Nina Tei¢holz, Shai'ster ~ Part I: The Diets of Shai et.al. and Extrapolating from Weight Loss Studies

Preface:  First, yes, I am aware that spelling Teicholz's name with a ¢ and using my Shai'ster play on words will rile my critics.  But I'm going to do it anyway because it is just so fitting.   In many ways, The Big Fat Surprise is frighteningly similar to Gary Taubes' Good Calories, Bad Calories,  and were it written by someone other than a friend and mentee, I'm quite sure Taubes would have a strong case for plagiarism.  Thus ¢ is the mini-$.  

Secondly, having listened to or read now several interviews, and after finally completing her book, it becomes evident that most of Teicholz's "case" for why butter meat and cheese should be part of a healthy diet rests on a single dietary RCT.  I am, of course, speaking of the Israeli study known by first author Shai.  I apologize to Dr. Shai for my play on words, it is not meant in any way to disrespect her.  However her study is being misused and abused in so many ways by the author of this book, as Gary Taubes did before her, that I cannot help but resurrect the Shai'ster term either.

Nina Teicholz on Nightline, link to full story

For the sake of keeping posts to a reasonable length, and publishing more often as time permits, I'll break this up into parts.  There are so many ways in which Teicholz misuses this trial in her book, the specifics of which I'll address in future parts ... how many it is hard to tell.

 

Shai et.al.  

The Study:    Weight Loss with a Low-Carbohydrate, Mediterranean, or Low-Fat Diet, also Supplementary Appendix (freely accessible). 

The Diet Composition

This installment will focus on the diets, assessment of intake, etc.  From the original paper, so there is no confusion:

Low-Fat Diet

The low-fat, restricted-calorie diet was based on American Heart Association20guidelines. We aimed at an energy intake of 1500 kcal per day for women and 1800 kcal per day for men, with 30% of calories from fat, 10% of calories from saturated fat, and an intake of 300 mg of cholesterol per day. The participants were counseled to consume low-fat grains, vegetables, fruits, and legumes and to limit their consumption of additional fats, sweets, and high-fat snacks.

Mediterranean Diet

The moderate-fat, restricted-calorie, Mediterranean diet was rich in vegetables and low in red meat, with poultry and fish replacing beef and lamb. We restricted energy intake to 1500 kcal per day for women and 1800 kcal per day for men, with a goal of no more than 35% of calories from fat; the main sources of added fat were 30 to 45 g of olive oil and a handful of nuts (five to seven nuts, < 20 g) per day. The diet is based on the recommendations of Willett and Skerrett.21

Low-Carbohydrate Diet

The low-carbohydrate, non–restricted-calorie diet aimed to provide 20 g of carbohydrates per day for the 2-month induction phase and immediately after religious holidays, with a gradual increase to a maximum of 120 g per day to maintain the weight loss. The intakes of total calories, protein, and fat were not limited. However, the participants were counseled to choose vegetarian sources of fat and protein and to avoid trans fat. The diet was based on the Atkins diet (see Supplementary Appendix 2).22

Teicholz introduces this study first in her discussion of the Mediterranean diet, and describes the diet used here as the "real" Mediterranean diet.  I point this out because there is some controversy over whether Willett's version is truly representative.   It is interesting to note that back when this was first published, Dr. Eades, who also had a hand in manuscript review of BFS, noted it odd that a diet based on Atkins would counsel vegetarian sources of fat and protein.   The sample menus provided in the Appendix are enlightening, and, frankly, I wonder how anyone stuck to these diets.   Two slices of bread and a slice of cheese for dinner?  Sour cream -- just sour cream -- for breakfast?  I've taken the liberty to include screenshots below for one of each sample day's foods.  You can click to enlarge.

           

I would note that the veggie sources of protein and fat are not reflected in either of the low carb menus.  While there is vegetable fat, there is also ample animal fat, and veggie protein seems almost entirely absent.   This really does not start this study off on the right foot at all.   

Intake Reporting & Assessment

This study was conducted in a workplace environment.  The main attractiveness of that was that in Israel lunch is the largest meal of the day.  Therefore, the cafeteria facilities could be used to make it convenient to select appropriate meals for one's assigned diet:  "The self-service cafeteria in the workplace provided a varied menu and was the exclusive source of lunch for the participants."   This aspect has been enthusiastically interpreted to be something that it is not, a factor I'll cover in a future segment.  But for now, there is no indication that participants were served only appropriate meals or that there were any prohibitions against a participant selecting an inappropriate meal.   It is nice that for one meal, foods were labeled with complete nutritional information including a color coded system to help assist in quickly identifying appropriate meals for the diet.

The dietary intake was reported quite extensively.  This is a screenshot for the first part of Table 2, there is more on the next page of the PDF concerning types of fat and fiber.  It is reported to the tenth of a calorie, tenth of a gram, etc.  

But one red flag ... for all the information given ... we are not given the initial absolute dietary intake at baseline.  Mathematically this can be calculated from the information given, but when trying to do so, the results were inconsistent.   

