Verlana.com - Copy of CARBOHYDRATE MANAGEMENT

COPY

link to holh6.googleusercontent.com/vuHfeynWmrHAHk2fAAhYuTpx6YmfvhMV1TIptVpYZi8A74VLmchfz2ZPOOd85fF_iiDMO8A5P9X62qgbzeVo8CpzjrecqblXShNGIIxLOlHIakKHo0xvORA8zbrU85XYZQ=w1280 me

PRAISE FOR NUTRITION IN CRISIS

“Fascinating book, full of original, interesting material by one of the original low-carb

researchers whose grounding in the field goes back decades. Feinman combines a

unique perspective with the experienced eye of a scientist. He’s had a front row seat

to developments in the field of carbohydrates and metabolism over the past decades,

which has given him key insights into the latest research.”

—Nina Teicholz, author of The Big Fat Surprise

“With his usual cutting wit and verve, Dr. Feinman hacks through the tangled thicket

of nutritional science to show us how we got to the sorry state we find ourselves in

today, deploying his vast experience in biochemistry, nutrition, and the turgid mind¬

set of academia to demystify mainstream nutritional research and explain how it has

currently gone ofF the rails. With a sprinkling of humor and literary allusions, along

with a deep dive into some of the nutritional literature, he tells readers what they

need to know to shed excess body fat, lower blood sugar, and restore their health. A

must-read for anyone with a serious interest in health and nutrition.”

—Michael R. Eades, MD, author of Protein Power

“We have arrived at the most important nutritional crossroads in all of human history:

We must choose to either keep the dietary status quo despite every bit of evidence

that it has failed, or demand a change in order to salvage what is left of our collective

health. In Nutrition in Crisis, Dr. Richard Feinman has dropped the opening salvo

in what is sure to be a battle for the ages—one that determines the future of how we

eat and the prevalence of disease in the years to come. Join the revolution!”

—Jimmy Moore, health podcaster; bestselling author of Keto Clarity

“Every scientific discipline needs a Dr. Richard David Feinman. He is just as skeptical

about what he believes as what he disbelieves. His writing bears eloquent testimony

to why he is such a scientific treasure.”

—Timothy Noakes, MD, PhD, emeritus professor,

University of Cape Town; founder, The Noakes Foundation

As a physician who has worked with thousands of patients unraveling the mysteries

and mythology around diet and nutrition, it is refreshing and timely to come across

a sane voice among the misinformation perpetuated by a misaligned food industry.

We have moved far beyond the simple ‘calorie in, calorie out 7 mentality and can now

embrace biochemical individuality and updated research while avoiding the dogma

of certain dietary camps.”

—Nasha Winters, ND,FABNO,

author of The Metabolic Approach to Cancer

“I have devoted my medical career to normalizing blood sugars in people with diabe¬

tes. This is only possible with very low carbohydrate diets. Dr. Feinman explains

why low-carb is wise even for people without diabetes. This cant-put-it-down title

is the closest thing to a complete, popular analysis of the biochemistry of human

nutrition that you will find and a superb lesson in how scientific studies have been

manipulated to prove fiction . 77

—Richard K. Bernstein, MD,FACN, FACE, CWS,

author of Dr Bernsteins Diabetes Solution and The Diabetes Diet

“Professor Feinman has been a leading pioneer in educating the academic community

and lay public on the health benefits of low-carb and ketogenic diets for years. In

addition to being a fantastic comprehensive review of practical nutritional biochem¬

istry, his book also tells an informative and entertaining story about what went wrong

in the nutrition establishment and advocates for a rational solution to the problem.”

—Dominic D'Agostino, PhD, leading scientist

on ketogenic metabolic therapies

“Low-carb? Low-fat? Count your calories? If you have whiplash from trying to follow

contradictory health headlines, you’re not alone. WeVe gotten ourselves into quite a

nutritional mess thanks to poorly done research that was even more poorly reported

to the public. With his trademark wry humor, Dr. Feinman exposes just how shoddy

much of the research has been and outlines how to take control of your own health.”

—Amy Berger, MS, CNS, NTP,

author of The Alzheimer s Antidote

“A distinctively insightful treatise on the sad state of affairs in nutrition written by

one of the true thought leaders. In his unique Brooklyn style, teacher and intellectual

Dr. Richard Feinman fights against ignorance by enlightening people on the real

science and application of nutrition.”

—Jeff S. Volek, professor,

Department of Human Sciences, Ohio State University

“I’ve had the honor and privilege of knowing and working with Dr. Feinman for many

years. IVe watched him employ his deep knowledge of biochemistry and metabolism

to tackle the entrenched, misguided nutritional advice cemented by thirty-five years

of the food pyramid. His book says it all.”

—Eugene Fine, professor, Department of Radiology,

Albert Einstein College of Medicine

“Dr. Feinman has spent his professional life dedicated to exposing the many flaws

found in most reporting in nutrition science. Nutrition in Crisis walks the reader

through the fog of dubious statistical analysis and into the light of day. In doing so

Dr. Feinman examines the science behind the low-carb debate, using humor and

metaphor to engage the reader in the process.”

—Miriam Kalamian,EdM,MS,CNS,

author of Keto for Cancer

“Making the lifestyle and nutritional changes necessary for lasting health is difficult.

Nutrition in Crisis supplies the intellectual imperative to take control of your own

health without resorting to symptom-suppressing medication. Read it, learn what

you need to do, and you will be mentally empowered and physically fortified for life!”

—Dr. Sarah Myhill, author of Sustainable Medicine and

Diagnosis and Treatment of Chronic Fatigue Syndrome and Myalgic Encephalitis

NUTRITION

in CRISIS

Flawed Studies,

Misleading Advice,

and the Real Science of

Human Metabolism

Richard David Feinman, PhD

CHELSEA GREEN PUBLISHING

White River Junction, Vermont

London, UK

DISCLAIMER: The information presented in this book is provided as an information

resource only and is not to be used or relied on for any diagnostic or treatment purposes.

Please consult your health care team before implementing any strategies herein.

No part of this book may be transmitted or reproduced in any form by any

means without permission in writing from the publisher.

Originally published by NMS Press and Duck in a Boat LLC in 2014 as

The World Turned Upside Down: The Second Low-Carbohydrate Revolution .

This edition published by Chelsea Green Publishing, 2019.

Acquisitions Editor: Makenna Goodman

Developmental Editor: Nick Kaye

Copy Editor: Deborah ITeimann

Proofreader: Beth Kanell

Indexer: Linda Hallinger

Designer: Melissa Jacobson

Printed in Canada.

First printing February, 2019.

10 9 8 7 6 5 4 3 2 1 19 20 21 22 23

Our Commitment to Green Publishing

Chelsea Green sees publishing as a tool for cultural change and ecological stewardship. We strive to align our book

manufacturing practices with our editorial mission and to reduce the impact of our business enterprise in the environ¬

ment. We print our books and catalogs on chlorine-free recycled paper, using vegetable-based inks whenever possible.

This book may cost slightly more because it was printed on paper that contains recycled fiber, and we hope you'll agree

that it’s worth it. Chelsea Green is a member of the Green Press Initiative (www.greenpressinitiative.org), a nonprofit

coalition of publishers, manufacturers, and authors working to protect the worlds endangered forests and conserve

natural resources. Nutrition in Crisis was printed on paper supplied by Marquis that contains 100% postconsumer

recycled fiber.

Library of Congress Cataloging-in-Publication Data

Names: Feinman, Richard D., 1940- author.

Title: Nutrition in crisis : flawed studies, misleading advice, and the real science of human metabolism / Richard

David Feinman, PhD.

Description: White River Junction, Vermont: Chelsea Green Publishing, 2019.

Identifiers: LCCN 20180484311 ISBN 9781603588195 (paperback) | ISBN 9781603588201 (ebook)

Subjects: LCSH: Nutrition—Popular works. | Diet—Popular works. | Health—Popular works. | Diet in disease—Popular

works. | Metabolism—Popular works. | BISAC: HEALTH & FITNESS / Nutrition. | MEDICAL / Endocrinology

& Metabolism. | HEALTH Sc FITNESS / Diets. | HEALTH Sc FITNESS / Weight Loss. | HEALTH Sc

FITNESS / Diseases / Diabetes. | HEALTH Sc FITNESS / Diseases / Cancer. | MEDICAL / History.

Classification: LCC RA784 .F45 2019 | DDC 613.2—dc23

LC record available at https://lccn.loc.gov/2018048431

Chelsea Green Publishing

85 North Main Street, Suite 120

White River Junction, VT 05001

(802)295-6300

www.chelseagreen.com

RECYCLED

Paper made from

„ recycled material

TTSG

www.fsc.org FSC* Cl03567

The book is dedicated to the memory of my father,

Max L. Feinman, MD,

who taught me about science and about honesty

and how much they were the same thing.

CONTENTS

Introduction 1

Part 1: Setting the Stage

1. Handling the Crisis: The Summary in Advance 19

2. Whaddaya Know? 35

3. The First Low-Carbohydrate Revolution 61

Part 2: Nutrition and Metabolism

4. Basic Nutrition: Macronutrients 73

5. An Introduction to Metabolism 91

6. Sugar, Fructose, and Fructophobia 105

7. Saturated Fat:

On Your Plate or in Your Blood? 117

8. Hunger: What It Is, What to Do About It 123

9. Beyond “A Calorie Is a Calorie”:

An Introduction to Thermodynamics 131

Part 3: The Low-Carbohydrate Diet for Disease

10. Diabetes 151

11. Metabolic Syndrome: The Big Pitch 167

Part 4: The Mess in Nutritional Science

12. The Medical Literature:

A Guide to Flawed Studies 175

13. Observational Studies, Association, Causality 185

14. Red Meat and the New Puritans 197

15. The Seventh Egg:

When Studies Defy Common Sense 215

16. Intention-to-Treat:

What It Is and Why You Should Care 219

17. The Fiend That Lies Like Truth 233

Part 5: The Second Low-Carbohydrate

Revolution

18. Nutrition in Crisis 245

19. Cancer: A New Frontier for Low-Carbohydrate 253

20. The Future of Nutrition 261

Acknowledgments 265

Notes 267

Index 275

Introduction

I Ve always had a weight problem. I would rarely have been considered

fat, but I was always trying to lose weight. When I was eight years

old, I wanted to get thinner so I could look sharp in my Brooklyn

Dodgers uniform to impress Barbara Levy, who was the most beautiful

girl in the world as far as I was concerned. I don't recall having any great

success, and it was only fairly recently that I found out that Barbara Levy

is now Barbara Boxer, former senator from California. In any case, I always

knew that starch made me fat—oddly, I was less afraid of sugar because

I mistakenly believed that there wasn't that much in Coca-Cola and the

other sodas I drank. I grew up with what is usually called a poor self-

image, and as the old joke goes, inside of every Botero is a Giacometti

trying to get out.

I knew from early on that it was important to cut out starch and obvi¬

ous solid sugar, and I made other observations about diet—for example,

that cold cereal for breakfast made me slightly sick. Its difficult for me

to remember exactly what I did eat in the morning. At least some of

the time it was bacon and eggs, which, in those days, was just one of

the things that people ate. Nobody recoiled in horror at bacon. The only

dietary advice at the time was to eat from the different food groups, which

were represented by a pie chart with unique symbols in each slice. The

bottle of milk was one that stuck in my mind. I felt early on that it was not

interesting, and I was sure that I didn't need an “expert” to tell me what

to eat. When the USDA food pyramid was introduced, I knew it was a

crock and I assumed that others did, too. My principles were simple: If

you have a weight problem, bread will make you fat, and if you don't have

a weight problem, why do you need the USDA? I thought everybody

was in agreement on that, but obviously that wasn't the case. I'm not sure

why people went along with all the “expert” advice. After all, everybody

has a great deal of experience with food. We all do three experiments in

“nutritional science” each day.

Nutrition in Crisis

People's compliance with dietary standards probably has to do with the

history of medicine. Among the turning points in that history was the

discovery of vitamins. Unlike poisons and microorganisms, vitamins were

stuff that you had to take if you didn't want to get sick. Another inflection

point was the identification of cigarette smoke as a causal agent in lung

disease. In that case, even though there was a toxic agent, the associations

were subtle and one needed statistics or other expert insights to see the

connection. This subtlety might have given people the idea that there were

experts who could see harm where they couldn't.

In my youth, I simply ignored the “expert" advice. I thought that I knew

what to eat (I was mostly right), and I saw obesity as a personal rather than

professional question. Decades later, when I began teaching metabolism, I

had to confront the interaction between science and nutrition. It proved to

be more difficult than I would have guessed.

This book is the story of my encounter with the world of nutrition, a

story of the science of biochemistry and metabolism—how you process

the food that you eat. It is about the application of science to daily life,

which is what I like about the subject. If you know a little chemistry,

you can appreciate the way that human evolution has reached into the

mixing pot of chemical reactions to obtain energy from the environ¬

ment, and even if you don't know chemistry, you can see the beauty in

the life machine.

But there is another side to the story. In the contentious and continually

changing stories of nutrition in the media, I encountered a discouraging

example of the limitations of human behavior in facing truth and prevent¬

ing harm. The story of nutrition has proved to be an almost unbelievable

tale of poor and irresponsible science within the medical community, one

of the most respected parts of our society.

However hard it is for scientists to distrust experts, it is even harder

for the general population. I was astounded when I saw a question on

an online diabetes site that said, “My morning oatmeal spikes my blood

glucose. How much carbohydrate should I have?” People with diabetes

cannot adequately metabolize dietary carbohydrate (starch and sugar) so

it seemed like an easy question. The answer from the experts, however, was

waffling and tedious, and it didn't include the obvious advice: “Limit your

oatmeal consumption to a level that doesn't spike your blood glucose.”

Introduction

Chemistry

When I was eight years old, my father taught me about atoms. I have one

of those memories that might or might not be accurate: I am sitting in my

father’s car, and he is telling me that the whole world is made of atoms

in the same way that the apartment building across the street is made of

bricks. Whether or not the scene really took place, it was a major influence,

and chemistry has long been a defining feature of my life. (Other vivid

memories of my early life in Brooklyn—being at Ebbetts Field and seeing

Jackie Robinson hit an inside-the-ballpark home run—turned out not to

be true. He had hit only one, in 1948, before I had ever seen a live game).

The crux of atomic theory, the thing that captures everybody’s imagina¬

tion when they are first exposed to it, is that it is a global and absolute

theory—it explains everything that has been done in the laboratory, the

kitchen, or anywhere else. Various fields of chemistry get at that same sense

of universal understanding with differing degrees of intellectual rigor, but

eventually I recognized that biochemistry was a good place to be for a

young person who didn’t know what career he wanted to end up with.

You can do drug design or theoretical chemistry or animal behavior or

nutrition and still call yourself a biochemist.

Teaching Nutrition

I have worked in a number of fields in biochemistry, but it was teaching

metabolism to medical students at SUNY Downstate Medical Center that

led to my professional interest in nutrition. Metabolism is the study of the

way food is processed and of the biochemical reactions that control life

functions. It is a fairly complicated subject—those parts that we understand

at all. Because there are so many individual biochemical reactions, students

tend to see the subject in the same way that somebody once described

the study of history: just one damn thing after another. There are general

principles and big concepts, of course, but you do have to know the details.

When I began teaching metabolism, I used the low-carbohydrate

diet—at the time primarily a weight-loss diet—as a central element in

my teaching. Control of blood glucose and insulin, the hormone whose

release is controlled by glucose, is central to many different processes in

Nutrition in Crisis

biochemistry. In the complicated network of biochemical reactions, insulin

stands out as a major point of regulation. The ups and downs of insulin are

what we try to control through the use of dietary carbohydrate restriction

as a therapeutic method. So, low-carbohydrate diets provided some unify¬

ing theme in teaching. I still teach metabolism in this way, though I now

emphasize diabetes where impaired ability to handle dietary carbohydrate

is the salient feature. The low-carbohydrate diet and its more thorough

form, the ketogenic diet, are popular—periodically very popular—and

while they remain controversial, the number of adherents, and possibly the

desperation of the detractors, suggests that low-carbohydrate must inevi¬

tably be accepted as I will describe it: the “default” diet for diabetes (the

one to try first) and the best diet for weight loss for many people. It is likely

that its current popularity can't be turned back. Myself and others who use

this teaching method have published papers about how understanding the

real-world benefits of low-carbohydrate—getting control of your health

and regulating your weight—can help you learn chemistry. 1

Around 2000, one of our second-career medical students who had been

a dietitian suggested that we include formal nutrition in our biochemistry

course, and she provided subject matter from standard nutritional practice.

I cannot really describe what it was about—probably, even at the time,

it was so vacuous that I couldn't retain it in my memory. In any case, I

objected because whatever it was, it wasn't biochemistry. The way I saw it,

criticizing how lectures are given is like complaining about how the dishes

are done: Everybody sees an immediate solution. Despite my protests, I

wound up having to give formal lectures in nutrition, and I really didn't

know the literature. I had long ago found that carbohydrate restriction was

best for me, and while low-carbohydrate diets provided me with a good

framework for teaching metabolism, applied therapies do not always have

a close relation to theories, so teaching nutrition required a certain amount

of background study on my part.

My first lectures on nutrition were neutral. I simply tried to cover

the basic aspects of low-carbohydrate and low-fat diets—the two main

choices, really—presenting the pros and cons of each approach in a simple

way. Low-fat diets are not based substantially on biochemical mechanisms;

instead, they follow from observed correlations between cardiovascu¬

lar disease and the presence of cholesterol or other lipids in the blood.

Introduction

More recently, low-fat has morphed into a prescription for obesity, and

proponents have started emphasizing the point that fat has more calories

per gram than other things, peddling the idea that the more calories, the

greater the effect on body weight—the ill-conceived idea that “you are

what you eat,” which hangs over everything. While I could explain at that

time how metabolism, and specifically the role of the hormone insulin,

accounted for the benefits of a low-carb diet, I could not provide a well-

organized review of the relevant studies in the medical literature. So, my

initial lectures were rather simple and straightforward while I tried to get

a grip on the scientific literature.

As I dug into that literature, however, it didn’t take long to see that

something was terribly wrong. In simply trying to grasp the facts, I had

stepped into a world of bad science, self-deception, and a scandal equal to

any in the history of medicine.

The Nurses’ Health Study

Science is very specialized. Although I had been doing research on blood

coagulation, which is related to cardiovascular disease (CVD), I did not

pay much attention to the diet-heart hypothesis—the idea that fat and

cholesterol in the diet raise blood cholesterol, which, in turn, leads to CVD.

I was suspicious of such a theory, though, because biology tends to run on

hormones and enzymes—that is to say, on control mechanisms rather than

on mass action (the principle that chemical processes are determined by

how much reactants are put into them). The grand principle in biochem¬

istry is that there is hardly anything that is not connected with feedback.

If you try to lower your dietary cholesterol, for example, your liver will

respond by making more. Simply adding more or less is not guaranteed

to produce much change at all, once feedback is taken into account. I was

therefore skeptical, if not well-informed.

Whatever my misgivings about the diet-heart hypothesis, I didn’t ques¬

tion it very deeply at first. However, when I went back to the original

literature to find the evidence supporting low-fat recommendations,

as I had to do in preparation for my lectures, it was a rude awakening.

