, the Native Americans of the Southwest were observed between 1898 and 1905 by the physician-turned-anthropologist Aleš Hrdlička, who wrote up his observations in a 460-page report for the Smithsonian Institute. The Native Americans he visited were eating a diet of predominantly meat, mainly from buffalo, yet, as Hrdlička observed, they seemed to be spectacularly healthy and lived to a ripe old age. The incidence of centenarians among these Native Americans was, according to the 1900 US Census, 224 per million men and 254 per million women, compared to only 3 and 6 per million among men and women in the white population. Although Hrdlička noted that these numbers were probably not wholly accurate, he wrote that “no error could account for the extreme disproportion of centenarians observed.” Among the elderly he met of age ninety and up, “not one of these was either much demented or helpless.”  {KL 305-311}

There can be NO doubt that Teicholz attributes the observation of eating buffalo to Hrdlička and the SWNA.   Again, per previous discussion, the free full book is digitized on Google.  I searched the book and found no reference to dietary use of buffalo at all among the SWNA.    So I tweeted a question to Teicholz:

Well, I DID buy a copy of this book and I have been reading it, but her response is typical.  

Now Teicholz has a peculiar notation and referencing style ... but not an unfamiliar one to anyone who has read Taubes' books.  I am convinced that the only reason to employ such a system is to make it as difficult as possible for anyone who choses to do so, to check references.  This is anathema to the spirit of science journalism, especially that of the "everything you thought you knew is wrong" variety.  In each chapter you have "linked" notes, there are four in Chapter 1 that explain some topics a bit further.  Then, towards the end of the book (from location about 6100 to about location 8100), we have Notes.  Here, chapter by chapter, blurbs and wordings are attributed to their sources in the following format -- I'll use the applicable notes to the SWNA excerpt as an example:

wrote up his observations in a 460-page report: Aleš Hrdlička, Physiological and Medical Observations among the Indians of Southwestern United States and Northern Mexico, Smithsonian Institution Bureau of American Ethnology Bulletin 34 (Washington, DC: US Government Printing Office, 1908).

were eating a diet predominantly of meat: Joseph M. Marshall III, The Day the World Ended at Little Bighorn: A Lakota History (New York: Penguin Books, 2006).

“no error could account for”: Hrdlička, Physiological and Medical Observations, 40– 41.

“not one of these . . . demented or helpless”: Ibid., 158.  {KL 6173-6179}

It would be difficult to flip back and forth in an electronic text let alone a hard copy book! Nevermind that the notes section in Kindle is one third the length of the text itself. Note that these are neither numbered for easy location nor linked to within the text, though thankfully in the electronic copy, they do link back to the text location. Finding them in the Kindle book was made easier using the search feature. I would note that in many, but not all, page numbers are provided. Not for the relevant citations here, and we're talking lengthy books, not journal articles. Before going on, there is then a third set of references in the form of an alphabetically ordered list of all references. These are neither numbered nor linked back or forth to their usage in the book. All of which, as with Taubes, makes for what appears to be meticulous, considerable and comprehensive research ... as long as you don't look too closely.

I have highlighted part of the middle note in red, because, THIS is the source of Teicholz' buffalo claim.   The Day the World Ended at Little Big Horn: A Lakota History was written in 2006 by Marshall who was born and raised on the Rosebud Sioux Reservation.  The book is described as a different look at the Battle at Little Big Horn.   As per the map on the right, the Lakota are a Central Plains nation, an estimated 1000+ North East of the SWNA (depending on whether one were to go edge-to-edge, or center to center, and when boundaries are drawn).

So firstly, Teicholz didn't just mistakenly attribute a different first-hand accounting of food traditions to the wrong author as an omission.  Marshall is no Hrdlička to the Lakota.   One could almost chalk this up to a mistake was there no notation citing the Marshall book.  But there was.  Buffalo was the major food consumed by the Lakota, but to infer this was necessarily a high fat culture is pretty absurd.  Wild buffalo meat is lean and fairly low in saturated fat.  So yeah, had Teicholz accurately cited this tribe, they make a case for meat consumption.  But buffalo isn't today's beef (even grass fed), and butter and cheese?  Sorry.   Further, the Lakota eat quite a lot of tubers plentiful to their region.  All a rather moot point because Teicholz never mentions the Central Plains Indians in general, nor the Lakota specifically, as she claims their diet was that which Hrdlička observed in what is now the South West United States.  UNBELIEVABLE!!!  I guess Simon & Schuster can't be bothered with any fact checkers?  

Teicholz could have provided me with the source from her footnotes.  Instead she said "read the book" ... which is the all too common retort.  I understand that one can't properly review a book without reading substantial portions of it, and for that I did purchase the book.  But I should not have had to purchase it to get the information on the reference.  She knew it wasn't to be found in Hrdlička's book.  This is journalistic malfeasance and dishonesty.  The book is FULL of it.