Before I go on, however, as mentioned when I first blogged on this study, WHAT DO THESE NUMBERS MEAN???   This was a free-living study.  The participants had some counseling support, etc., they even got spouses involved in a subgroup.  But where did this seemingly detailed intake information come from?

Adherence to the diets was evaluated by a validated food-frequency questionnaire24 that included 127 food items and three portion-size pictures for 17 items.25 A subgroup of participants completed two repeated 24-hour dietary recalls to verify absolute intake (data not shown). We used a validated questionnaire to assess physical activity.26 At baseline and at 6, 12, and 24 months of follow-up, the questionnaires were self-administered electronically through the workplace intranet. The 15% of participants who requested aid in completing the questionnaires were assisted by the study nurse. The electronic questionnaire helped to ensure completeness of the data by prompting the participant when a question was not answered, and it permitted rapid automated reporting by the group dietitians.

So the FFQ where the data reported in the original study came from, was for a single day, at 4 time points over a 2 year period.  No weighing.  No measuring.  No real-time logging.  

What of this 24-hour recall?  Still highly fallible but it sounds like these were two reports.  Well, I found another paper from this same study:  Adherence and Success in Long-Term Weight Loss Diets: The Dietary Intervention Randomized Controlled Trial (DIRECT).  (I have the full text but it is imprinted with information about the source so I cannot share.  If anyone provides me with a shareable copy, I'll do that.)   Finally!  The baseline intakes ... and this 24-hour recall data that should have at least been published in the Appendix.    This makes matters quite a bit worse.   Assessments by this method were done in 67 subjects:  27 LF, 22 MED and 18 LC.   The 12 month timepoint overlaps in these two assessments so I constructed the table below:

Baseline calories were between roughly 2300 and 2350 which seems kind of low for a group of mostly (~85%) middle aged 200 lb men.  But we'll work with that.  In the second set of data, I used the reported differentials from Table 2 and subtracted them from baseline.  The third set comes from the Adherence paper.  The change in intake of 300-600 calorie deficit at one year out is difficult enough to believe given that the LF and LC groups were regaining from the 6 month point.  But the 24 hour recallers are reporting a deficit of around 1000 calories per day!!  

It is obvious, now, why the absolute intake information was not provided in the original paper, and why the 24-hr recall data was not reported.  Because this data is so inconsistent.

Relevance of Study Diets to Claims:

The bottom line here, is that Teicholz is holding this study up as definitive proof of the superiority of high fat diets for health and weight control.  I'm going to point out a few of the many errors in this logic.

• All dieters, at all points, reported DECREASING absolute intake of ALL THREE macronutrients.

This is important, because whatever health benefits Teicholz wishes to attribute to the "high fat" diet in this study were not obtained by eating more of anything, or high anything.    

• FFQ Reported:   The difference in fat intake between "low fat" and "high fat" was 14 grams, roughly a tablespoon.
• 24-hr Recall:  The difference in fat intake between "low fat" and "high fat" was 26 grams, under two tablespoons.

Now while the 24 hr recall should be more accurate, it is in no way more believable than the more reasonable (though still questionable) FFQ data.  The differential in fat intake is higher only because the low fat dieters reported really cutting their fat intake by some 40 odd grams, while the "high fat" group reported cutting its fat intake by over 25 grams.  That's right.  There were 18 Atkins dieters who reported cutting almost 2 tablespoons of fat out of their diet. 

• The low carb group (original study data, not on my table) increased saturated fat intake by 0.28 to 0.78 grams per day.  

Those decimal points are not typos.  Under one gram per day absolute increase in fat intake.  

• THE FAT INTAKE REPORTED DURING THE TWO YEARS OF THIS STUDY NEVER EXCEEDED 40% OF CALORIES OR 80 GRAMS OF FAT.

---

Impacts of Hypocaloric (Weight Loss) vs. Eucaloric (Maintaining/Usual) Diets on Health:

Before I sum this installment up, I am inserting this aside.  So many of the metabolic consequences of a crappy diet are directly attributable to chronic overconsumption and excess weight (fat) gain (or maintenance of that fatter self with higher caloric intake).

Nobody can be in weight loss mode for years on end.  A "normal" person in our modern society where access to food is virtually "free" is relatively weight stable and will only be in minor surplus or deficit for a few days perhaps weeks at a time. 

Health improvements seen in clinical trials involving weight loss are simply NOT useful in determining what constitutes a healthy diet for the general population.  This is where the much maligned dietary epidemiological studies are important.  At least if I am of European descent, if I look at the traditional diets of healthy people from my heritage, I have some assurances that such a diet will be a healthy one for me.   If not my heritage, then at least a diet (preferably with comparable lifestyle and environment) observed and followed for decades by healthy cultures is a good place to look for ideas on what might be a healthy diet.  This is why the high fat dietary extremism in vogue these days is so ridiculous, but I digress ...