My assumption that there was at least a grain of truth in the diet-heart

hypothesis turned out to be overly optimistic. If the hypothesis is not a

Nutrition in Crisis

Replace Lower Risk Higher Risk

Saturated fat ._

-4 fco/_ nf onprnvl

with carbohydrates

l /o ui cl ici y yj

Monounsaturated fat

with carbohydrates

Polyunsaturated fat

with carbohydrates

H- --- ■ - //—

Saturated fat with

monounsaturated fat

Saturated fat with

polyunsaturated fat

-1-1 1

1—-1-1-1- I 1 ' 1

-SO -60 -40 -20 0 20 40 60 80

Change in Risk [%)

Figure 0.1. Estimated changes in risk of coronary heart disease associated with isoca¬

loric substitutions [error bars show 95 percent confidence interval]. Adapted from F.

B. Hu et al„ “Dietary Fat Intake and the Risk of Coronary Heart Disease in Women," New

England Journal of Medicine 337, no. 21 [1997]: 1491-1499.

total sham, it is pretty close. One of the first papers that I came across in

my literature survey was a report from the Nurses' Health Study (NHS).

Centered at Harvard, the NHS is one of the largest epidemiological stud¬

ies with more than one hundred thousand participants. It has produced a

large number of studies on nutrition and other aspects of lifestyle.

Walter Willett, head of the NHS and follow-up studies, and his associ¬

ate Frank Hu examined the association of different kinds of fat, as well

as carbohydrate, with the risk of CVD. 2 I found the result astounding.

Figure 0.1, redrawn from their paper, shows the effects of substituting one

type of fat for another, and of substituting carbohydrate for fat. Replacing

saturated fat with either polyunsaturated fat (vegetable oils) or monoun-

saturated fat (olive or canola oil) reduced risk substantially. That's what

the nutritional community had been saying, so I saw no surprise there.

Introduction

However, when the polyunsaturated fat was replaced with carbohydrate,

Hu et al. found an average 60 percent increase in risk. What? Carbohydrate

is worse than fat for cardiovascular risk? Thats not how it was supposed to

be. What about saturated fat? Surely that’s a bad guy. Replacing saturated

fat with carbohydrate did provide some benefit according to the figure, at

least on average, but there is more to the story. In this kind of figure, the

error bars (horizontal lines) show the spread of individual values, which

was quite large in this case. In other words, even though there was an

average improvement from replacing saturated fat with monounsaturated

fat—their main selling point—some subjects experienced greater benefit

than the average, and some much less benefit than the average. When satu¬

rated fat was replaced by carbohydrate, some of the study’s subjects were, in

fact, going in the opposite direction—that is, they were at greater risk for

CVD, which contradicted the supposed benefit of reducing fat. It wasn’t

just a few subjects, either. The breakdown was about 60:40: 60 percent

of subjects experienced reduced risk of CVD by replacing saturated fat

with carbohydrate, and 40 percent experienced greater risk. It gets worse.

Without getting too caught up in the statistical details, the rule is that if

the (horizontal) error bar crosses the zero line, then there is no significant

effect of the substitution. So, based on the study’s findings, substituting

carbohydrate in place of saturated fat is at best neutral, or more precisely, it

is as likely to increase risk as it is to lower it. The same is true of substitut¬

ing carbohydrate for monounsaturated fat.

Looking at figure 0.1, it is hard to see a risk of fat, but hasn’t risk from

fat been the message all along? Certainly the idea that carbohydrate is a

risk is not found in the media or the pronouncements of health agencies.

And then there is the authors’ summary of the paper:

Our data provide evidence in support of the hypothesis that a

higher dietary intake of saturated fat . . . is associated with an

increased risk of coronary disease, whereas a higher intake of

monounsaturated and polyunsaturated fats is associated with

reduced risk. These findings reinforce evidence from metabolic

studies that replacing saturated fat . . . with un-hydrogenated

monounsaturated and polyunsaturated fats favorably alters the

lipid profile, but that reducing overall fat intake has little effect.

Nutrition in Crisis

This conclusion is not accurate. It’s at best misleading, and at worst

outright deceptive. The measured risk of saturated fat intake was dependent

on what it was replaced with, yet there is no mention of carbohydrate as a

replacement. The most striking thing to me was that if you looked at the

risk from carbohydrate in comparison to the risk from saturated fat—that

is, the risk of substituting one for the other—there was no difference. Even

worse, substituting carbohydrate for other fats increased risk. How could

this be? Fat out. Carbohydrate in. Wasn’t that the clear recommendation

for improved health from just about every health agency and expert? Yet

the data said it didn’t matter. Was it dishonest not to make this clear in

the discussion section of the paper? At best, it was an error of omission.

The authors from the Harvard School of Public Health were, and still are,

the more modest among those vilifying fat, insisting that it is only the

type of fat that we need worry about. Most recently, the American Heart

Association (AHA) has come around to the same point of view as if they

had just discovered it. I was probably not alone, but I began using the

term lipophobes long ago for proponents of low-fat. It’s a wiseguy term, and

since I was still something of an inside player in the nutrition world, I was

reluctant to put it into print until Michael Pollan started using it without

any sense of irony. 3 (I started saying “... as Michael Pollan calls them.”)

By the time I read the NHS paper, my professional involvement in the

field of nutrition was cemented. I did not, however, adequately attend to

the sense of being sucked into a whirlpool from which it would be hard to

escape. The data supporting low-carbohydrate were there for everyone to see,

I thought, even if the authors had chosen to downplay the strongest result.

A trip to the supermarket today demonstrates that the results from

the Nurses’ Health Study had little effect. The low-fat story is still with

us. More striking is that two meta-analyses (averages of many studies)

came to the same conclusion regarding the ineffectiveness of replacing fat

with carbohydrate. Siri-Tarino et al. concluded that “there are few epide¬

miologic or clinical trial data to support a benefit of replacing saturated

fat with carbohydrate,” and in March of 2014, yet another meta-analysis

found similar results. 4 What turned out to be most remarkable about all of

these studies was that they presented a reanalysis of studies that had found

no effect of saturated fat to begin with. One has to ask why the results

were not accepted when first published. Some of the included studies are

Introduction

twenty years old. How is it possible that, in the most scientific period in

history our society runs on incorrect scientific information? That's one of

the questions that I will try to answer in this book—or at least describe, as

Fm not sure that there is a clear-cut answer. Looking ahead, I will intro¬

duce the revolutionary idea that, except in cases of well-defined genetic

abnormalities, there is no predictable effect of diet on heart disease based

on the current research. It is a hypothesis, and we might know more as we

understand the genetics, but no effect is certainly more plausible than the

diet-heart hypothesis, which remains only a conjecture without experi¬

mental support. This lack of effect will be one of the themes in this book

and one of the battlegrounds as the crisis in nutrition plays out.

About This Book: Who It’s for and Why I Wrote It

Food and chemistry have been two of the largest influences in my life.

The beauty of biochemistry is that it relates the movement of electrons to

whats on our plates—and this is a connection I thought I could explain. I

like writing about biochemistry. It allows us to see how things fit together,

but it also exposes the things we don't know—the things that evoke within

us curiosity, the defining feature of the scientific life. If you want to indulge

that curiosity, you are the person I had in mind when I started the book.

This is a book for scientists. Not specifically for people with atomic-

force microscopes in their labs, but for those who want to look at nutrition

from a scientific point of view. Science is less about sophisticated measure¬

ments than it is about basic honesty. It is true that scientific fields can be

very mathematical or intellectually rigorous, but all sciences, even those as

complex as quantum mechanics, are tied to logic and common sense, and

are frequently directly accessible to lay audiences. Part of the game, most

researchers understand, is to make the results easy to understand. Einstein

is widely quoted as having said that we want to make it simple but not

too simple. Modern medicine, despite its reliance on technology, explicitly

accepts an obligation to explain things logically to the patient. It doesn't

always fulfill this obligation well, but the goal remains nonetheless. In this

book I will try to fill some of the gaps and define words, but for the tough

spots, you will have to read like a scientist. How do we read? We re all

specialists so most of us cant read technical articles, even those in our own

Nutrition in Crisis

fields, without some bumps. Skip over the bumps and see if you can get

the big picture. You can always come back to them later, and many of the

details are just a Google away.

If you write a book about biochemistry, it’s about chemistry, but if you

write a book about nutrition, it’s about everything. Not every chapter in

this book is for everybody. I have tried to provide a continuous, easy-to-

follow thread, but different subjects require different kinds of discussion,

and some of these discussions are necessarily technical. You can skip them,

but I do suggest giving them a shot.

Although this is primarily a book about the science of nutrition, you

can’t escape the sociology and politics of medicine. Establishment medi¬

cal journals, private organizations, and government health agencies have

insisted on low-fat, low-calorie dogma despite the scientific evidence to

the contrary. This politically motivated breakdown in scientific practice is

deeply discouraging to me and was an additional motivation for my writing

this book. The corruption of science goes beyond principle, too what s at

stake is the health of patients.

The breakthrough in understanding metabolism that underlies much of this

book comes from the realization that many superficially unrelated patho¬

logic or disease states and associated conditions are intimately connected

at the physiological and biological level. Equally important, control of

these states rests in a major way with diet. Diabetes, obesity, cardiovascular

disease, states of hypertension, and numerous other physiologic states are

all tied together. The promise is that, examined together, they might provide

a global theory of metabolism and with it a common cure.

A major focus of this book is the concept of metabolic syndrome, which

is a collection of clinical markers—including overweight, high blood pres¬

sure, and the so-called atherogenic dyslipidemia (the lipid markers that

are assumed to contribute to cardiovascular disease)—that together and

in combination indicate risk of disease. The identification of metabolic

syndrome constitutes, in my view, a great intellectual insight. That the

common effector is likely the hormone insulin points to the importance of

controlling dietary carbohydrate, the major stimulus for insulin secretion.

The resistance of the medical profession to dietary carbohydrate restric¬

tion in the treatment of metabolic syndrome, and even more obviously,

Introduction

in the treatment of diabetes, I find incomprehensible. Everybody knows

somebody with diabetes. Echoes of the early days in Brooklyn made it very

upsetting to see pictures of Jackie Robinson taken shortly before his death

from diabetes complications at age fifty-two. Because it is progressive, the

disease is an underappreciated source of suffering. Clinicians will tell you

that it is like cancer in its devastating effects. Diabetes is the major cause of

amputations after accidents and the major cause of acquired blindness.That

is a motivation for writing this book and why you might find it important.

This resistance is a scandal at the level of Ignaz Semmelweis, an

early-nineteenth-century Viennese physician. To reduce the incidence

of puerperal fever (infection after childbirth), Semmelweis suggested

that physicians wash their hands after performing autopsies and before

delivering babies. They refused; it was too much trouble. But that was the

nineteenth century, before the germ theory was established, and thats

some kind of excuse. It's hard to know how we will look back on the

actions of the American Diabetes Association (ADA), who believe that

for people with diabetes, “Sucrose-containing foods can be substituted for

other carbohydrates in the meal plan or, if added to the meal plan, covered

with insulin or other glucose-lowering medications.” 5

The most difficult part of writing this book was trying to understand—if

such a feat is even possible-—how the whole field of medical nutrition could

be wrong. Way wrong. Totally off the mark. As misguided as the alchemists'

pursuit of the creation of gold. This disconnect from true science is not only

bizarre; it is a source of real harm to patients.The phenomenon is particularly

hard to explain because the widespread respect for the medical professionals

is based on real accomplishment and expertise, and it is hard to see why they

would go so wrong in nutrition. For me, having precedents makes it easier

to grasp, if not completely comprehensible. Here's one example of self-

deception and refusal to accept evidence that I keep in mind. The following

is a passage from Abraham Rabinovitch's writing on the Israel Defense

Forces and intelligence in the days before the Yom Kippur War (1973):

The intelligence chiefs believed they knew a deeper truth ... that

rendered irrelevant all the cries of alarm going up around them.

Zeira and his chief aides were to demonstrate the ability of even

brilliant men to adhere to an idee fixe in the face of mountains

Nutrition in Crisis

of contrary evidence.... They clung to their view even though

the Egyptian deceptions were contradicted by evidence of war

preparations that AMAN's [military intelligence] own depart¬

ments were daily gathering. . . . But the deception succeeded

beyond even Egypt's expectations because it triggered within

Israel's intelligence arm and senior command a monumental

capacity for self-deception. 6

The Israelis could have lost it all. They could have lost the whole country

due to their refusal to accept the evidence.They were largely saved, however,

by a couple of field commanders who were wild and crazy guys most

notably Ariel Sharon, who attacked an Egyptian emplacement by disobey¬

ing orders not to cross the Nile. Audacity and the refusal to follow orders

might be what save nutrition as well.

Finally, this book is for the person (and those for whom she spoke) who

posted on my blog wondering how she could determine which nutritional

studies are flawed and which are not, especially at a time when we are

inundated with so many conflicting recommendations. “Where are the

true studies that are NOT flawed,” she wrote, “and how do I differentiate?

She was right to be suspicious. It is not always easy. There are so many

nutrition papers that try to snow you with technical detail, and those are in

fact the ones to be most suspicious of. Scientific papers will necessarily have

technical components, but researchers shouldn't be making their results

more difficult to understand than they need to be—and some of the papers

are simply not true. Most researchers know that if you make up the data on

a federally funded grant, you can go to jail, but when it comes to interpret¬

ing the data, they can say just about any damned thing. In this book, I

explain how to interpret nutrition papers. In particular, I explain what the

statistics mean, how they can be misused, and how to navigate the literature

as someone who doesn't necessarily have a background in statistics.

The Second Low-Carbohydrate Revolution

The killer app, so to speak, of the low-carbohydrate diet is still the treat¬

ment of diabetes. Intuitively obvious, proved in many experimental

trials, and widely used anecdotally and clinically, there are virtually no

Introduction

contraindications. Resistance to its use appears to be spurred on entirely

by pressure from political organizations, primarily the American Diabetes

Association (ADA), which, looking for a way to save face, still refuses to

endorse low-carbohydrate strategies. Many identify the influence, direct

or indirect, of drug companies and food companies as a culprit as well.

Whatever the motivation, not encouraging physicians to at least offer

carbohydrate restriction is seen by many who have had success with the

approach as “criminal.” The latest guidelines from the ADA emphasize

“individualization,” presumably as a way of softening their previous oppo¬

sition to low-carbohydrate. The word individualization is used twenty-one

times in their position paper, 7 but the actual principles to be applied for

each individual are not stated. The foolishness of not explicitly restricting

carbohydrate for people with a disease whose most salient manifesta¬

tion is an inability to adequately metabolize carbohydrate is astounding.

Individualization, in my view, can best be described as a cop-out.

Despite the resistance to low-carbohydrate, we have at the same time

a constant flow of blog posts and books that show the low-fat diet-heart

hypothesis for the intellectual and clinical disaster that it really is. The most

recent and most complete, a book called The Big Fat Surprise* is surprising

in its description of the depths of self-delusion, if not dishonesty, in keeping

low-fat alive. While the pace of criticism is increasing, these exposes docu¬

ment that the diet-heart hypothesis has been debunked since its inception.

If you step back and look at the data, the concerns, the voices on

Huffington Post, or the numerous blogs belonging to dietitians, it shines

through that the easiest way to lose weight is the low-carbohydrate diet.

The concerns, voiced for forty years, have never been effectively substanti¬

ated, and the real-world tests of carbohydrate restriction come out in its

favor. There are now dozens of successful implementations, though the

Atkins diet is still the best known, having attained a somewhat generic

status, like Kleenex.

Metabolic Syndrome: The Big Pitch

There is almost nothing in biology that is not connected with feedback.

This idea is fundamental yet widely ignored. Reducing dietary intake of

cholesterol will have limited effect because of compensatory synthesis.

Nutrition in Crisis

Likewise, if you stop eating carbohydrate, your body will respond by

synthesizing glucose and making other fuels available. This grand idea puts

severe limitations on what you can do (as in the case of cholesterol or, look¬

ing ahead, trying to starve tumors by reducing glucose), but it also points

to some opportunities. When you consume carbohydrate, the hormone

insulin turns off the feedback system in the liver that produces glucose

from glycogen or gluconeogenesis. Understanding that diabetes represents

a breakdown in this feedback system—that the liver of a person with type

2 diabetes will not respond to insulin (insulin resistance)—makes clear

why you should not add more insulin. Nonetheless, the compensatory

feedback response to many drugs and foods does call for caution in jump¬

ing to conclusions.

The key point is that there is a global effect of the hormone insulin.

We -can get very far simply by regulating this hormone. The role of

feedback is part of the picture, but the effects of manipulating insulin

can be highly predictable, which is the main theme of this book. Always

in the background of this discussion is metabolic syndrome (MetS).

Metabolic syndrome is rooted in the observation, generally credited to

Gerald Reaven, an endocrinologist who died recently, that a collection

of seemingly different physiologic effects—overweight, high blood pres¬

sure, high blood glucose, high insulin, and the collection of blood lipid

markers referred to as atherogenic dyslipidemia (high triglycerides, low

HDL)—are all tied together by a common causal thread: disruption in

the metabolic response to insulin. 9 The physiologic markers of MetS

predict progression to the associated disease states (obesity, diabetes,

hypertension, and cardiovascular disease), all of which respond to dietary

carbohydrate restriction. That is the big pitch. This observation confirms

that it really, is a syndrome (since it has a common underlying cause)

and simultaneously points us to the most effective treatment. No dietary

approach is better than low-carbohydrate and no drug will target all of

the markers together.

There are, in fact, critics of MetS who question the practical significance

of the syndrome. What they’re really suggesting is that the effects have to

be treated with a collection of drugs: drugs for diabetes, drugs for heart

disease, drugs for high blood pressure. A low-carbohydrate diet, which is

already widely accepted as effective for weight loss, is likely the strategy

Introduction

to treat the different facets of MetS without using this cocktail of drugs.

Acceptance of such a notion is the goal of the revolution.

Oddly enough, the bright light on the horizon is the ketogenic diet for

cancer. I say “oddly” because carbohydrate restriction for diabetes is already

a slam-dunk, and should have been the crystallizing point for change. Of

course, as I’ve already discussed, the resistance to low-carbohydrate for

people with diabetes has been extensive. Somehow treatment of cancer

is not encountering the same obstinacy, despite limited research on the

subject. In chapter 19 I describe work by my colleague Dr. Eugene J. Fine,

targeting insulin in the treatment of cancer. I see this study, despite its

small size, as a reason for hope and a sign of future progress. 10 If it turns

out that we learn to treat diabetes by learning to treat cancer, it would not

be the strangest thing that ever happened in science.

How to Approach This Book

If we had ham, we could have ham and eggs , if we had eggs,

—Old American idiom

This book represents the view of a practicing biochemist, and as such, it

approaches nutrition as applied biochemistry. Biochemistry is not all there

is to nutrition, but it represents a more scientific and logical perspective

than “you are what you eat”; it tells us instead that we are what our metab¬

olism does with what we eat. I will explain why low-carbohydrate diets are

the default diet (the one to try first) for diabetes and metabolic syndrome

and why you need to understand this idea even if you are not suffering

from either condition. To be clear, I am not an advocate of anything. In the

end, you have to be your own nutritionist. My job is to give you some tools

for sorting out the army of nutritional “experts” out there and the studies

they produce.

I am not an expert on politics, but, as in the aphorism most often attrib¬

uted to.John Adams (I think he stole it from the ancients), I study politics so

that my children can study biochemistry and nutrition. It’s all tied together.

The science is not divorced from the politics. The Framingham Study, 11

the first very large population trial, tested not only a scientific principle—

whether dietary fat and cholesterol were related to risk of cardiovascular

Nutrition in Crisis

disease (they were not)—but also whether the recommendations of health

agencies were a rush to judgment (they were). The study had such a large

political component that, as striking as the scientific outcome was—there

was no effect of dietary total or saturated fat or cholesterol on cardiovas¬

cular disease—it couldn’t be seen to fail. The results were buried for years

until a statistician rediscovered them and finally had them published. This

intertwining of the politics and science is a persistent pattern, and I try

to tell both sides of the story and explain how they connect to each other.

My main principle, however, is that basic science comes first. I give

preference to the demonstrations in nutritional and medical practice that

are based in the fundamentals of biochemistry, of hard science. Big clinical

trials have to be judged on their inherent strength, but, if they contradict

basic science, the authors have an obligation to explain why. And science is

continuous with common sense. It doesn’t matter how many statistical tests

you run: If the results violate common sense, it is unlikely to be science.

The poor research published by prestigious individuals and institutions

suggests the nature of science itself has to be investigated. To do this, we

will have to define scientific principles, explain how to read a scientific

paper, and determine whether peer review has done its job. But first, in

chapter 1,1 give you the bottom line—the practical consequences of the

science. The rest of the book will serve to justify these statements and

recommendations.

PART 1

Setting the Stage

-CHAPTER 1-

Handling the Crisis

The Summary in Advance

W hat should I eat? This question invariably comes up during

my lectures to medical students, during presentations at

conferences, and even in private conversations about scientific

experiments. Depending on my audience, I will go into different levels of

technical detail, but the bottom line is always the same: The best diet is the

one that works. Any expert can tell you how his or her diet theoretically

conforms to widely accepted science, and how it is “healthy,” “moderate,”

and a “bargain”—but if it doesn’t work in practice, if you don’t lose weight,

if your blood sugar doesn’t go down, then it’s no good at all.

There is little evidence that the diets recommended by government

and private health agencies have provided much help for the current

epidemic of obesity and diabetes, but they keep pushing them anyway.

Defenders usually tell you that it is because people are not really

following the guidelines. What they don’t say is how they know the

recommendations are good if nobody follows them. So here I’ll give you

three basic rules of nutrition that I propose as a guide, and I’ll show you a

few principles that will help you follow them. My recommendations are

likely different from what your doctor has told you. The rest of the book

will justify these principles.

The Three Rules

The following are three simple, fundamental principles for getting control

of your diet:

Rule 1. If you’re OK, you’re OK.

Nutrition in Crisis

Rule 2. If you want to lose weight, don’t eat. If you have to eat,

don’t eat carbs.

Rule 3. If you have diabetes or metabolic syndrome, carbohydrate

restriction is the “default” approach—that is, the one to try first.

RULE 1. If you don’t have a weight problem, if you feel OK, if you

are healthy, and if you don’t have a family history of disease, there is

no compelling reason to change your diet. Rule 1 is actually surprising

to many people. You might want to find out more about nutrition and

biochemistry, but the idea that there was once a Garden of Eden diet that

we all ate until somebody first brought high-fructose corn syrup into the

world, along with all our woe, seems unlikely.

Not everybody has this view. There is the idea, not always stated explic¬

itly, that, analogous to Freud’s The Psychopathology of Everyday Life , we are

all doing something wrong and that life is a continuous battle between

what our bodies really need and the pressures of civilization. It’s not like

that. We evolved to be adaptable. Lots of dietary approaches work and,

when it comes down to it, none of us are going to get out of this alive.

And then there are the Dietary Guidelines for Americans from the

USDA. Congress specifically charges the department with providing

advice to people who are healthy—that is, people who don’t need advice. In

Brooklyn, we call this fixing something that ain’t broke. Like psychoana¬

lysts, these government nutritionists feel endowed by their Creator with

the intuitive power to penetrate unspoken, unmeasured, deep truths, and

believe that they are able to tell us that we are not eating the right thing and

that we are at risk for some future disease. They are, however, quick to take

offense if you suggest that their inability to control the epidemic of obesity

and diabetes makes it very unlikely that they know what the right thing is.

RULE 2. If you want to lose weight, don’t eat. If you have to eat, don’t

eat carbs. I first said this as a joke at a conference, but there is a great deal of

truth in it. This is not to say that starvation is a good long-term strategy—

the danger is that you will lose muscle mass along with your fat but too

frequently we think that it is important to eat all the time. That is not true,

and intermittent fasting, which is garnering a certain amount of inter¬

est, might be a very useful strategy for weight loss. There are exceptions:

Some medical conditions, diabetes in particular, might require individual

Handling the Crisis

variation. (Calorie reduction is beneficial for diabetes, but the need to avoid

ups and downs means that there are other considerations.) The problem,

though, is that we tend to think that hunger is some kind of physiologic

signal telling us that our body needs food. We think that this signal must

be answered immediately. Its not like that. Chapter 8 explains how hunger

only means that you are in a situation where you are used to eating. This

situation, such as a business lunch or a tailgate party, might have little to

do with your state of nourishment. The hunger pangs that you feel might

be real enough, but you are not compelled to answer them—and they don't

even last that long.

Calories are a measure of the total energy available from burning food.

The less food you eat, the less energy you will have. Not all calories are the

same, though. Many experiments that show success with calorie reduction,

on closer analysis, reveal that the effect was due to the de facto reduc¬

tion in carbohydrate. Some dietary strategies will waste more energy than

others (as heat and other unproductive effects).These nuances make up the

second part of Rule 2, but the fundamental truth remains that if you don't

eat, you will get thin.

RULE 3. If you have diabetes or metabolic syndrome, carbohydrate

restriction is the “default” approach—that is, the one to try first. Almost

everyone is now within two degrees of separation of somebody who has

diabetes. Diabetes is a disease of carbohydrate intolerance. In type 1, there

is a substantially reduced or total inability of the pancreas to produce

insulin in response to carbohydrate. Type 2 is characterized by insulin

resistance—the inability of the body's cells to properly respond to the

insulin that is produced—as well as progressive deterioration of the insu¬

lin-producing cells in the pancreas.The defining symptom and major cause

of the pathology is high blood sugar. The idea for treatment is simple: If

diabetes is a disease of carbohydrate intolerance, it is reasonable to assume

that adding dietary carbohydrate would make things worse, and restricting

dietary carbohydrate would make things better. It makes sense, and this

expectation is generally borne out. There are people with diabetes who can

tolerate greater or lesser amounts of carbohydrate but, for most people,

this simple treatment works exactly as it's supposed to—and dietary carbo¬

hydrate restriction might even be best for those people who can tolerate

higher carbohydrate. There are no experimental or clinical data that show

Nutrition in Crisis

a contradiction. The fact that people with diabetes are not typically offered

a low-carbohydrate diet as the default treatment, let alone as an option at

all, is in my view a major scandal in the history of medicine.

Keep in mind that if you are on diabetes medication, you have to discuss

getting on a low-carbohydrate diet with your physician. Carbohydrate

restriction will lower blood glucose in the same way as many medications,

so a low-carbohydrate diet in conjunction with your current medication

regime might pose a risk of hypoglycemia (low blood sugar). Generally

your physician, if he or she has any experience with carbohydrate restric¬

tion, will reduce or eliminate your medication before putting you on a

low-carbohydrate diet. Some people see this as the single best argument

for carbohydrate restriction: In most diseases, reduction in medication is

considered a sign of success.

Doing It: Eat to the Meter

“Eat to the meter” is the principle used by people with diabetes. The

meter is the glucometer, which, as you can probably guess, measures blood

glucose. If the food you just ate causes a spike in blood sugar, it is a sign to

avoid that food. Simple as that. Oddly, diabetes educators might tell you

that if a food spikes your blood glucose, it means you need more insulin to

deal with that food—a truly misguided notion.

Those of us with a weight problem might sensibly eat to what I call

the ponderometer: the bathroom scale. If the experts tell you to eat more

whole grains, but the reading on the scale keeps going up, then stop!

Doing It:

The Best Exercise Is the One That You Do

The nutrition world can’t agree on much, but there are few who would

dispute that exercise is inherently beneficial. Exercise by itself is not

particularly effective for weight loss unless you’re a professional athlete or

in basic training, but it does enhance the benefits of dieting. Some of the

effects are vascular—that is, physiologic rather than biochemical so it is

hard to pin down the relation to nutrition, but finding any agreement in

this contentious field is a good thing.

Handling the Crisis

A Brief Primer on Carbohydrate Restriction

Calories count, but the advice by experts that only calories count

is wrong. There are great advantages to diet strategies that go

beyond calories. For most people the best strategy for weight

loss is to reduce carbohydrate intake. Insistence on the idea

that “a calorie is a calorie,” as it is usually stated, is one of the

crystallizing points of the nutritional crisis. The benefits of a low-

carbohydrate, higher-fat diet are due partly to the more satiating

effects of protein and fat, or, more precisely, the poor satiation

from most carbohydrate. Its important to keep in mind that

“a calorie is a calorie” is not true as a general principle; instead,

it is the macronutrient composition of the diet that affects the

amount of weight gained or lost. There are extensive data in the

medical literature to support the unique effects of carbohydrate

reduction, but the best evidence might be anecdotal. In the

field of weight loss, anecdotal evidence is actually fairly reliable.

Everybody knows somebody who lost a lot of weight on the

Atkins diet. Some report that the pounds seem to u melt away”

There is no guarantee that you will have the same experience,

and everybody hits a plateau, but it is still your best bet. Those

who give dire warnings about low-carb diets likely haven’t actu¬

ally recommended them to patients or tried them personally.

Low-carbohydrate and ketogenic diets follow* basic biochem¬

istry. The key factor is improved control of insulin, the anabolic

(building up) hormone that controls the major events in

metabolism—storing fat and carbohydrate as glycogen, encour-

aging protein synthesis—and insulin is most reliably controlled

by carbohydrate. Diets based on reduced carbohydrate are

consistently successful and that's why people keep using them

in the face of ^concerns” of the medical establishment. As in

any diet, people might quit at a certain point—we have six

hundred million years of evolution and a lifetime of behavioral

Nutrition in Crisis

conditioning telling us to eat anything that tastes good—but

even those who fall off the wagon will usually return to a low-

carbohydrate diet. Nutritionists will tell you that u yo-yo dieting”

has some risk, but there is no evidence for that, and most of us

are happy for any period where we are thin, however long it lasts.

Whatever your concerns about low-carbohydrate, for most

people, low-fat diets are even worse than doing nothing. The

hunger will simply be too much to deal with. Sure, if you can

get yourself into the frame of mind where you like the “lean

and hungry” feeling, then anything that reduces calories will be

okay. However, most people arerit able to deal with that feeling

and will not be able to lose any significant amount of weight on

a low-fat diet.

The scientific literature backs up the anecdotal evidence

in support of low-carbohydrate. Jeff Volek, one of the major

researchers in carbohydrate restriction, put this spin on it:

“Nutrition research is hard. Too many things change and its

easy to come up with nothing. When you study low-carb diets,

people lose weight. Put people on a low-carb diet and you get

real data.” As I finish this hook, the long-standing refusal of

the nutritional establishment to face the data is finally falling

away. It will be increasingly easy to follow" a low-carbohydrate

approach and to have the support of a physician in doing so.

Experimentally, carbohydrate restriction has better compliance

than anything else because it gives immediate results, possibly

due to the greater satiety of protein and fat, or, more likely,

due to the poor satiety of most carbohydrates. You can test for

yourself, at least at the level of perception.

Later chapters will describe experimental studies that

support these conclusions about low-carbohydrate strategies,

but there is one truly remarkable phenomenon that tells you

about the edge low-carbohydrate diets have in satiety When

diet comparisons are carried out experimentally, most often the

Handling the Crisis

protocol is to allow the low-carbohydrate group the freedom to

eat ad lib as long as they follow the restrictions on carbohydrate.

Low-fat diets, on the other hand, are restricted to a fixed number

of calories. The reason that the Atkins diet and other low-

carbohydrate approaches put no limit on consumption is that

fat and protein intake is self-limiting when carbohydrate is low.

Setting up the experiment this way, however, constitutes poor

experimental design because you are now testing two things;

the ability of a low-carbohydrate diet to limit caloric intake as

well as the proposed difference in physiologic effects of the type

of diet—the metabolic advantage. The low-carbohydrate diet

almost always wins in such comparisons, but because of the poor

experimental design, you can t see how much of the success can

be attributed to the greater satiety of low-carbohydrate diets as

opposed to the greater weight loss per calorie. The results do tell

you one thing, though: As advertised, you dont have to count

calories in a diet based on carbohydrate restriction.

Perhaps the greatest virtue of carbohydrate restriction rests

with its fail-safe feature. If you are not rapidly losing weight,

or if you seem to hit the wall, it is a way of eating that gives

you freedom from the sense of fighting a war against fat. You

will almost never gain weight and you will escape that over¬

bearing feeling that every meal is a battle. This feeling might

actually be the greatest threat of obesity. As a threat to health,

excess weight is probably exaggerated. Mortality correlates with

weight only at the very extremes. The major threat is the sense

of loss of control. I am not a health care provider, but I get

emails from executives, officers in the military, and other people

who hold dominion over their professional world and who have

trouble controlling their own body mass. The struggle can be a

tremendous burden and can take over entire lives. Cut out most

of the carbohydrates in your life, and you stand a chance of

getting rid of the burden.

Nutrition in Crisis

The best exercise, like the best diet, is the one that you can get yourself to

do regularly. For that reason, exercise is highly individualized. Proponents

of each type of exercise think that the others are not good, but the impor¬

tant thing is to find one that fits your style, or, more important, one that

you get in the habit of doing. I am personally a believer in slow burn (slow

repetition with heavy weights). This type of exercise is probably best for

people who think that they are entering old age, in that it is efficient: You

can get perceptible benefit from one set of ninety seconds. 1

An encouraging development in exercise physiology is the technique

called periodization, which is really just a pretentious way of saying that

it is good to mix up different types of exercise on different days. You now

have official sanction for switching between different types of exercise

because you are bored.

Doing It:

Prepare for Battle, Prepare for Victory

Even the easiest diet has problems, ups and downs, and temptations. You

have to plan out how you are going to handle different situations. If you

know that the celebratory business meeting will be serving pizza, go in

with a plan (eat beforehand, decide if you are comfortable enough to eat

the toppings without the crust, etc.) There are techniques for staying

on your diet and dealing with hunger, but you also have to prepare for

success—that is, what to do if you actually aren’t hungry. Suppose you’ve

started a low-carbohydrate diet, you sit down to the recommended broc¬

coli and steak, and you are full after eating one quarter of what you usually

eat. You should be prepared to stop. In restaurants, doggie-bags are a good

strategy unless you’re on a first date, but then you’ll probably be on your

best behavior anyway. What should you do if the birthday cake looks

disgusting and you don’t want to eat it? It is, after all, your boss s birthday.

There are many techniques. Put the birthday cake on your plate and walk

around the room stabbing at it with your fork, for example. After a while

you can set it on the sideboard with the other half-eaten pieces of cake.

If you are worried about wasting food, ask for a doggie bag—tell them,

“this is so good, but I can’t eat another bite”—and then compost it when

you return home.

Handling the Crisis

Doing It:

Minister to Yourself, Cross-Examine the Experts

Doctor: Therein the patient

Must minister to himself.

Macbeth: Throw physic [medicine] to the dogs; Til none of it .

—William Shakespeare, Macbeth

Doctors dont study nutrition. Nutritionists dont study medicine. Neither

studies much science. This is an exaggeration, of course—many great scien¬

tists were trained as MDs—but it is not without some truth. Having an

MD does not qualify a doctor on anything except their medical specialty.

Experts have the obligation to justify their opinions, and you have the right

to expect that justification. In the end, though, you are your own expert.

Strive to learn as much as you can, and go for results rather than experts.

Perhaps most troubling is the proposition that the whole field of profes¬

sional nutrition is fundamentally flawed. It is genuinely hard to understand

how the progression of warnings about the lethal health effects of red

meat or white rice can continue. These conclusions are not only contrary

to our own intuitions, but when you look at the underlying data, they are

actually scientifically meaningless. It is difficult to accept that the dozens

of publications coming out of the Harvard School of Public Health and

published in The New England Journal of Medicine are of extremely poor

scientific quality. Where s the peer review? Where’s the expert training?

The failure of experts and the lack of genuine peer review represent a major

component of the crisis in nutrition and in medicine at large. In this book

I will give you the tools to see how establishment medicine went off track,

and I will show you how the three rules will help you minister to yourself.

Doing It: The Low-Carbohydrate Principles

My personal preference is for principles over formal diets. However, if you

like diets with precise instructions, there are millions. For low-carbohydrate

and ketogenic diets, The New Atkins for a New You , 2 Protein Power [ and

The Art and Science of Low-Carbohydrate Living are good places to start.

Paleo diets might also be of interest—they are like the Mediterranean diet

Nutrition in Crisis

in that nobody really knows what they are, but they are generally low-

carbohydrate and employ a higher ratio of science to emotion than other

options. Numerous books and websites will give you precise instructions

and recipes for any of these diets. If your MO is to just fit the general

low-carbohydrate strategy to your own lifestyle, on the other hand, these

are the big principles.

Your Carbs Come from Vegetables

The simplest way to break into carbohydrate restriction is by brute force:

no rice, no potatoes, no bread, no pasta, and no dessert beyond a small

amount of fruit. What you have to sacrifice is dependent on your current

diet. If you normally eat a steak with potatoes and broccoli, leave out the

potatoes. This might be all you have to do. If you are full from this smaller

amount of food, put the rest in the refrigerator, or compost it if its going

bad. If you still have an appetite, you can have more steak, but in fact most

people don’t want more steak even if they can afford it—so it’s usually

more broccoli. Vegetables contain some carbohydrate, but the important

thing is that you don’t really have to count anything. You are likely to have

real success with this simple approach. If you have diabetes or metabolic

syndrome, you are virtually guaranteed to get better—although, again, if

you are taking medication, you have to do this with a physician who can

lower your medications appropriately.

To delve a bit deeper, you need to understand a couple of things: First,

there is a graded response, which means that the benefit is roughly propor¬

tional to the amount of carbohydrate that you remove from your eating;

and second, there is a breakpoint, where there is enhanced weight loss.

The breakpoint is generally marked by the presence of ketone bodies in

the blood (ketosis) and urine. Strictly speaking, the presence of ketone

bodies in the urine is called ketonuria (although frequently taken as a sign

of ketosis). The ketone bodies indicate a significant switch from reliance

on carbohydrate as an energy source to a new, predominantly fat-based

metabolism. At this point, you will have to attend more precisely to how

much carbohydrate you actually ingest. The simple rules above will likely

get you in the carbohydrate range of 100 grams per day, which is a big

switch for most Americans. To go into ketosis, you will have to go below

20-50 grams per day, though different people have different cutoffs.

Handling the Crisis

Though the Atkins diet has assumed the status of Kleenex and is

considered the generic low-carbohydrate diet, it actually has a more

specific strategy than you might realize. The Atkins diet’s recommenda¬

tion is to achieve ketosis for two weeks, followed by gradually re-adding

carbohydrate into the diet, presumably for reasons of taste or possibly on

the principle that you want to eat similarly to how you used to but with¬

out making yourself fat. Many people stay in ketosis indefinitely, but this

requires more attention and usually a period of adaptation.