If you argue that the results of weight loss diets are indeed evidence to identify a healthy lifelong diet, then you are saying that eating McDonald's is a healthful diet because both Fat Head and this guy saw health improvements when losing weight consuming a calorie controlled diet of mostly McD's foods.  Or perhaps there needs to be a Chapter 11:  Why Twinkies are Good For You added to BFS.  After all, remember Mark Haub?  He lost over 25 lbs and improved his health markers eating junk foods.  So that means the American diet doesn't have enough Twinkies in it?  The USDA recommendations banishing Twinkies to the very tip of the pyramid (and not even a token sliver set aside on My Plate) are what caused our obesity epidemic?  By Teicholz's logic ....

---

Will the Real Low Fat Diet Please Stand Up?

In her book, Teicholz makes the unsubstantiated claim that Americans have reduced their fat intake from 43% of calories to 33% of calories.   She uses the word "dutifully" in the book and interviews  quite frequently, as if we're a bunch of dietary drones acting on every recommendation that comes from some governmental or institutional agency.  The NHANES data only put our fat intake at 36-37% in 1970, and we've dropped that to 33%.  But have we decreased fat intake?  NO.  We just added more calories of carbs than anything so that the percentage decreased.  

This is the "high fat" trick of the weight loss world too.  The recent diabetes trial pitting a ketogenic LC diet against MCCR, the authors even marveled at the lack of change in LDL despite increasing fat from 38 to 58%!  Oh my!!!  Only the absolute fat intake declined an insignificant amount (112 to 110 grams).

As far as I'm concerned, anyone who uses percentages in this manner is guilty of intellectual dishonesty.  

So ...   In Shai, the fat intake of the Mediterranean dieters, as reported by either method, was roughly 33-34%.  In other words, a "low fat diet".  

Teicholz is guilty of journalistic and all manner of whatever you want to call it malfeasance in her presentation of this study comparing Atkins/"high fat" to the ABK/Mediterranean diet.  At most time points there is NO significant difference in the outcomes between these two arms.  Most of the statistical tests appear to have been done using the LF group as reference.    In most parameters the two groups were statistically similar, and the Mediterranean diet -- as noted in the abstract -- outperformed Atkins with diabetics in this study.   There's lots more to say on this in future installments regarding gender differences, drop-out rates and adherence rates, "successes" v. "failures", etc.

Long Term .... How About 6 Years Nina?

The most troublesome of all about Teicholz's claims is that her wording and the marketing of her work implies comprehensive study into all the research.  I am going to wager that Teicholz hasn't even really read Shai, let alone the 2009 Adherence paper from that same study.   There is no indication that she is at all aware of the 4 year follow-up to the study published in late 2012.  One might think that this was too late to change the book, but by her accounts, it does not appear that Teicholz has been working on this continuously for a decade.    There are quite a few references to papers published as recently as late 2013 in the book.   Even IF her over-exaggerated claims of "high fat superiority" -- she uses that word definitive a lot too -- were true, they all really go to pot by 6 years out.    From the book:  

"The one statement that seems safe to make is that the refined carbohydrates and sugars that we were recommended to eat by the AHA as part of a healthy, fat-avoiding diet , are not merely indifferent, “empty calories,” as we’ve long been told, but are actively bad for health in a variety of ways.  Moreover, the clinical trials in recent years imply that any kind of carbohydrate, including those in whole grains, fruits, and starchy vegetables, are also unhealthy in large amounts. Remember that the Shai study in Israel found that the Mediterranean diet group, eating a high proportion of calories as these “complex” carbohydrates, turned out to be less healthy and fatter than the group on the Atkins diet, although they were healthier than the low-fat alternative.  {Kindle Location 5456 of 12033}

Only this is not even true.  There was no difference in weight loss between the Atkins and Mediterranean diets from roughly 1 year on out.  NONE.  There was a statistically greater regain amongst the Atkins dieters vs. the Mediterranean dieters during the 4-year followup.

And last but not least, this is interesting.  If you adjust "baseline" to the 5 month nadir (I need to get rid of the y-axis marks.  What happens for the next 19 months of the intervention and additional 4 years of follow up is interesting.  It sure as heck doesn't demonstrate any sort of superiority for the Atkins diet.  I guess it really is insignificant after all that the Atkins Foundation partly funded this study after Shai met Westman and decided to add this third test group to her study.  

One last thing.  This wasn't a proper and true Atkins diet?  OK.  But you take the good with the bad and Teicholz portrays this as virtually unimpeachable proof of high fat glory.  But how about that low fat arm?  That wasn't low fat.  Not according to Yerushalmy and Hilleboe anyway.  Way more fat than Israelis were consuming in 1950.

---

More on Nina Teicholz and The Big Fat Surprise:

• Ladies and Gentlemen I present .... The new Gary Taubes. Nina Tiecholz
• Book Review ~ Big Fat Surprise by Nina Teicholz
• No Big Surprise ... A Compendium of Errata Etc. from The Big Fat Surprise by Nina Teicholz