My survey of an online support group called the Low-Carber Forums 5

found that the majority of people on low-carbohydrate diets eat all the

nonroot vegetables they want without counting grams of carbohydrate,

even though most of these individuals thought that they were on the

Atkins diet, which specifies precise grams of carbohydrate.

If you eat at home a lot, it is important to learn how to cook vegetables.

Like many tasks, cooking vegetables involves spending a great deal of time

thinking and procrastinating and a relatively small amount of time actually

doing the job. A good solution is to time yourself—the task likely ends up

taking less time than you’d imagined. So, find the appropriate vegetable

dishes for either quick cooking or for cooking in advance. Cauliflower, for

example, is a common component of low-carbohydrate diets, and you can

make steamed cauliflower in just a few minutes. Eat it straight or use it

later for other very quick recipes.

Don’t Be Afraid of Fat

People have always eaten fat. The antifat campaign is of recent origin, and

countless histories and exposes have proven its lack of grounding in scien¬

tific evidence. It has been one of the truly bizarre phenomena in the history

of science—in this most scientific of periods, we have simply ignored the

failures of the numerous experimental trials of the low-fat idea. The big

clinical studies—the Framingham Study, the Oslo Diet Heart Study, and

a dozen others culminating in the Womens Health Initiative—have found

no support for the theory. When pressed, most health agencies say that

there is nothing wrong with fat in general, only with saturated fat, but even

the evidence for removing saturated fat is missing from the big studies.

From the scientific point of view, however, these studies were successful.

They asked the real question: “Are low-fat, especially low-saturated-fat,

Nutrition in Crisis

dietary patterns beneficial for prevention of cardiovascular disease and

other health problems?” Many were well done, within the limitations of

large population studies, and they gave a clear answer: No. Low-fat diets,

especially low-saturated-fat dietary patterns, do not provide benefit. A

clear question was posed and the answer was equally clear. Of course, as

in court, you cannot be found innocent, only not guilty. There might well

be conditions where saturated fat is a risk, but we have not found them.

We suspect that high saturated fat in combination with high carbohydrate

might have risk but we haven't even been able to show that.

So, my advice—grounded in the existing evidence—is to not be afraid

of fat. If you are concerned about saturated fat, there are numerous

alternatives, but I would not advise trying to go low-carbohydrate and

low-fat unless you like being hungry. You might, in fact, like the feel¬

ing—feelings are different if we impose them ourselves rather than having

them imposed on us. Total calorie restriction does have general health

benefits, but, even there, the de facto reduction in carbohydrate is likely

the controlling factor.

Desserts and Sweets

Carbohydrates as a chemical class encompass simple sugars and their

polymers, starch and related compounds. Reducing sugar intake is part

of reducing carbohydrate intake, and sugar is an easier target for elimina¬

tion than other sources of carbohydrate: Candy is considered frivolous.

Because sugar assumes a more discretionary position, it might be the best

place to start reducing carbohydrate intake, and for some people, elimina¬

tion of sugar might be all that's necessary. It's important that you don't

put starch back in place of the sugar that you remove, though. What the

strange collection of bedfellows currently involved in the political move¬

ment that I call fructophobia, the attack on sugar, forgot to tell you is that

sugar is a carbohydrate. If you cut out sugar and replace it with “healthy”

high-grain, high-carbohydrate oatmeal, you are stacking the cards against

yourself (even if you think that oatmeal offers benefits in fiber that offset

the number of carbohydrates). Many people, anecdotally, gain benefit by

simply removing sugar-sweetened soda, but if you have a sweet tooth, you

might need some help. One strategy is to imagine that you are conducting

an experiment. The hypothesis from anecdotal observation is that cravings

Handling the Crisis

for sweets disappear after three days without sweets. Your experiment will

test whether that is true for you.

If you must have something sweet, there are several nonnutritive sweet¬

eners—some natural, some artificial. You might want to avoid artificial

sweeteners, however, especially those in diet soda, because they can sustain

your taste for sweets and provoke bad reactions in certain individuals. The

scare stories about the artificial sweeteners are not scientifically sound, but

the possible psychological effect of “sweetness” might be real.

You should do whatever works for you, but I think that for most people,

it is best to avoid the common nutritional advice that you are allowed treats

or that you are allowed a cheat day. You are not allowed treats. Nobodys

perfect and sweet things do taste good, and you will have some. We all

screw up periodically, and when this happens, you just have to get back to

the plan. However, sweets are not allowed in the sense of being a specific

feature of your diet. Consumed sweets are, scientifically speaking, experi¬

mental error. There are people who are able to incorporate some sweets as

part of their diet, who don't deviate from the single ice-cream bar that they

have every day as their lone large source of carbs. Some people can get

away with eating these sweets, and if you like food, you will certainly find

foods that are worth the risk to your diet, but they are not allowed in the

sense of a recommendation. There is a big difference between the attitude

that you cant be perfect and the attitude that you deserve an occasional

treat. But I am not suggesting that you should indulge feelings of guilt or

mentally beat yourself up. The psychologist Alfred Adler always advised:

“Do the wrong thing or feel guilty but not both.”

The current hysteria about sugar might help you cut back, but there

is great danger in characterizing things as forbidden fruit, as our very

earliest history shows, especially when someone is trying to ban that

forbidden fruit. Being told that you cant have something might make

it more appealing, and when you see the absurd lengths that the media

and researchers alike go to in order to demonize sugar, you might begin

to think it is safe. Lawyers call this the Reverse Mussolini Fallacy: Just’

because Mussolini made the trains run on time doesn't mean you want

them to be late. (The last time I was in Italy, the trains did run on time,

contrary to the Italian stereotype.) Just because the USDA says it's bad,

doesn't mean it's good.

Nutrition in Crisis

Fruit and Other Tricky Foods

All fruits contain sugar, and as with any type of food, it is best to stay with

the rule: Eat to the meter. If it doesn’t interfere with weight loss or blood

sugar, or whatever your goal is, then it is OK. I like Suzanne Somers’s

technique. Unlike many experts who might be thin themselves (what does

Walter Willett know about fighting fat?), Somers lost a part in Starsky and

Hutch because they told her she was “a little too chunky.” She recommends

eating fruit in pieces—half an apple now, the other half later—with the

goal of avoiding insulin spikes. 6

Mike and Mary Dan Eades, authors of Protein Power, ran a clinic for

many years. They had thousands of patients on low-carbohydrate diets.

Despite generally achieving great success, they had several patients who

complained that they had faithfully adhered to the diet but were not losing

weight. The Eades found that the three most common foods that caused

trouble were cheese, nuts, and nut butters. When these were reduced or

removed from the (already low-carbohydrate) diet, patients were able to

continue to lose weight. Cheese is probably a simple matter of overcon¬

sumption. Although apparently safe, moderation must be applied to cheese.

Dr. Atkins recommended that you restrict yourself to hard cheeses, 7 which

pretty much means serious cheeses that you might find in a gourmet cheese

shop. Although it’s not obvious what was behind his recommendation, the

greater intensity might be more satisfying, and the current price of good

cheese (e.g., two- or three-year aged Gouda) also ensures moderation for

most of us.

“Not a License to Gorge”

The phrase “not a license to gorge” appears in the original Atkins book

several times. 8 The idea is that, although there are no stated limits on what

you can eat as long as you keep carbohydrate low, overeating is not encour¬

aged. The principle that “a calorie is a calorie” is not correct, but calories

do count, and if you are concerned about your weight, you undoubtedly

already know that it is possible to defeat any diet. The nutritionists usually

give you exactly the wrong advice but are actually right about eating slowly.

Satiety sets in slowly. For people who like food, the prescription is simply to

not eat if you are not hungry. A possible exception is a situation where you

feel that you have the kind of cravings that will set your low-carbohydrate

Handling the Crisis

diet back. In such a case it might be useful to eat something thats allowed

to satiate the craving, even if it seems like overeating: some high-protein,

high-fat food (if you have an Eastern European butcher, real kielbasa is

best). Eventually, your cravings will stabilize. In general, the corollary to

the second part of Rule 2 is that if you have to overeat, don’t overeat carbs.

Diet Definitions

There are many variations of diets based on carbohydrate restriction. As

a treatment for diabetes, the principle is to keep carbohydrate as low

as possible. In less stringent conditions, there might be more room for

Table 1.1. Operational Definitions of Carbohydrate-Restricted Diets

Diet

Carbohydrate restriction

Very low-

carbohydrate

ketogenic diet

(VLCKD)

20-50 grams per day, or less than 10% of the 2,000

kilocalories per day diet. Generally, although not always,

accompanied by ketosis, this is the level of the early phases

of the plans in many popular carbohydrate books.

Low-

carbohydrate diet

Less than 130 grams per day, or less than 26% of a nominal

2,000 kilocalories per day diet. This corresponds to the ADA

definition of 130 grams per day. This is a generally accepted

number, likely derived from misinterpretation of Cahill’s study

of the onset of ketosis.

Moderate-

carbohydrate diet

26-45% of the 2,000 kilocalories per day diet. The upper limit

is chosen as the approximate carbohydrate intake before the

obesity epidemic (43%). Current consumption is about 49%.

High-

carbohydrate diet

Greater than 45% of the 2,000 kilocalories per day diet.

Recommended target on ADA websites. The 2015 Dietary

Guidelines for Americans recommend 45-65% carbohydrate.

For comparison

Pre-obesity

epidemic

(1971-1974-

NHANES 1)

Men: 42% carbohydrate

(-250 grams for 2,450 kilocalories per day)

Women: 45% carbohydrate

(-150 grams for 1,550 kilocalories per day]

Year 1999-2000

Men: 49% carbohydrate

(-330 grams for 2,600 kilocalories per day)

Women: 52% carbohydrate

(-230 grams for 1,900 kilocalories per day]

Nutrition in Crisis

maneuvering. If specific diets are referred to, table 1.1 provides the associ¬

ated guidelines that have been published in several peer-reviewed journals

by professionals with the credentials and experience.

These definitions are important. Authors in the nutritional literature give

themselves license to call anything that they want a low-carbohydrate diet.

With a straw man in hand, it is not hard to show that a low-carbohydrate

diet is dangerous to your health, but the results have very little to do with

the actual established guidelines.

Depending on your goal and who you are, you might experience a

graded response: The greater the carbohydrate reduction, the greater the

weight loss, and the greater the improvement in blood glucose. Particularly

in weight loss, however, there might be a threshold effect. The onset of

ketosis, which for most people occurs at a daily intake of about 30 grams,

will have a more dramatic effect.

-CHAPTER 2 -

Whaddaya Know?

F or several years, we gave incoming medical students a questionnaire

to assess their knowledge of nutrition. These students were among

the most accomplished in the country, but, like everybody else,

most of their nutritional information came from rumors and the fronts of

packages on supermarket shelves. We didn’t keep any quantitative feed¬

back on the quiz, but some of students liked the format, and you might,

too. No particular knowledge is assumed beyond your life experience as

a nutritional end user, an eater. The quiz only tests how you were able

to sift through all the information thats available on the internet and in

popular books, and it is intended as a teaching device: The answers provide

basic information. I provide the quiz first so you can see how you do, and

then, as I go through the answers, I will delve into the nutritional concepts

associated with each question. So, whaddaya know?

1. The most energy-dense food (most calories per gram) is:

□ carbohydrate □ protein

□ fat □ alcohol

2. For a slice of buttered bread, which is more fattening?

□ the butter □ You cannot tell from the

□ the bread information given.

□ Both are equally fattening.

3. During the epidemic of obesity and diabetes, the macronutrient

that increased most was:

□ carbohydrate □ All were about the

□ protein same. Calories increased

□ fat across the board.

Nutrition in Crisis

4. The macronutrient most likely to

raise blood glucose in people with

type 2 diabetes is:

□ carbohydrate

□ fat

□ protein

□ alcohol

5. The dietary requirement for carbohydrate is:

□ approximately

□ as much as possible

130 grams per day

□ There is no dietary

□ approximately

requirement for

50 percent of calories

carbohydrate.

6. The amount of carbohydrate recommended by the American

Diabetes Association and other health agencies is:

□ approximately

□ as much

130 grams per day

as possible

□ approximately

□ as little

50 percent of calories

as possible

7. A good source of monounsaturated fat is: (check all that apply)

□ butter □ avocado oil □ flaxseed oil

□ olive oil □ corn oil

□ canola oil □ soybean oil

8. The diet component that is most likely to raise triglycerides

(fat in the blood) is:

□ fat □ carbohydrate □ protein

9. In general, what effect does a low-fat diet have on HDL-C

(high-density lipoprotein cholesterol, i.e., “good cholesterol”)?

□ increase □ decrease □ no change

10. The dietary change that is most likely to increase the risk of

cardiovascular disease is:

□ unsaturated fat □ unsaturated fat □ carbohydrate

-> saturated fat -> carbohydrate unsaturated fat

Whaddaya Know?

□ carbohydrate

saturated fat

□ saturated fat

-> carbohydrate

□ saturated fat

unsaturated fat

The Calorie, the Calorimeter

Now that youVe taken the time to complete the quiz, let’s see how you

actually did. I encourage you to keep a tally as you go along and see how

you stack up against the incoming medical students.

1. The most energy-dense food

(most calories per gram) is:

□ carbohydrate

[El fat

□ protein

□ alcohol

Our first-year medical students do surprisingly poorly on the first

question, considering how basic it is. Typically only 85 percent of our

incoming class gets it right. Although they have not yet been through

the metabolism course, this is a highly educated group of people, so you

would assume that everyone knows that fat is the most energy-dense

macronutrient. The likely explanation is that they are not curious about

nutrition because they don't see it as part of medicine and because they are

mostly young, healthy, and thin.

The operational numbers in kilocalories per gram are 4, 4, 9, and 7 for

carbohydrate, protein, fat, and alcohol, respectively. Calories are a measure

of energy, and calories in nutrition represent the energy that we can obtain

by metabolizing ingested food. In physics, energy is defined as the ability

to do work, which is not so different from the common day-to-day idea. By

using a device called a calorimeter to measure heat—one form of energy—

we are able to determine the total number of calories in a sample of food.

The food is placed in a small container in an atmosphere of oxygen under

pressure, and it is then ignited. The container is surrounded by a water

bath that heats up when the sample is ignited. By recording the change

in temperature in the water bath, we can calculate the amount of heat

Student response (%)

Nutrition in Crisis

generated from the combustion, and this number can be assigned to the

oxidation of the food. This is the real definition of the nutritional calorie.

In physics, the definition of a calorie is given as the amount of heat

required to raise the temperature of water by one degree. The dietary “calo¬

rie” is equal to a physical kilocalorie (kcal)—that is, 1,000 physical calories.

Use of “kcal” in nutrition is increasing, and in this book we use kcal when

referring to quantity for the sake of clarity This is one small step toward

helping nutrition become scientific.

The definition I have given for a calorie is quite simple, but there are

some very important nuances: The calories assigned to a food represent the

energy for complete combustion of that food in oxygen. Calories refer to

the following chemical reaction, not to the food itself:

Food + Oxygen -> Carbon Dioxide + Water

Food + O 2 CO 2 + H 2 O

Heat produced in the calorimeter measures the energy for this specific

reaction. Again, the energy is in the reaction, not in the food. It is not like

particle physics where the mass of a particle is given in units of electron-

volts, a measure of energy, (because of E = me 2 ). We will come back to this

point when we delve into thermodynamics and consider what is wrong

with the idea that a “calorie is a calorie” in chapter 9.

Fat is the most energy-dense macronutrient at 9 kcal per gram (keal/g),

and for that reason, it is considered inherently fattening by nutritionists.

This is the basis of traditional recommendations for low-fat diets for

obesity. There is more to the problem than caloric density, however, and it

turns out that there is more to getting fat than total calories. To see how

this all plays out in a real situation, consider the next question.

2. For a slice of buttered bread,

which is more fattening? Student response (%)

□ The butter. 11

□ The bread. 85

□ Both are equally fattening. 4

[xj You cannot tell from the information given. 0

Whaddaya Know?

While we do not normally ask trick questions, this falls into that

category. An argument could be made that because butter has an energy

density of 9 kcal/g, it is inherently more fattening, but you really need

more information. If you put a lot of butter on the bread, it would indeed

be more fattening, but in fact, people rarely put more than one tablespoon

of butter (approximately 100 kcal) on a slice of bread (also approximately

100 kcal). Yes, fat is the most calorically dense food, but caloric density, like

any density, can be very misleading: Density is a measurement per amount

of material—per unit of volume, per gram, per something—so it matters

how much of that something you actually have. Calories per gram is not

informative unless you know how many grams. Things like caloric density,

or any density for that matter, are measures of intensity, and technically are

called intensive variables. An intensive variable does not depend on how

much you have: One tablespoon of butter has the same energy density

(kcal/g) as two tablespoons of butter, but obviously two tablespoons have

twice the total calories. Such total, absolute measurements are called exten¬

sive variables. It ? s like that riddle: Which is heavier? A pound of uranium

or a pound of Styrofoam? Of course they are the same. A pound is a pound.

Uranium has an extremely high density (about 1.6 times that of lead), so

a pound of uranium is only a little bigger than a major league baseball,

whereas a pound of Styrofoam would fill up an entire room.

Fraction, or percent, is another kind of measure of intensity that can be

misleading. Looking ahead, this is one reason that providing the percent

increase of risk in a clinical test is not helpful, however much the media

and scientific literature might seize on such numbers. After all, your odds

of winning the lottery are increased by 100 percent, or doubled, by buying

two tickets instead of one—but should that really make you want to play?

Bottom line: When you hear people say that fat has more calories per

gram, know that it is irrelevant. You have to know how many grams you

are actually eating.

The Obesity Epidemic

So, what did we eat during the obesity epidemic of the past thirty or forty

years? What kinds of macronutrients were in our diet? You might know

this better than our students, who probably did not attend to the problem.

Nutrition in Crisis

3. During the epidemic of obesity and diabetes,

the macronutrient that increased most was: Student response (%)

0 carbohydrate 53

□ protein 4

□ fat 2

□ All were about the same. 41

Calories increased across the board.

At this point many students have caught on to where I’m coming from,

and there are probably more votes for carbohydrate than if this had been

the first question. However, many people, students included, still think that

calories increased across the board. The epidemic of obesity and diabetes

has been accompanied by a substantial decrease in the percentage of fat in

peoples* diets, and, at least for men, the absolute amount of fat consumed

also went down slightly.

Figure 2.1 shows data from the National Health and Nutrition

Examination Survey (NHANES), which was conducted at the intervals

listed on the horizontal axis. The vertical axis is the absolute amount of

calories consumed, and the large increase in energy consumed appears to be

due entirely to a dramatic increase in the consumption of carbohydrate. The

percent change shown along the top indicates the expected decrease in fat,

but, at least for men, the absolute amount of fat (total calories) and, notably,

the absolute amount of saturated fat went down. (The extent to which this

gender difference plays out in real life is not clear. Anecdotally, it is easier

for men to lose weight, but the major effects of metabolism are qualitatively

the same across sexes, and the benefit of controlling the glucose—insulin axis

applies to women as well. Obviously, other hormones might play a role.)

There remains a lot of error in these kinds of surveys, but it is quite clear

that there was no increase in fat intake. The clearest conclusion is that

an increase in carbohydrate consumption, rather than fat consumption, is

associated with greater total intake, and that a “Western diet,” as they call

it in the nutrition literature, does not mean a high-fat diet. It is widely

said that association does not imply causality. The more accurate state¬

ment is that association does not necessarily imply causality. Few people

would deny that the association of dietary calories and body mass is causal,

however nonlinear the association might be. Whether the association

Whaddaya Know?

3500

3000

15 2500

% 2000

> 1500

I .000

500

NHANES Men

CHO:FAT:PROT

42:37:17 40:33:16

Overall intake

+ 6 . 8 %

Carbohydrate

Fat

Saturated fat

1971-74 [I) 1988-94 (III) 1976-80 [II]

Years

+23.4%

-5%

-14%

1999-2000

3500

3000

= 2500

>. 1500

w 1000

500

NHANES Women

CHO:FAT:PROT

45:36:17

70 75 '80 '85 '90 '95 '00

Years

Overall intake

52:33:15

+21.7%

Carbohydrate

Fat

Saturated fat

1971-74 [I] 1988-94 (III) 1976-80

Years

+38.4%

+ 11 %

— +3%

1999-2000

Figure 2.1. Consumption of macronutrients during the epidemic of obesity and

diabetes. The horizontal axis represents the periods in which data was collected. The

left vertical axis is the absolute energy input in kilocalories. The numbers on the far

right refer to the percent change in calories. The ratios of macronutrients are shown

along the top. Note: CHO = Carbohydrate. Data from the National Health and Nutritional

Examination Survey.

Nutrition in Crisis

between increased carbohydrate intake and increased calories is causative

is one of the central questions explored in this book. The argument will

be that, given the effectiveness of low-carbohydrate, high-fat diets as a

treatment and sometimes a virtual cure for diabetes, it would be surprising

if carbohydrate was not involved in some causative role.

Finally, there is an obvious association between the official advice of

the USDA, the AHA, and other authorities to reduce fat and increase

carbohydrates, and what people actually did. They reduced fat, at least as a

percentage of calories, and they dramatically increased carbohydrate.

Looking ahead, one way to test whether there is a causal link between

carbohydrate intake and obesity is to simply reduce carbohydrate under

conditions of fixed calories while monitoring weight and incidence of

diabetes. There are some good experiments that test this (see descriptions

ofVolek’s experiment in chapters 7 and 11). Whatever else can be drawn

from the NHANES data, the association between increased carbohydrate/

decreased fat intake and obesity and diabetes is the single result that makes

the largest impact on our medical students, and it remains an undercurrent

in any analysis of the role of macronutrients.

4. The macronutrient most likely to raise blood

glucose in people with type 2 diabetes is: Student response (%)

(El carbohydrate 83

□ protein Not available

□ fat Not available

□ alcohol Not available

This question is, or should be, obvious. The correct answer was chosen

by 83 percent of our students. Statistics on the wrong answers were lost,

but the surprise is that anybody got it wrong. Diabetes is fundamentally a

disease—or, really, several diseases—of carbohydrate intolerance. Glucose

normally stimulates secretion of the hormone insulin from beta cells of the

pancreas. Insulin is a kind of master hormone, controlling fat and protein

metabolism as well as maintaining normal blood glucose, and in that

sense, carbohydrate controls other aspects of metabolism. People with type

1 diabetes cannot produce insulin in response to blood glucose. People

with type 2 diabetes have progressive deterioration of the beta cells. They

Whaddaya Know?

do produce insulin but their cells respond poorly—a phenomenon called

insulin resistance. Diabetes is as much a disease of fat metabolism as of

carbohydrate metabolism: The primary effect of insulin is on synthesis and

breakdown of fat, and a person with type 2 diabetes might have excessive

fatty acids in their blood. Nonetheless, the most obvious characteristic and

the major cause of other symptoms remains the hyperglycemia (high blood

glucose). Different carbohydrate-containing foods raise blood glucose to

different extents, but the general principle holds that carbohydrate is the

primary influence on blood glucose in people with diabetes.

The Dietary Requirement for Carbohydrate

There is no requirement for dietary carbohydrate as there is for the so-called

essential amino acids or essential fatty acids. This does not mean that I

recommend doing without them altogether, even if this were possible (it

isn't: even meat has a small amount of carbohydrate).

5. The dietary requirement for carbohydrate is: Student response (%)

□ approximately 130 grams per day

□ approximately 50 percent of calories

□ as much as possible

IE] There is no dietary

requirement for carbohydrate.

The fact that there is no dietary requirement for carbohydrate means,

in a practical sense, that if you do want to reduce carbohydrate, there is

no biological limit on how much you can restrict the intake. The extent

to which you actually do restrict carbohydrate intake will depend on your

personal reaction and taste, but you do not need to consume any at all.

Nutritionists will emphasize that the brain needs glucose, but your body

is capable of making glucose from protein through the process known as

gluconeogenesis and supplying glucose from storage as glycogen. There

are also alternative fuels available in the form of ketone bodies. If you did

need dietary carbohydrate, you would die if you went without food for a

week—after all, you store a lot of fat but not much carbohydrate. I write

more on glycogen and gluconeogenesis in chapter 5.

Nutrition in Crisis

6. The amount of carbohydrate recommended

by the American Diabetes Association and

other health agencies is:

□ approximately 130 grams per day

M approximately 50 percent of calories

□ as much as possible

□ as little as possible

Student response (%)

It is hard to believe that a diabetes agency would recommend any

significant amount of carbohydrate, yet their 2008 dietary guidelines

contain the rather remarkable advice that “Sucrose-containing foods can

be substituted for other carbohydrates in the meal plan or, if added to the

meal plan, covered with insulin or other glucose-lowering medications.

Care should be taken to avoid excess energy intake. (A)” 1

The truth is that starch, which is a polymer of glucose, is undoubtedly

worse for people with diabetes than sugar. (Sugar is half glucose and half

fructose, and fructose does not stimulate insulin release.) It is the phrase

“added to the meal plan,” and the advice that follows, that is most jarring.

To many people it would seem that the ADA is saying that it is okay to

make things worse as long as you take more drugs. The “(A)” mark indi¬

cates that they consider this advice to be their highest level of evidence.

They don't cite that evidence, but it is surely not experimental. While this

book was being written, the ADA quietly dropped this passage from their

2013 guidelines, 2 but they have not explicitly stated that it was wrong. It

is unknown whether the rank and file of ADA membership ever read the

organizations follow-up statements, and if so, whether they thought they

were of high quality or simply political statements that no one had time

to fight. One of the organizations statements from 2012 reads as follows:

“Although brain fuel needs can be met on lower-carbohydrate diets, long¬

term metabolic effects of very low-carbohydrate diets are unclear.” 3

In fact, the long-term effects of low-carbohydrate diets are clear. Very

clear.There are trials spanning one or two years, and internet sites and forums

make apparent that low-carbohydrate is a way of life for many people with

diabetes. Although personal stories are hard to document, we would have

heard if there were any indication of long-term problems. The medical

establishment has been obsessed with finding fault with low-carbohydrate

Whaddaya Know?

diets, and over forty years, they have found nothing. Zero. Zilch. Zip. More

important, there is no reason to suggest that there would be any long-term

effects. In science, you don’t start from scratch. You don’t assume that there

is harm unless there is a reason to. Nothing about reducing carbohydrate

suggests harm. The ADAs use of the word unclear implies conflicting data,

but there are no conflicting data and there is no reason to expect any. I

suggested to a spokesperson for the ADA that the organization was stron¬

ger on what they were opposed to than on anything positive they had to

offer. She admitted that it was a fair criticism. So, why are they opposed

to carbohydrate restriction? The ADA guidelines of 2010 say: “Such diets

eliminate many foods that are important sources of energy, fiber, vitamins,

and minerals and are important in dietary palatability.” 4

If “care should be taken to avoid excess energy intake,” then why would

you need carbohydrate as an “an important source of energy?” Diabetes is a

severe disease—does anyone think that taking extra vitamins and minerals is

as important to a person with diabetes as maintaining healthy blood glucose?

There is also a subtle switch from using the word “carbohydrate” to refer to

the micronutrient, to using it to mean carbohydrate-containing food. Finally,

I would suggest that dietary palatability is not the ADA’s area of expertise.

The guidelines from the ADA, as from other health agencies, are supposed

to be serious, but, in fact, have the character of an infomercial. Their stan¬

dards are those of selling a product rather than presenting a scientific case.

It seems that we are supposed to just take what they say at face value. The

ADA is a private organization, but they receive public support, at least in

that they are tax-exempt. Their experts are often federally funded as well.

Are they really free to say whatever they want without justification? This is

the kernel of the crisis in nutrition. Science is still continuous with common

sense. It is not okay at all to, as the ADA says, eat more carbohydrate if

“covered with insulin or other glucose-lowering medications.” If you can

take less medication, that is considered a benefit in every disease I know.

Lipid Chemistry

The second half of the quiz is about lipids. The discussion of what to eat

frequently boils down to how much fat one should consume. The medi¬

cal establishment—including the AHA, ADA, Harvard School of Public

Nutrition in Crisis

Glycemic Index

Politically Correct Low-Carbohydrate

The original questionnaire given to medical students asked one

question about the glycemic index. Intellectually, the glycemic

index was an important idea. It followed the same principle as

low-carbohydrate diets, and was seemingly of practical value.

The intention of the glycemic index was to address the experi¬

mental effect of carbohydrate on blood glucose. The glycemic

index addresses the old idea, pretty much a dogma when I was

in school, that simple sugars would cause a rapid rise in blood

glucose, but complex carbohydrate—which at that time still

meant polysaccharides (starch)—would not. The idea was ques¬

tioned at some point, however, and it turns out that when you

actually measure the effect of foods on blood glucose, it’s not

easily predictable—that is, it must be determined experimen¬

tally. Glycemic index (GI) is precisely defined as the area under

the blood glucose time curve during the first two hours after

consumption of 50 grams of carbohydrate-containing food. In

other words, it is the total amount of blood glucose for a fixed

time period after ingestion.

Whatever its promise, low-GI diets have evolved to be a

politically correct form of carbohydrate restriction, and it is

questionable if they have any value at all. Eric Westman, who

has experience with both kinds of diets, put it well: “If low-GI

is good, why not no-GI?”

GI is mainly influenced by the absolute concentration of

glucose in the food, the extent to which glucose appears in the

blood (not necessarily from the food itself), and the quantity

of other nutrients such as fat or fiber in the carbohydrate-

containing food that might slow the digestion or absorption

of carbohydrate. The individual personal variation makes it

Whaddaya Know?

doubtful that GI diets are useful at all. In comparison to simply

reducing carbohydrate, low-GI strategies are complicated and

require looking up and calculating values, a feature that might

be appealing to some, but is probably annoying to most.

The difference between intensive variables, such as caloric

density, and extensive variables, such as total carbohydrate

eaten, was brought out at the beginning of the quiz. GI is an

intensive variable. Two bowls of cereal have the same GI as one.

If there is not much carbohydrate (or really much glucose) in a

food, it will have a low GI, but it could still have a large effect

if you consume a lot. The glycemic load attempts to correct this

problem. The glyccmic load (GL) is defined as the GI multiplied

by the grams of carbohydrate in a sample of a particular food.

Obviously, GL is still an intensive variable. You still have to

know how much is consumed. More important, are you really

sure that GL will really be different from total carbohydrate,

which is easier to calculate?

There is also the overall character of using GL A slice of

white bread has a high GI. The GI will go down if you smear a

tablespoon of butter on the bread. It will go down still further

if you add two tablespoons of butter. If you could somehow

butter infinitely, until for all intents and purposes you have pure

butter, you would have a GI = 0, which is probably not helpful

for those who want to use the GI as a guide to eating.

One final ambiguity: GI measures blood glucose. Fructose, a

sugar of great current interest (because it is 50 percent of sucrose

and slighdy more than 50 percent of high-fructose corn syrup),

is partially converted to glucose in two hours, which is why the

GI of fructose is 20 and not zero. In fact, more is converted after

that time, severely compromising any assertion about the diff er¬

ences in effect of the two sugars. Sucrose has a GI of 70, wliich

is roughly the average of glucose and fructose. Thus, ice cream

has a lower GI than potatoes. Yet now we cant recommend ice

Nutrition in Crisis

cream because of the high fructose. Lower GI or lower fructose?

How can you do both without saying “low-carbohydrate” out

loud? Tliis tangled web is woven out of the failure to face scien¬

tific facts. This aspect of the nutritional crisis is probably best

addressed by ignoring glvcemic index altogether.

Health, and National Institutes of Health (NIH)—has been on an antifat

crusade for fifty years. No experiment will make them change their point

of view—or more precisely, clearly describe their point of view, which they

can apparently mold and reform to fit any challenge. Low-fat products are

still everywhere and low-fat is still recommended in one way or another.

The diet-heart hypothesis holds that lipid in your diet, particularly satu¬

rated fat, will raise one or another lipid fraction in your blood, which will

in turn cause you to have a heart attack. TheyVe put it to the test—big,

expensive clinical trials with tens of thousands of subjects, hundreds of

millions of dollars—and it consistently fails. To figure out whats going on,

and to avoid the path to poor health, you have to understand the details.

Some of the details involve more chemistry. Many terms in the popular

media are used incorrectly and contribute to the current crisis. A little

precision will help you understand the problem.

First, you need to know that the term lipid refers to a diverse collection

of chemical compounds, all of which are sparingly or not at all soluble in

water. The group includes fatty acids, fats and oils, cholesterol, and deriva¬

tives of these compounds. Directly applicable here are the fats and oils and

their constituent components, the fatty acids and glycerol. WeTl look at fat

structure and the meaning of saturated and unsaturated fat.

The crux of the crisis is the packaging of lipids into complex units, the

lipoproteins (LDL, HDL, and others) that transport cholesterol and fat.

These particles are targeted as surrogate markers for cardiovascular risk.

As surrogates, they are of limited value. The extent to which they truly

indicate risk is controversial. Much of the revolution in nutrition revolves

around facing the fact that a surrogate is not the same as actually getting

Whaddaya Know?

sick. It is still important to be familiar with them, however. They will be

part of your lipid workup when you get a clinical lab test.

The big payoff will be wrapping your head around how fat interacts with

carbohydrate, and looking ahead, we will try to understand how carbohy-

drate can be converted to fat, but, to

a large extent, fat cannot be converted

to glucose. We will want to grasp how it is that we cannot use our fat stores

to keep glucose at normal levels and how it is that the amount of dietary

carbohydrate might be more important than the amount of dietary fat in

determining how much body fat we

have. Now, back to the quiz.

7. A good source of monosaturated fat is:

(check all that apply)

Student response (%)

□ butter

M canola oil

□ corn oil

□ flaxseed oil

0 olive oil

IE) avocado oil

□ soybean oil

You hear the terms saturated fat and polyunsaturated fat often, but they

are not quite precise; only fatty acids can be unsaturated or saturated. All

dietary and body fats and oils are triglycerides (TG), or, more correcdy tria-

cylglycerols (TAG).The name tells you about the structure: There are three

acyl groups (pronounced “ay-seal”). Acyl is the adjective form of acid, and

the components are fatty acids whose three acyl groups are attached to the

compound glycerol. Fats have an is-shaped structure. The three arms of the

E are the fatty acids, and the backbone is the compound glycerol. As shown

in figure 2.1, only the fatty acids can be saturated (SFAs) or unsaturated

(UFAs). Saturatedfat simply means that the fat contains a higher propor¬

tion of saturated fatty acids. For unsaturated fat and its variations, mono- or

poly-, the chemical bonds that attach the fatty acids to the glycerol are

called ester bonds. You only need to know the term ester because when the

fatty acids are found alone, especially in blood, they are referred to as free

fatty acids (FFAs) or—because they are no longer attached to the glycerol

part by the ester bonds—nonesterified fatty acids (NEFAs). So FFA and

Nutrition in Crisis

Glycerol

Fatty acid

Fatty acyl- (adjective)

Q Carbon

Oxygen

Hydrogen

Triglyceride (TAG)

Figure 2.2. Fat structure. Fats and oils are triglycerides (TGs), formally triacylglycerol

(TAGs). There are three ester bonds to glycerol.

Stearic Acid-Saturated Fatty Acid

Oleic Acid-Monounsaturated Fatty Acid

Linoleic Acid-Polyunsaturated Fatty Acid

Figure 2.3. Single and double chemical bonds as seen in common fatty acids.

Whaddaya Know?

Trans isomer

“across”

★

A=A trans-2-butene

Cis isomer

“adjacent”

cis-2-butene

Figure 2.4. Orientation around double chemical bonds. The two carbons in a double

bond each have a hydrogen atom and another atom, a carbon, or in the case of fatty

acid, a carbon chain. If these are on the same side of the double bond, this constitutes

the cis- configuration. Otherwise, the bonds are called trans-.

NEFA are the same thing. Fatty acids are long chains of carbon atoms

with a carboxylic acid group. The fatty acids provide the real fuel in fat in

the long hydrocarbon chains, much like gasoline. Carbon—carbon double

bonds are more chemically reactive and can be converted to single bonds

(e.g., with hydrogen atoms, in which case they are called saturated—that

is, saturated with hydrogen). Saturated means that all the carbon-carbon

chemical bonds are single bonds.

Saturated fats, again, have a high percentage of SFAs in the arms of

the E structure. Similarly, unsaturated fats have high amounts of MUFAs

(monounsaturated fatty acids) and PUFAs (polyunsaturated fatty acids).

For some fats, however, it is not clear that these terms are useful. One

thinks of lard as a kind of pure high-saturated fat, but in fact it is only 41

percent saturated and mostly (47 percent) MUFA, predominantly oleic

acid, the main fat in olive oil. From the survey, our students did not have

a good sense of what MUFA was. So it is a question of whether you think

that lard is half full of SFA or half empty. Most important, as in the epide¬

miological study described in the introduction, it has been impossible to

demonstrate any risk for cardiovascular disease associated with consuming

saturated fat—and yet official health agencies continue to insist that there

is. If anything, the studies show that carbohydrate is a greater risk than

saturated fat, but both correlations are too weak to attribute a significant

role to either. I will explore this more as we go along.

Nutrition in Crisis

The structure of the different kinds of fatty acids are shown in figure

2.3. The major monounsaturated fatty acid is oleic acid. Everybody thinks

that monounsaturated fats—those with a high content of MUFA, such as

olive oil—are protective of cardiovascular disease, but it is not so clear-cut.

Few fats have only SFAs. Coconut oil is the exception, but those are

medium-chain fatty acids (12—16 carbons)—that is, the arms of the E in

figure 2.3 are shorter than the more common fatty acids. Because of the

interest in ketogenic diets, coconut oil has recently become more popular:

Medium-chain fatty acids form ketone bodies more readily. In reaction,

and presumably because they don’t want the public drawing their own

conclusions, the American Heart Association has decreed that coconut

oil is dangerous, without providing any significant proof of this claim.

Consumer Reports and the popular media have echoed the AHAs opinions,

as if the AHA had come upon a new experimental discovery.

Saturated fats (again, those with a fairly large number of SFAs) tend

to be solid, while those with more UFAs tend to be liquid—generally, the

more saturation the higher the melting point. To understand why, we need

to look deeper into the structure of fatty acids, which brings us to the issue

of trans-fats.

Trans-Fats and the Meaning of Trans-

When vegetable oils are hydrogenated—the process by which some of the

unsaturated fatty acids are turned to saturated—a side reaction can occur

that changes the configuration of some of the unsaturated fatty acids from

cis- to trans-.

Let me first explain what trans- means, since the nutritional Murphy’s

Law dictates that confusion will be introduced wherever possible. The

carbon-carbon double bond has rigid geometry, so that unlike the single

bond, there is no rotation around the bond. Imagine a chain of carbon

atoms as in a hydrocarbon such as gasoline or the backbone of a fatty acid.

If the bonds are all saturated (single) bonds, then you can think of the

molecule as somewhat floppy because of free rotation around the bonds. If

there is a double bond in this chain, however, there are two ways to arrange

the structure: The two carbons in a double bond can have hydrogen atoms

on the same side of the bond (cis-) or on opposite sides (trans-). Figure 2.4

shows the geometry around double bonds.

Whaddaya Know?

Almost all naturally occurring fatty acids have the cis- configuration,

but it is important to understand that, by itself, this is just a designation

for the millions of double bond-containing compounds in the world.

Figure 2.3 shows that SFAs have less structure than UFAs, and this lack

of structure is why saturated fats tend to be solid: They are easier to pack

into a solid. (By analogy, it is easier to pack T-shirts into a box than to pack

model Eiffel Towers or heads of Nefertiti.)

MUFAs and the Mediterranean Diet

The Mediterranean diet is widely recommended for its health

benefits, but the data are very weak—if there are any at all—and it

is not obvious that anybody knows what the diet consists of beyond

pouring olive oil on everything. The idea probably originated with

Ancel Keys, generally considered the father of lipophobia. Keys

originally found a good correlation between tat consumption

(actually fat availability) in six different countries and the incidence

of heart disease in diose countries. 5 It wasn’t long, however, before

the Secretary of Health in New York State and a professor at

Berkeley published a paper showing that there were data from

countries other than the six that Keys had studied, which would

have significantly weakened the correlation. 6 Keys has generally

been characterized as a zealot, although he was probably more

open-minded than some of his followers. He was, however, not

easily embarrassed and undertook a second study of seven countries.

The Seven Countries Study on dietary availability of fat had

an interesting result: The two countries with the highest intake

of fat were Finland, which had the highest incidence of CVD,

and Crete, which had the lowest 7 It was deduced that this had

to do with the type of fat: saturated in the case of Finland,

and unsaturated, in the case of Crete. Tilings were further

complicated when it was later pointed out that there were large

differences m CVD between different areas of Finland that had

Nutrition in Crisis

the same diet. This information was ignored by Keys, a pioneer

in such approaches to dealing with conflicting data. In any case,

the finding immediately led to the recommendations to lower

saturated fat, although for most people there was a lingering

idea that it was good to reduce fat across the board, despite the

lack of correlation in the Seven Countries Study. Subsequently,

health agencies were stronger in stepping up the pressure on

saturated fat, but not so good at admitting the error in their

earlier recommending of low-fat across the board. This is still

the state of affairs. The real problem, however, is that even the

link between saturated fat and heart disease has been impossible

to establish: Direct tests failed immediately and continue to fail.

The story of the political triumph of an idea that was clearly

contradicted by the science has been told numerous times, 8 and

yet the phenomenon still persists. Every time you see a low-fat

item in the supermarket, you are looking at an artifact of one of

the most bizarre stories in the history of science.

One of the rarely cited responses to die Seven Countries

Study was a letter written by researchers at the University of

Crete and published in the journal Public Health Nutrition: The

important part of the letter is this:

In the December 2004 issue of your journal . . .

Geoffrey Cannon referred to ... the fact that Keys and

his colleagues seemed to have ignored the possibility

that Greek Orthodox Christian fasting practices could

have influenced the dietary habits of male Cretans in

the 1960s. ... Professor Aravanis confirmed that, in

die 1960s, 60% of the study participants were fasting

during the 40 days of Lent , and stricdy followed all

fasting periods of the church . .. periodic abstention

from meat , fishy dairy products , eggs and cheese^ as well

as abstention from olive oil consumption on certain

Whaddaya Know?

Wednesdays and Fridays ... this was not noted in the

study, and no attempt was made to differentiate between

fosters and non-fasters. In our view this was a remark¬

able and troublesome omission, (Emphasis added)

The whole sorry tale has now been told many times, most

recently and completely by Nina Teicholz in her expose of the

low-fat fiasco, The Big Fat Surprise} 1

Returning to the Composition of Dietary Fat

Going back to figure 2.3, the composition of different dietary fats turns

out to be somewhat surprising. It is true that there is a lot of oleic acid

(the major monounsaturated fatty acid) in olive oil (73 percent) and canola

oil (58 percent). Less well known is that the highest amount is found in

avocado oil—and probably most surprising is that oleic acid makes up

almost half of the fatty acids in beef tallow and lard (44 percent and 47

percent, respectively). Beef tallow (rendered fat) was what McDonalds

used to use to fry their French fries in—at the time, they got the thumbs

up from Julia Child—until they were pressured to switch to vegetable oil

in a movement spearheaded by Michael Jacobson. Currently the executive

director of the Center for Science in the Public Interest, Michael Jacobson

might best be described as humorless, uptight, and puritanical. I have been

accused of inappropriate behavior in making this characterization, but

you can check out his interview with Stephen Colbert for proof (Colbert:

“What is the latest thing that you’re warning people not to enjoy?”) Of

course, when McDonald s did switch to vegetable oil, the amount of trans¬

fat went up, and that got Jacobson riled up again. In any case, most of the

saturated fat in beef is stearic acid, which is considered neutral or “heart-

healthy,” but the question is at least ambiguous—and again, beef fat is only

half SFA; the other half is mostly oleic acid, as in the Mediterranean diet.

Canola oil, it turns out, is named for CANadian Oil, Low Acid, an oil

isolated from rapeseed {rape, from Latin for turnip in this context), which also

Nutrition in Crisis

contains euricic acid. This fatty acid is the star of the 1992 film Lorenzo's Oil ,

but it is not as beneficial as the movie suggests. It is generally considered toxic,

and for that reason, it is removed in making canola oil. My original vision of

the Quebecoise in their native costumes picking canola fruit from the canola

tree turned out not to be correct There is no canola tree, and the real source,

the rapeseed plant, was originally processed to remove the euricic acid.

The rapeseed plant has now been bred to have inherently low euricic

acid, and there now really is a plant called canola, the oil of which is one of

the major exports from Canada. Processing also produced trans-fatty acids,

but this has been removed from current versions of the product.

Back to the quiz. We asked students about the “cholesterol" species that

are reported in your lipid profile and that are supposed to be indictors

of risk of CVD. What is called cholesterol in this context is actually a

supramolecular (more than one type of molecule) known as a lipoprotein,

which contains lipid protein.

Lipoproteins: “Good” and “Bad” Cholesterol

There were a few questions from the original quiz that are not reprinted

here. Most of our students knew that low-fat diets lower low-density

lipoprotein cholesterol (LDL), the so-called “bad cholesterol.” LDL is

considered bad because it is assumed to correlate well with heart disease, but

the correlation is not as strong as many believe. Less than half the people

who experience a first heart attack have high cholesterol, and less than half

have high LDL (although high is, of course, subject to interpretation).

Cholesterol is not a cause of heart disease—we don t know what the

cause is—but total cholesterol is considered a very poor risk marker,

although it is still treated as a factor associated with incidence of disease.

There are, however, better risk markers, including triglycerides and the

“good cholesterol,” HDL.

8. The diet component that is most likely to

raise triglycerides (fat in the blood) is:

0 carbohydrate

□ fat

□ protein

Student response (%)

Whaddaya Know?

The phenomenon of carbohydrate-induced hypertriglyceridemia

(high blood triglyceride) has been known for at least sixty years. A major

contributor is the process of de novo fatty acid synthesis, more usually

called de novo lipogenesis (DNL), in which fatty acids are made from

other components, mainly carbohydrate. It is significant that the fatty

acid that is made in DNL is palmitic acid, the sixteen-carbon saturated

fatty acid. Bottom line: Carbohydrate in the diet raises saturated fat in

the blood. This was demonstrated most convincingly by experiments at

the University of Connecticut 11 described in chapter 7. These experiments

revealed that saturated fatty acid in the blood might be harmful, but that

its presence is more dependent on the intake of dietary carbohydrate than

dietary fat. Our students completely missed the boat on this one.

How much of a risk factor is high triglycerides? Well, it is impossible

to tell. The AHA tends to downplay the importance of triglycerides. This

is probably related to the need to avoid talking about low-carbohydrate

diets, since dietary carbohydrate restriction is the most effective method

of reducing high triglycerides except perhaps for total starvation. The

AHA has been pretty clear on not endorsing low-carbohydrate diets. On

the other hand, triglycerides have clearly become a focus of the ADA,

evidenced by their alarmist attitude toward sugar, or fructose in particular.

Sugar is carbohydrate, and regardless of whether an increase in fructose is

more or less effective than glucose in elevating triglycerides, both will raise

triglycerides.

9. In general, what effect does a low-fat diet

have on HDL-C (high density lipoprotein

cholesterol, the “good cholesterol”)? Student response (%)

□ increase 31

EE] decrease 39

□ no change 30

A low-fat diet reduces cholesterol, both “good” and “bad.” The bottom

line on the cholesterol problem: The literature tends to show that a

subtype of the LDL particle, the smaller LDL, is generally found to be

most atherogenic (contributing to a highest risk for CVD). High levels of

Nutrition in Crisis

small, dense LDL are referred to as “pattern B” and this pattern is most

dependent on the level of carbohydrate, rather than the level of fat. This

critical observation has had little effect on official positions of the AHA

or other agencies. It would appear that they don’t think LDL size matters,

but they have not said exactly why not. The AHA did, however, in 2000,

quietly remove their proscription against total fat. You didnt know that?

LDL particle size is not generally measured in a standard lipid profile,

and your physician is most likely to look at total cholesterol, or total LDL

cholesterol, to determine if you are at risk for heart disease. The recognized

surrogate for pattern B is the ratio of triglycerides to HDL. The cutoff is

3.5. 12 If your value is below that mark, you are at limited risk for cardio¬

vascular disease.

It is assumed that the reduction in triglycerides and increase in HDL

that is a consequence of a low-carbohydrate diet has a protective effect. As

suggested in the introduction, however, it is possible that, outside of well-

defined genetic abnormalities, what you eat might have no effect on your

risk of heart disease—or, more precisely, individual variations that might

predict a link between diet and heart disease have not yet been discov¬

ered. We will come back to this theme. The response to doubt separates

science from medicine and religion. It is revolutionary to even consider

the possibility that the diet-heart hypothesis as it s currently presented is

completely wrong. In science, doubt represents an opportunity. In religion,

you pray for relief from doubt.

10. The dietary change that is most likely to

increase the risk of cardiovascular disease is: Student response (%)

□ unsaturated fat saturated fat

0 unsaturated fat -> carbohydrate

□ carbohydrate -> unsaturated fat

□ carbohydrate saturated fat

□ saturated fat -> carbohydrate

□ saturated fat unsaturated fat

This is one of the most important observations, because it has been

known for so long. I described in the introduction my early research into

the literature and my dismay at the results of the Nurses’ Health Study, 13

Whaddaya Know?

which demonstrated that there was an increase in CVD incidence when

fat was replaced by carbohydrate, regardless of the type of fat being

replaced. This was astounding given the persistent low-fat message. One

would expect that the study, done more than fifteen years ago, would

have been the stimulus for a shift away from that low-fat message. That

didn't happen.

Assessing the Quiz Results

How did you do? Some critical facts were not known to most of our medi¬

cal students. In summary, whatever you knew before, the information that

you should take forward is this:

• The majority of the increase in calories in the epidemic of obesity and

diabetes has been due to a dramatic increase in carbohydrate consump¬

tion. The association between the observed behavior of the population

in macronutrient consumption and official advice to reduce fat and

increase carbohydrates might be causal. How could the advice not have

played a role? The bottom line, that reducing carbohydrate is the best

treatment for diabetes, suggests that carbohydrate must have played

some role in the origins of the epidemic, but this remains unknown.

What is established is that the progression of culprits—saturated fat,

red meat, white rice—that are “proved” daily by epidemiologic studies

to be causes can probably be excluded.

• The crux of the problem in controlling the epidemic of diabetes can be

summarized in the following statements:

• Dietary carbohydrates raise blood glucose in people with diabetes

more than other macronutrients.

• There is no biological requirement for carbohydrate (for anybody).

• Despite evidence to the contrary, health agencies recommend high

carbohydrates (more than 40 percent of total calories).

• The glycemic index and glycemic load are a weak form of low-carb strategy

The logical problems and the limited experimental proof of their efficacy

make their use questionable as a primary strategy. They might, however,

be of some use, since they still encourage carbohydrate restriction.

Nutrition in Crisis

• Calories are about processes, not substance, and looking ahead, different

processes (oxidation in the calorimeter versus metabolism) make differ¬

ent use of the calories.

• On the technical side: It’s good to pay attention to the difference between

intensive properties, such as calorie density, and extensive properties,

such as total calories. When you hear people say fat is inherently more

fattening, you need to know that doesn’t mean anything.

The major points about lipids from the second part of the quiz include:

• The terms saturated and unsaturated can only be applied to fatty acids,

the constituents of fats and oils. The composition of common fats and

oils is different from popular conceptions (e.g., beef tallow is almost half

oleic acid, the main fatty acid of olive oil.)

• The dietary change that has the greatest effect on cardiovascular risk

factors is replacement of fat with carbohydrate. Low-fat diets reduce

LDL, but low-carbohydrate diets reduce the important subfractions,

the pattern B, that are more atherogenic. Reducing carbohydrate also

improves HDL. The real question, however, is whether reducing carbo¬

hydrate diminishes the actual incidence of CVD. It would be difficult

to answer that question, but it must be considered. What you might not

have known before the quiz is that our current state of knowledge does

not provide evidence that what you eat will make any difference in your

risk of heart disease.

• The ambiguity or, more precisely, the near absolute failure of the diet-

heart hypothesis contained the seeds of the first low-carb revolution.

We’ll look at this in the next chapter, and I will explain why another

revolution is needed.

-CHAPTER 3-

The First Low-

Carbohydrate Revolution

T he first low-carbohydrate revolution dates from about 2002. As is

frequently the case in politics, the revolutionaries saw themselves

as a loyal opposition and probably didn’t think of their ideas as

particularly iconoclastic. Dr. Atkins was an established physician who was

undoubtedly only trying to help. He was surprised at the vehement backlash.

Unfortunately his response to criticism was less like that of John Adams

than that of John’s cousin, Samuel Adams, who was described in Don’t

Know Much About History as being better at brewing dissent than beer. It is

doubtful that much could have been done, though. Like political revolutions,

scientific revolutions usually have to be won more than once. Gary Wills

described the Gettysburg Address as a statement that the Civil War was a

second American Revolution. 1 People of my generation might see the civil

rights movement as a third. It’s often the same for revolutions in science:

We are taught that atomic theory comes from John Dalton, the Manchester

schoolteacher who proposed it in 1799, but atoms were not truly accepted as

real things, rather than convenient models, until Einstein nailed it in 1905.

The idea that dietary carbohydrate, sugars, and starches have some unique

power to make animals fat is very old. It would be hard to identify the first

farmer who fattened animals for market by feeding them grain. Brillat-Savarin,

the father of modern gourmet cooking, generalized the fattening principle to

human beings, and claimed that there were folks who were “carbophores”—in

fact, he admitted to being one himself. 2 The mechanism—the anabolic effects

of the hormone insulin, stimulated primarily by the sugar glucose—was a

well-established physiologic phenomenon before the first low-carbohydrate

revolution. The scientific literature provides many examples of weight loss

from carbohydrate restriction and links it to something beyond the reduction

Nutrition in Crisis

in calories that usually comes with such a diet. Although it has been around

in one form or another for a long time, carbohydrate restriction only became

revolutionary with the ascendancy of a kind of low-fat nutritional-medical

monarchy, backed by powerful influences.

The first edition of the original Atkins book 3 appeared in the 1970s,

right around the time of the codification of low-fat as the desirable diet.

The Atkins diet was so vehemently denounced that it gave rise to congres¬

sional hearings. An amusing moment occurred when the American

Medical Association (AMA) asserted that one of the dangers of a low-

carbohydrate diet for weight loss was that it might lead to anorexia. Overall

it was an indication of the unwillingness of the medical professions to

tolerate dissent.

As in political revolutions, the first low-carbohydrate revolution was

stimulated by a kind of manifesto, a document that historians now describe

as being a call to action. The equivalent of Thomas Paine s Common Sense ,

which fired everybody up for the American Revolution, was a 2002 article

by Gary Taubes in the New York Times Magazine titled “What If It s All

Been a Big Fat Lie.” 4 Later expanded into the book Good Calories, Bad

Calories , s the article documented the political ascendancy of the low-fat

paradigm and the establishment of something like the Court of Low~Fat.

The AMA, the AHA, and a number of influential physicians were all

received at court. The media and the government, including the McGovern

committee, went along with it. Tom Naughtons comedy documentary Fat

Heacf* includes a clip of McGovern explaining in 1977 that Congress did

not have the luxury of waiting for all the science to be in.

The McGovern hearings began a pattern of ignoring dissenting voices,

like that of Philip Handler, head of the National Academy of Sciences,

who testified that there was little science behind this rush to judgment. I

recognized Handler as part of White, Handler, and Smith, the group of

authors that wrote one of the few comprehensive biochemistry texts at the

time—that is to say, he was a very well-known biochemist.

The Lipophobes and Their Opposition

There were many experimental and clinical trials that set out to prove

the diet-heart hypothesis. The first, the Framingham Study, which

The First Low-Carbohydrate Revolution

continues today, is a massive survey of the behavior of residents in this large

Massachusetts town. The original results showed no effect of dietary total

fat, saturated fat, or cholesterol on CVD. The study did initially find an

association between blood cholesterol and CVD, but the correlation wasn't

a knockout, and it became weaker as the study continued. This original posi¬

tive but unsustainable association is now a classic in epidemiology, taught in

statistics classes. The fact that diet did not correlate with CVD is less often

discussed, and in the later data on cholesterol, some age groups (men, 48-57

years old) actually displayed greater risk with lower cholesterol.

The results of the Framingham Study were buried for years until the

statistician Tavia Gordon had them published in 1968, almost ten years

after the data had been analyzed. Publication of the data should have been

the death of diet-heart hypothesis right there. If fat was as bad as they

said, there wouldn't have been a single study that failed to prove it. Not

one. In actuality, almost every one of the dozen or so large trials conducted

since Framingham has also failed. Science, however, was not the major

force behind the denial of evidence. The lipophobes, as Michael Pollan

calls them, 7 continued to dismiss each experimental failure as the loss of

a minor battle in a war where victory would surely fall to them: just one

more big clinical trial, just another hundred million bucks and you will see

how bad fat is. Even in 2001, when the AHA removed its proscriptions

against total dietary fat, it was done without fanfare. As a result, it is likely

that most consumers think that AHA still recommends reduction in over¬

all dietary fat. They're still down on saturated fat, and of course, trans-fat

(the latter being a moot issue since it's been almost entirely removed from

the food supply), but the truth is that the AHA has given up on total fat.

Although preceded by other exposes, Taubes's Good Calories, Bad

Calories was the most compelling presentation of how nutritional science

had been taken over by lipophobes. Numerous retellings have followed.

Nina Teicholz's The Big Fat Surprise* is of comparable literary quality to

Good Calories, Bad Calories , and is more explicit in its condemnations of

the players. Ultimately, with control over the NIH, the low-fat mafia could

now resist all scientific argument and dismiss all of the experimental fail¬

ures to provide any sort of reasonable case. The ascendancy of low-fat was,

and still is, coupled with a special hatred for low-carbohydrate diets and

especially for their main exponent, Dr. Atkins, even after his death.

Nutrition in Crisis

The low-fat idea wasn’t good to begin with, but of all the tests of the

idea that failed, one after another, nothing was more embarrassing than

the Womens Health Initiative (WHI), which reported in 2006: “Over

a mean of 8.1 years, a dietary intervention that reduced total fat intake

and increased intakes of vegetables, fruits, and grains did not significantly

reduce the risk of coronary heart disease (CHD), stroke, or CVD in post¬

menopausal women.” A multicenter, nearly $400 million study, the WHI

had assigned 19,541 postmenopausal women to the dietary intervention

and had a control group of29,294 women in a free-living setting. As such,

its failure should have been a bombshell.

It was not long before Dr. Elizabeth Nabel, director of the National

Heart, Lung, and Blood Institute of the NIH, appeared on television to

assure the nation that the recommendations had not changed regardless

of the study’s findings. You really did still need to reduce saturated fat,

she insisted. Nothing’s changed, despite the study they funded explicitly

showing that change was needed.

The situation was serious. The refusal to accept the failure of a scientific

test, and the stubborn insistence on doctrine, caused palpable harm. The

WHI women weren’t getting any better, and the population at large, doing

its best to adhere to low-fat, was getting fatter and more diabetic during

this period. Refusal to see the WHI for what it was represented a clear

statement that the lipophobe movement, starting at the top at the NIH,

was going to stonewall any effort to change.

“The Shot Heard ’Round the World”

IfTaubes’s “What If It’s All Been a Big Fat Lie” was the Common Sense

of the first low-carbohydrate revolution, then the “shot heard ’round the

world” was the report by Gary Foster and coworkers 9 showing that the

Atkins diet actually improved markers for cardiovascular disease, the lipo-

phobes’ main “concern” about low-carbohydrate diets.

Foster’s demonstration had a big impact because he spoke for the whole

nutritional establishment. He later described, in public lectures, how

he and his collaborators had been having lunch at a scientific meeting,

bemoaning their inability to sweep the Atkins diet from their sight. They

decided to get a grant to trash the diet, and so they did. One suspects that

The First Low-Carbohydrate Revolution

their intent was clear in the grant application: not to test the efficacy of

the Atkins diet—that wouldVe been nearly impossible to fund—but to

show just how bad it was. So they carried out a one-year study comparing

a low-carbohydrate diet modeled on the Atkins diet with a low-fat diet.

What they found, contrary to the conclusion they’d hope for, was that “the

low-carbohydrate diet produced a greater weight loss (absolute difference,

approximately 4 percent) than did the conventional diet for the first six

months.”This part was not a surprise even to the authors. Everybody knew

somebody who had lost a lot of weight on the Atkins diet, and nutritionists

had more or less accepted the idea that low-carbohydrate diets were good

for weight loss, though they usually insisted that it was “just a reduction

in calories.” (If youVe tried to lose weight, you know that there is no “just”

about it.) The kicker, however, was that Foster reported:

After three months, no significant differences were found

between the groups in total or low-density lipoprotein cholesterol

concentrations. The increase in high-density lipoprotein cholesterol

concentrations and the decrease in triglyceride concentrations were

greater among subjects on the low-carbohydrate diet than among

those on the conventional diet throughout most of the study.

Both diets significantly decreased diastolic blood pressure and

the insulin response to an oral glucose load. 10 (Emphasis added)

In other words, the low-carbohydrate diet was better on HDL (“good

cholesterol”) and especially on triglycerides. Most importantly, there was

no increase in LDL, which is what your doctor uses as the traffic light for

determining whether you need to be prescribed statins.

More Work Needs to Be Done

The conclusion of the Foster study: “Longer and larger studies are required

to determine the long-term safety and efficacy of low-carbohydrate, high-

protein, high-fat diets.” This strange conclusion indicates the persistent

difficulty in making progress. Low-fat diets do worse on most markers,

and are at best a draw on the others, yet we are supposed to be worried

about the low-carbohydrate diet? If the low-fat diet is worse, shouldn’t

Nutrition in Crisis

we be worried about long-term safety and efficacy of that diet instead?

Somehow the conclusion consistently drawn from Foster s experiment was

that the “diets were the same at one year”—that it was a draw. There are

probably sporting events where the champion keeps the title in the case

of a draw, but the idea that the low-fat diet was some kind of champion

with a long-term record of success is absurd. The allegedly “prudent” and

“moderate” low-fat diet is exactly the one that gave us the epidemic of

obesity and diabetes in the first place.

Ad Lib Versus Calorie Restriction

Beyond the obvious bias, Foster's study was compromised in its experi¬

mental design. People in the low-fat group were directed to consume an

explicitly low-calorie diet: They were required to eat what they were told. In

the low-carbohydrate group, on the other hand, participants were allowed

to eat anything that they wanted as long as they kept carbohydrates low.

(Even if you believe that the diets were actually equal, which diet would

you go for?) This protocol was used because Foster et al. were not testing

the principle of reducing insulin fluctuations as a means of controlling

metabolism. The study was not testing, as in the title, “a low-carbohydrate

diet for obesity.” Instead it was testing the Atkins diet and perhaps, in the

authors' minds, Atkins himself. (After all, the Atkins diet said that you

didn't have to count calories—anathema to traditional nutritionists.) The

experiment, in reality, was testing two principles: that carbohydrate restric¬

tion means greater satiety, allowing you to regulate calories implicitly; and

that the Atkins diet, calorie for calorie, is more effective for weight loss.

Testing both at once was more demanding than isolating the variables.

The low-carbohydrate diet's better results might have been due to either or

both principles, but its impossible to know because of the flawed experi¬

mental design.

The dietary protocol was not the only problem. Data were analyzed

according to a bizarre method known as intention-to-treat (ITT). In this

method, data from the subjects who had dropped out of the study were

included in the results by “imputing” values based on previous measure¬

ments. The difference between “imputing” and making stuff up is hard to

figure out. ITT doesn't make any more sense than you'd think from this

The First Low-Carbohydrate Revolution

brief description, but that’s simply how things are (see chapter 16 for a

more in-depth discussion). Beyond the obvious lack of common sense, an

ITT will always make the better diet look worse than it actually is. In this

event, though, those in favor of carbohydrate restriction were sufficiently

happy to see the positive outcome, and they were disinclined to be too

critical of the methods. At face value, the lipophobes had their shot and

they lost. They tried to maintain a facade of impartiality while still putting

the burden of proof on low-carbohydrate, but it still didn’t help their case.

So, what happened to the first low-carbohydrate revolution? Why didn’t it

move forward? How did low-fat loyalists prevail in the face of such strong

scientific evidence?

What Stopped the Revolution?

It was an opportunity. The public had a chance to see if the low-

carbohydrate idea would work. Many did try it, and many had

great success. Popular articles were written about the phenomenon.

Low-carbohydrate wasn’t equally effective for everyone, but frequently,

it seemed miraculous. People described how the “pounds melted away.”

So why didn’t it move forward? First, there was poor understanding of

what actually made the diet work; and second, there was a proliferation

of products designed to make low-carbohydrate “easier,” because it was

perceived, incorrectly for the most part, as a difficult diet strategy. This

allowed the company Atkins Nutritionals and other similar ventures to

sell a lot of products, many of which were not always helpful. They are

still doing that. These products help some dieters stay on track, but their

artificial character made them suspicious, and the move toward natural

food has made their reputation worse. The low-carbohydrate community

itself mostly emphasized low-carbohydrate and, oddly, there was little

disappointment when Atkins Nutritionals declared bankruptcy in 2005,

though it has since been resurrected and continues to offer substitutes for

the carbohydrates that you are giving up. Particularly troubling was the

proliferation of products containing sugar alcohols—carbohydrates that

are digested slowly, if at all, and were therefore presumed to not contrib¬

ute to blood glucose. Untested and poorly understood, sugar alcohols

gave some people intestinal problems, but more importantly, they cast

Nutrition in Crisis

what should have been a straightforward diet in a slightly bizarre light.

Yet in the end, it was not Atkins Nutritionals or any other company, but

the nutritionists and the professors of medicine who stopped the first

low-carbohydrate revolution. They had a million objections and they got

the media on their side.

The Problem with Dietitians

What was remarkable about the whole state of affairs was that the low-fat

strategy had failed in competition with a real alternative. Low-fat could

not compete with low-carbohydrate, even with the experiment set up in

its favor and the authors putting a positive spin on the data. It was a direct

challenge to nutritional orthodoxy, but the stalwarts of the old nutritional

order did not go gentle into that good night. What torpedoed the first

low-carbohydrate revolution were the nutritionists. They had the chance to

tell the public, Tf you do want to try a low-carbohydrate diet, this is what

we recommend.” Instead, they acted as if Fosters paper had never existed

and they ignored those studies further supporting carbohydrate restriction

that followed it.

Nutrition has never been highly thought of. The field derives from the

practical job of making institutional menus. The advances of physiology

and biochemistry meant that nutrition increasingly overlapped with more

solid science, but the field was, and is, very slow to change. Nutritionists

have attempted to put on the mantle of professionalism. They are currently

trying to establish the newly renamed Academy of Nutrition and Dietetics

(AND), and to secure their status as the sole voice on nutrition, with the

right to legally repress anybody else. Recently they tried to stop low-

carbohydrate proponent Steve Cooksey from “offering counseling” on his

blog despite appropriate disclaimers. Thankfully the courts decided that

the First Amendment was still the law of the land and left Cooksey s

accusers with egg white on their faces. However, the case, now replayed in

several similarly ugly conflicts and described in later chapters, highlights

the adversarial state of nutritional science and the intolerance of dissent.

For those familiar with computer logic and with a taste for computer-nerd

humor, my own group may rename itself OR (Objective Research) and

plans to refer to the Cooksey affair as NAND-gate.

The First Low-Carbohydrate Revolution

Underlying all of the resistance is the idea that only long-term, large-

scale studies are important, an inaccurate assumption that nutritionists use

as a basis for ignoring the smaller studies that are better controlled and

provide more information. Small studies are not worse—they are generally

better. More important, the quality of a study is not simply determined

by the size or length of the study. But once again, the implicit assump¬

tion remains that the long-term studies have supported a low-fat diet—a

prudent diet, a diet of moderation, a diet with proven success, a diet that

can work for all of us. There is no such diet and there never was. The long¬

term studies have failed, almost every .single one: the Framingham Study,

the Oslo Diet Heart Study, 11 the Western Electric Study, 12 The Women's

Health Initiative, 13 and probably two dozen others. They showed no value

in reducing dietary fat or saturated fat for prevention of heart disease or any

other health conditions, and in the biggest trial of all, the “full-population

trial” of Americans during the obesity epidemic, it was increased carbohy¬

drate, not fat, that was actually harmful.

To be fair to nutritionists, doctors also played a part in the deception.

Undeterred by their lack of training or experience in biochemistry or

nutrition, it was de rigueur for junior faculty in a department of medicine

to write a review trashing the Atkins diet. Some of these critiques even

stated that the main flaw of low-carbohydrate diets was their failure to

conform to the USDA dietary guidelines and other institutional recom¬

mendations. In other words, they faulted a diet whose central premise was

that the USDA recommendations were bad, for not conforming to those

recommendations. This is a known logical fallacy that was called “begging

the question” before the phrase lost its original meaning: using the ques¬

tion as part of the answer.

Failure to Accept Failure

Stepping back and looking at the big picture, the most striking thing was

the inability of low-fat diets, even those low-fat diets that did lower choles¬

terol, to provide a significant impact on cardiovascular outcome, or really,

on anything else. Very large, very expensive clinical trials of low-fat dietary

strategies failed, and yet our tax dollars continue to pay for similar trials.

Even in those cases where we didn't have outcome data about heart attacks

Nutrition in Crisis

or deaths, and instead had to look at the risk factors—the different choles¬

terol forms: HDL, LDL, and their subfractions—it turned out that the

effect of reduced fat was at best ambiguous, whereas dietary carbohydrate

typically had the major effect. As carbohydrates were increased, most of the

risk markers got worse. (I should note that these markers and their associa¬

tion with outcome were not sufficient to attribute cause, but that had not

stopped such interpretations when it was low-fat that reduced risk factors.)

If we go beyond the original idea of total blood cholesterol as a major

risk factor (less than half of the people who have a first heart attack have

high cholesterol), and if we instead look at the different forms of the

lipoprotein particles that are actually measured in clinical tests of blood

cholesterol, carbohydrate restriction becomes the “default diet,” the one to

try first for general health. It is important to acknowledge, however, that

while low-carbohydrate diets look better for CVD risk factors, we have the

same problem that we have with fat: There is little in the way of evidence

that lowering carbohydrate can actually prevent CVD. Given its success

in treatment of the collection of health markers referred to as metabolic

syndrome, it would be surprising if reducing carbohydrate did not help in

prevention—but at this point, what we know is very little. We are left with

the real possibility that there is nothing at all to the diet—heart hypothesis,

and that diet might not be a major player at all in CVD, except in cases of

well-defined genetic conditions. Its very surprising, given our current view

of things, and its likely to change as we learn more—but you have to go

with the data.

PART 2

Nutrition and

Metabolism

-CHAPTER 4 —

Basic Nutrition

Macronutrients

C arbohydrate, fat, and protein represent the major macronutrients

because of the large quantities in which they are consumed.

Micronutrients include vitamins and minerals, which are only

taken in small amounts. It has become common to refer to foods as either

“nutrient-rich” or “nutrient-dense,” or not, according to whether they re

thought to have high amounts of micronutrients. Some people, including

me, are annoyed by this lack of precision. After all, macronutrients are also

nutrients, and it would make more sense to say “micronutrient-rich” if that

is the intended meaning. Readers should be aware of this lack of precision

moving forward.

This chapter describes the primacy of macronutrient composition for

metabolic effects. During the years from 1970 to 2000, roughly the period

when observers began to notice an obesity epidemic, people consumed

an excessive number of calories, the majority of which came from carbo¬

hydrates. The total amount of protein, usually the most stable part of the

American diet, did not change. Fat, if anything, went down.

The Basics of Carbohydrate Chemistry

Chemically, the class of compounds called carbohydrates includes: simple

sugars (monosaccharides), such as glucose and fructose; combinations of

two simple sugars (disaccharides), such as sucrose, which is made up of one

glucose and one fructose; polymers (polysaccharides), such as starch and

glycogen; and derivatives of the sugars (e.g., the so-called sugar alcohols).

Alcohol—that is, ethanol—is not a carbohydrate despite what you

might hear on YouTube. Whatever the extent to which sugar can make

Nutrition in Crisis

Starch {corn, wheat, potato, etc.)

Sucrose (table sugar)

Enzymatic

digestion of

corn starch

Gl digestion

starch

Gl digestion

Glucose (e.g., corn syrup)

Glucose

Fructose

Enzymatic

conversion of Glucose,

glucose to fructose Fructose Mixture

(e.g., HFCS)

Figure 4.1. Structure and transformations of the common carbohydrates. Starch is

a polymer of glucose. In digestion, the glucose units are released and absorbed as

such. Sucrose is a dimer of fructose and glucose, and digestion produces the two

monosaccharides.

you as loopy as alcohol, the two compounds are simply not the same

on a chemical level. A horse is not a dog. One of the reasons that we

make Pre-Meds study organic chemistry is in hope that the precision in

naming organic compounds will carry over into pharmacology. It is likely

that manufacturers started spelling klonopin (an antidepressant) with a

k because they didn't want physicians to accidentally prescribe clonidine

(an antihypertensive).

In terms of formal chemistry, sugars are polyhydroxy aldehydes and

ketones. The most common sugar is glucose, and it almost always cyclizes

(folds up) in the form of a hexagon, at least in aqueous solutions. Fructose

can also form a six-membered ring, but it is more likely to cyclize in a

pentagonal shape, as shown in figure 4.1, which provides a simplified

representation of the common sugars and related polymers. You can see

from the figure that starch is a polymer of glucose (polysaccharide), and

that it breaks down into simple sugars during digestion. If you never did

the experiment in grade school, you can try chewing a piece of bread for

Basic Nutrition

several minutes. The sweetness that develops is the result of digestive

enzymes in saliva that catalyze the conversion of the bread’s starch into

sugar (glucose). Not all starch molecules are as simple as the one depicted

in figure 4.1: While some starch molecules, called amylose, do have a linear

structure, other types, called amylopectin, have many branch points.

Glycogen: Glucose Savings Account

Glycogen is a highly branched polymer of glucose that serves as the stor¬

age and supply depot for body glucose flux. The liver, a kind of command

center of metabolism, provides glucose for other organs, and muscle, the

main consumer of glucose, stores glycogen for its own use. The liver has the

highest concentration of glycogen, but there is more muscle tissue in the

body, so it is muscle that possesses the highest total amount of glycogen.

Glycogen is a dynamic storage site. The extensive branching means that

there are a lot of ends from which glucose units can be chopped off as

needed. It also means that glycogen occupies a lot of space. Children with

one of the inborn errors of metabolism known as glycogen storage diseases

will have visibly distended abdomens due to the liver increasing in size

(hepatomegaly) to accommodate glycogen stores, which in these diseases

can no longer be broken down.

We think of glycogen as desirable because of its association with endur¬

ance in sporting events. This association is the basis for carbohydrate

loading the day before a marathon, but, even in the area of athletics, things

are not clear-cut: Marathons are mostly run on fat, even if you are not

adapted to a low-carbohydrate diet. Jeff Volek and Steve Phinney have

recently trained marathon runners and other elite athletes to perform on

very low-carbohydrate ketogenic diets. The athletes do well. They win

races, describe perception of greater stamina (not “hitting the wall”), and,

surprisingly, they maintain glycogen stores—that is, they run on fat instead

of depleting glycogen. It should be noted, however, that a ketogenic diet

for athletes depends on a training and accommodation period.

On a low-carbohydrate diet, glycogen storage tends to be reduced, typi¬

cally by around 60 percent with very low-carbohydrate intake (< 100 g/day).

However, as I discussed earlier, metabolism does not run on mass action

(how much is available) but rather on hormones and enzymes—so under

Nutrition in Crisis

conditions of low-carbohydrate intake, glycogen will be replenished by the

glucose produced from gluconeogenesis, thus maintaining storage. It is

important to understand that gluconeogenesis and glycogen metabolism

are really one process: Glucose synthesized from protein might be stored as

glycogen and then only later appear in the blood. An important feature of

a carbohydrate-restricted diet, however, is the switch to a metabolic state

that runs on fat rather than carbohydrate.

Glucose is at the center of metabolism. Looking ahead to chapter 5,

the main theme in human biochemistry is that there are two major fuels:

glucose and acetyl-CoA (derived largely from fat and pronounced ASS-a-

teel Co-AY). Two fuels and two goals: provide energy and maintain blood

glucose at a constant level. Too little blood glucose (hypoglycemia) is not

good because some tissues, particularly the brain and central nervous

system, require glucose—but too much (hyperglycemia) is also a problem.

Glucose is chemically reactive and interacts with several biomolecules

through the process of glycation. In the case of protein—the major

target—glycation might inhibit biologic activity or may lead to clearance

from the cell or circulation. You need glucose, but again, you don t have to

ingest any—your liver makes it from protein and other compounds.

A major point that will reappear throughout this book is that carbo¬

hydrate and protein can be turned to fat, but while glucose can be made

from protein, with few exceptions, you cant make glucose from fat. In the

context of the two fuels, this means that glucose can provide acetyl-CoA,

but acetyl-CoA cannot be converted to glucose.

Lipid Chemistry: Good Fats, Bad Fats

Most of the fatty acids have common names because they were discovered

before we had systematic chemistry and professional panels to set the

rules. Some of the names tell you how they were discovered—for example,

palmitic acid is found in palm oil, oleic in olives, and caproic and capryllic

acids smell like goats (genus Capra ).

You 11 hear a lot of disclaimers about good fats and bad fats, but in one

way or another, the government, private agencies, and individual research¬

ers are still recommending that you reduce the total amount fat. The

recommendation for reduced fat might be accompanied by an explanation

Basic Nutrition

that the type of fat is more important than the total amount, but it usually

boils down to some contradictory statement like: “Fat is not bad. Only

saturated fat is bad. Eat low-fat foods.” The big targets are saturated fat,

and of course, trans-fat. The juxtaposition of saturated fat, a natural part

of the food chain that has existed throughout human metabolic evolution,

and trans-fat, a by-product of industrial modification that was never seen

before the twentieth century, is an indication that it is politics, not science,

that is at work here. (Note that, as in figure 2.4, trans- is a general chemi¬

cal term, and there are naturally occurring trans-fats that might actually

be beneficial. However that is not what is discussed here. As described

below, trans-fats means the by-product of turning unsaturated oils into

solid form.)

The AHA website provides a truly maniacal cartoon video on the

hideous Sat and Trans brothers: “They're a charming pair, Sat and Trans.

But that doesn't mean they make good friends. Read on to learn how they

clog arteries and break hearts—and how to limit your time with them by

avoiding the foods they're in.'' 1

What's missing from the website is the story behind trans-fat. The

crusade against dietary saturated fat, which the AHA and other health

agencies fought so vigorously, led to a search for alternatives to butter and

lard. It is important to understand that this mission was led by physicians,

rather than physiologists or biochemists. Some accused those in charge of

trying to carry out a grand experiment with the American people as guinea

pigs, but there's no stopping zealots. Butter was seen as the quintessential

high-saturated-fat food, and it was clear that no progress could be made

without deposing it and installing a substitute.

Wide availability of vegetable oils provided a potential alternative to

butter, lard, and other sources of saturated fat. Vegetable oils, however,

are liquids, and at least for baking, have to be converted to a more useable

form such as margarine or Crisco. You can do this through the process

of hydrogenation: Unsaturated oils are “unsaturated” with respect to how

much hydrogen is attached to the carbon atoms. Hydrogenation turns

some of the unsaturated fatty acids to saturated fatty acids, in effect

converting the oil to a solid form. Unsaturated fatty-acid molecules have

more rigid structure and are harder to pack into a solid—which is why

unsaturated fats tend to be liquid. Converting some of the unsaturated

Nutrition in Crisis

fatty adds into saturated fatty acids made the material more solid and

easier to work with. A side reaction in the manufacture of hydrogenated

oil, however, is the conversion of some of the oil to the tram- form (the

names refer to structure).

As you saw in figure 2.4, double (unsaturated) bonds can be cis - or

trans-, most naturally occurring fatty acids are cis-, so trans-fatty acids are

not normally processed to a great extent. (Note: trans-fats are unsaturated;

cis- or trans- can only refer to unsaturated bonds.) Some biochemists—

notably Mary Enig—tried to stop the introduction of products containing

trans-fatty acids, but most chemists did not know much about trans-fatty

acids and with little support, the low-saturated-fat forces prevailed. At

least for a while. When it turned out that trans-fat was the form that

correlated best with cardiovascular disease, it was a ready-made scape¬

goat, and presumably because it was used in so many artificial products,

it became the target of health agencies. Of course these groups did not

mention that trans-fat in the food supply arose from their own campaign

against saturated fat. In the end, trans-fat is a very small part of the diet,

and its risk is probably greatly exaggerated. However there is widespread

support for its removal among the public, and it is required by law to be

removed. Because there’s nothing inherently good about trans-fat, nobody

wants to defend it.

In contrast to trans-fat, saturated fat has always been part of the

human diet and is a normal part of metabolism. Saturated fatty acids

are synthesized in your body through a process that is stimulated by a

high-carbohydrate diet. This has been known for years. The process

is called de novo lipogenesis and is in the biochemistry textbooks, and

while it is acknowledged that high dietary carbohydrate led to de novo

synthesis of saturated fatty acids, the idea is immediately forgotten when

official dietary recommendations are written. So where did we get the

idea that fat—and saturated fat, in particular—is unhealthy? Again, the

story has been told many times, most succinctly in 7 he Rise and Fall of

Modern Medicine , 2 most engagingly in Good Calories Bad Calories , 3 and The

Big Fat Surprise , 4 but the death knell for the low-fat idea was, or should

have been, the meta-analyses from several groups. 5 A collection of studies,

some going back twenty-five years, was not able to find any risk in dietary

saturated fat.

Basic Nutrition

The Glucose-Insulin Axis

It remains extremely difficult to eat a zero-carbohydrate diet. It is only

very recently that the so-called carnivore diet has become popular. There

are no real studies—in the absence of believable or reliable guidelines,

self-experimentation is common. Carbohydrate is an inherent part of

almost all human diets. There is, however, no biological requirement for

carbohydrate in the same way there is for protein, or, to a lesser extent,

for fat. Biologically speaking, we’ve yet to figure out whether differences

in carbohydrate type matter, but we do know one simple principle: “High

in carbohydrates” is bad advice for people who are overweight and espe¬

cially for people who have diabetes or metabolic syndrome. For many of

them, reducing carbohydrate can constitute a cure—and while we don’t

know for sure, it is likely that excessive carbohydrate consumption plays

a role in how people get fat and diabetic in the first place. We have a

grasp on the basic underlying science: Carbohydrate, directly or indi¬

rectly, through the hormone insulin, controls the response to other foods.

Hormonal systems are very complex but surprisingly, in metabolism,

insulin has an overpowering effect. It is an anabolic hormone, meaning

it stimulates the buildup of body protein and storage of body material.

Insulin is predominant in the storage of nutrients: It encourages the stor¬

age of fat and carbohydrate, and increases the synthesis of body protein.

Hormones communicate with living cells through the docking proteins

known as receptors, and stimulation of the receptors, in turn, triggers the

metabolic machinery within the cell.

Carbohydrates, especially glucose, constitute the major stimulus for

secretion of insulin. Persistent high-insulin fluxes will bias the body

toward storage, and particularly storage of fat. So, even though dietary fat

is important, it plays a more passive role than is generally said. The rate at

which fat gets stored ultimately depends on the hormones that are pres¬

ent. A high-fat diet with high carbohydrate is very different from a diet

with the same amount of fat but lower carbohydrate. Higher carbohydrate,

through its effect on insulin, might make the effect of the fat deleterious

instead of beneficial. That’s the bottom line.

We teach medical students that the flow of dietary fat into stored fat,

or into the lipid markers used to characterize cardiovascular risk, is like

Nutrition in Crisis

the flow of water through a faucet. Carbohydrate controls the faucet. If

carbohydrate intake is low, the flow stops and fat is oxidized. In the pres¬

ence of high carbohydrate, on the other hand, insulin increases the rate of

fat storage. This is simplified, of course, but not radically so. The main idea,

that it is a control problem, not a too-much-stuff problem, is on target.

Diet Comparisons and the Medical Literature

Fat metabolism is only part of the picture. Insulin is a global hormone

that regulates carbohydrate and protein metabolism. The evidence for the

regulatory function of insulin is there, but not everybody wants to face it.

If you confront naysayers, they tell you that the problem is very complex.

So, how does it actually play out in the real world? It should be easy to

get the answer. Lets consider one experiment that is pretty clear, or at

least one that should have been clear: Bonnie Brehm et al. assigned fifty

healthy, slightly obese women to an ad lib low-carbohydrate diet or an

energy-restricted, low-fat diet, for four months. 6 The results were that the

low-carbohydrate women lost significantly more weight.

Brehm et al.’s study yielded both good and bad news. The bad news

first: The actual spread of individual values was very large, bigger than

what is shown in the paper. Put simply, there are different ways of showing

variation in the data, and the method used in Brehm et al.s paper—the

standard error of the mean (SEM)—always makes data look better than

they are. The actual spread of values is about four times the size indicated

in the publications. So the bad news is that outcomes on the two diets

are not highly predictable and from the presentation of the data—group

statistics—you cant tell who did what.

The good news is that the big winners must have been the people in

the low-carbohydrate group. You don’t know if most people in the low-

carbohydrate group did better than most people in the low-fat group, or,

alternatively, if there were a few really big winners in the low-carbohydrate

group that tipped the scale, but you can at least be sure that low-carbohydrate

has the possibility of the biggest payoff.

And then there’s the really good news: If you read the Methods section,

you’ll find the experiment followed the same protocols as Foster’s famous

2002 study: “One group of dieters was instructed to follow an ad libitum

Basic Nutrition

diet. , . . The other group of dieters was instructed to follow an energy-

restricted, moderately low-fat diet with a recommended macronutrient

distribution of 55% carbohydrate, 15% protein, and 30% fat.”

In other words, if you were in the low-fat group, you had to count calo¬

ries or follow the low-calorie meal plan that they gave you, whereas if you

were in the low-carbohydrate group, you could do whatever you wanted

as long as you kept carbohydrates low. Even if it s a tie and the results are

the same, most of us are going to be happier doing the low-carbohydrate

diet since there is no restriction on calories. This is how it was done in the

landmark Foster paper and the same pattern has continued since.

As a scientist, you dont always read the Methods section of a paper in

great detail unless you are planning to repeat the experiment yourself or there

is something unusual in the way the experiment was carried out. If there is

something important in the methodology, it should be described in the body

of the paper. I admit that when this study was published in 2005,1 didn’t

realize that the standard methodology was a low-calorie diet pitted against an

ad lib low-carbohydrate diet. The rationale for this, of course, was that since

the Atkins diet did not restrict calories, participants in the low-carbohydrate

arm should only be instructed to reduce carbohydrate intake. This might be

reasonable from a clinical point of view, but it does make “the diet,” rather

than carbohydrate—which is easier to control—the independent variable.

Although Brehm et al.’s experiment is far from the best, it is repre¬

sentative of the type of experiment in which the low-carbohydrate diet

does better. However, you cannot ignore the big spread in the values. In

2006 a meta-analysis—that is, a reexamination of previous studies—was

performed by Alain Nordmann et al. As HI explain further in chapter 17,

meta-analysis is a weak, possibly useless method. The idea is to average

previous studies, but most of us agree that averaging errors makes things

worse, not better. That said, a meta-analysis usually does give you a chance

to see the results from several different studies. The conclusion from

Nordmann et al.’s meta-analysis was that low-carbohydrate diets lead to

better weight loss. Most of the studies found the low-carbohydrate diet to

be more effective at six months but, again, it had no advantage by one year.

The reason things got worse after six months is that the experimenters let

them get worse: They didnt know how to keep everybody on track. When

its your personal diet, you wont let that happen.

Nutrition in Crisis

There are now numerous studies that present the same picture. Although

the shorter duration studies turn out best for low-carbohydrate, you almost

never see a study showing that low-fat is better. Unfortunately, nutritionists

tend to consider a draw between the two diets as a win for low-fat, whereas

when low-carbohydrate wins, the results are simply ignored, or, as in Foster s

study, its declared that “more work needs to be done.” And so we have the

recommendations that we do and we are in a current nutritional mess.

A common objection to these studies is that as they depend on diet

records—as most do—they are prone to error, and the subjects in the

studies often misreport what they eat. Errors in reporting have been

documented, but the data are not completely inaccurate—usually they

are about 80 percent accurate. All experiments have error, however. It is

only a question of how you deal with the error. For inaccuracy in dietary

reporting to account for the difference between diets it would be necessary

for subjects in the low-fat group to underreport what they ate and for the

low-carbohydrate people to overreport what they ate, or both. These errors

are certainly possible, but again, from a practical standpoint, it might be

good to be on a diet where you think you ate more than you actually did.

Animal Models, Human Subjects

To be fair to the low-fat doctrine, it is easy to be misled. Animal studies

are critical in biological science and it seems that mice, especially those

bred for laboratory work, will get fat on high-fat diets even without any

carbohydrate. 7 How is that possible? People dont usually get fat on high-

fat diets, or at least they dont overconsume fat to such a high degree unless

their diet is also high in carbohydrate. Mice provide a good model for

human metabolism in other cases, too, so this is a serious difference to

consider. While carbohydrate is key in human metabolism, the same is not

necessarily so in rodent metabolism, where a high-fat diet can bring on

obesity, diabetes, and cardiovascular disease even in the absence of carbo¬

hydrate. We dont yet have a theory to encompass the differences between

animal models and human subjects. If our understanding of the catalytic

role of insulin is correct, however, then it might well be that mice (who

normally live on high-carbohydrate diets) maintain a functionally high

level of insulin all the time. In other words, the bias toward an anabolic

Basic Nutrition

state that occurs with high-carbohydrate diets in humans might always be

“on” in mice.

Whatever the explanation, its hard for many to recognize that the

animal model system we have used so extensively to understand humans

might actually have the most impact precisely because there is a difference

between how people and mice respond to certain treatments. In this case,

animal models might offer a clue to human behavior and physiological

response through its differences from that of the mice.

The real problem, though, is that we have not faced the results of the

practical, experimental tests in humans. We have large, expensive clinical

trials with a consistent and reliable outcome: There is no effect of dietary

fat on obesity, cardiovascular disease, or just about anything else. The

unwillingness to face these failures makes this a remarkable phenomenon

in the history of medicine—that it persists in a period of sophisticated

science and technology makes it nearly unbelievable. Unbelievable is the

keyword. One understands that science can be incomplete or might have

flaws, but it is hard to understand how the whole establishment could

maintain such a misguided opinion on the diet-heart hypothesis. How

can they keep doing the same experiment over and over without success?

How can they get away with it, and just as importantly, why would they

want to get away with it?

In science, excluding a theory is always easier than showing consistency.

In this case, if the fat-cholesterol-heart connection was as inescapable a

risk as they make it out to be, then none of these big studies should fail. Not

one. In fact, almost all fail. Yes, there have been increasing admissions that

high-carbohydrate is not a good thing, but these admissions usually come

with a qualifier such as “especially refined sugar” or “particularly refined

starch,” despite the fact that no study has directly compared “refined” and

“unrefined” carbohydrates (high-GI and low-GI are not measures of refine¬

ment and they are, in any case, weak predictors.) The drastic increase in total

carbohydrate that has accompanied the obesity epidemic is its most salient

feature. Some may choose to ignore it if they don’t like it, but it is there.

So, does this mean we can add more fat to our diets, or is fat still bad?

Its probably fair to say that most people think that in some way fat it still

bad, but the role of fat in the body is itself controlled—directly or indi¬

rectly through hormones—by carbohydrate. Deleterious effects of lipid

Nutrition in Crisis

metabolism are ultimately dependent on carbohydrate intake. Perhaps

most surprising, the biochemistry shows that although it is the fatty acids

in your blood that are the problem, they are more likely to come from

dietary carbohydrate than from dietary fat.

Is Carbohydrate Fattening?

I don t understand. I went to this conference and they had a

buffet every night , and I really pigged out on roast beef and

lobster> but I didn't gain any weight

—-Jeffrey Feinman, the authors brother

Nobody manages to avoid weight gain after going on a cruise and pigging

out on pasta, but the same isn’t true for those, like my brother, who indulge

in roast beef and lobster. To return once again to the faucet analogy: Insulin

opens the faucet for fat storage but it shuts down the faucet for fat oxidation.

At this point, you might ask whether these details matter. Doesn’t it all even

out in the end? Isn’t it just calories in, calories out, or, as they always say in the

news releases, “a calorie is a calorie”—and don’t the laws of thermodynamics

tell us that? It is hard to tell the extent to which you can lose more weight,

calorie for calorie, by changing the composition of the diet. One clue, though,

is that when experiments show that one macronutrient is less efficient than

another (wastes calories as heat), it is usually the low-carbohydrate arm that

is less fattening. Critics say that these results are due to inaccurate reporting

of food intake, and that it is always just a matter of total calories consumed.

Low-carbohydrate diets, they claim, simply reduce total energy intake. It’s

true that food frequency records can have substantial error, as in the Brehm

study. If reduced calorie intake is indeed why the low-carbohydrate group

always wins in these face-offs, then as we know, low-carbohydrate partici¬

pants would have to be overreporting what they ate or low-fat comparisons

groups would have to be underreporting what they ate, or both. Wouldn’t

this be a positive point for low-carbohydrate though? There might be a real

benefit to being on a diet where you think you ate more than you did. After

all, there is no “just” about reducing calories.

The other hole in the critics’argument is that thermodynamics does not

predict that “a calorie is a calorie.” Most people who quote the “la>vs of

Basic Nutrition

thermodynamics” (they usually mean just the first law) have never studied

thermodynamics. The essential feature of thermodynamics rests not with

the first law, which is about energy conservation, but rather with the second

law, which says that all (real) processes are inefficient. Energy is dissipated.

The variable efficiency (the extent to which energy is wasted as heat) of fat,

protein, and carbohydrate is well known, but in the medical literature, it is

ignored at will. In cases where total calories turn out to be the controlling

variable, independent of macronutrient composition, it is because of the

homeostatic (stabilizing) mechanisms of biological systems, not because of

thermodynamics. Thermodynamics is my special interest, and we’ll come

back to the subject in chapter 9.

“The Atkins Diet Is a

High-Calorie Starvation Diet”

This quotation is from George Cahill, one of the pioneers in the study

of metabolism and the response to starvation. The idea is that the

reduction in blood glucose and insulin and the increase in glucagon

that accompany lower carbohydrate intake resemble the changes that

are associated with total reduction in calories. In starvation, insulin goes

down, glucagon goes up, fat oxidation increases, and—at some point—

ketone bodies are generated.

In 1992, Klein and Wolfe carried out a defining experiment. 8 The

subjects went without food for three days, were given a period of rest, and

then went through another three days of starvation. During the second

stretch without food, however, the subjects received intravenous injection

of a lipid emulsion (fat) that was designed to meet their resting energy

requirements. The first period represented a model of early starvation, and

the second represented an absence of food intake with adequate energy.

Klein and Wolfe measured several physiologic parameters during each

period, and as shown in table 4.1, there was not a great difference between

the two periods despite the very large discrepancy in energy intake. The

levels of free fatty acids were expected to be different—in the first period,

breakdown of body fat was required for energy, so free fatty acids would

be high; and in the second period, free fatty acids would be lower since

injected lipid would be used for energy—but, in fact, the values were the

Nutrition in Crisis

Table 4.1. Similarity of Starvation and Carbohydrate Restriction

Free

fatty adds Fat oxidation Glucose (l-Hydroxybutyrate

(pmol/I) (pmol/kg/min] (mg/dl] ImM)

84-hour fast .92 1.94 68 2.56

84-hour fast 1.02 1.67 66 2.54

+ lipid

Source S. Klein and R R Wolf, "Carbohydrate restriction regulates the adaptive

response to fasti ng”4meffcan journal of Physiology 262, no 5 [1992]: E631-636

same. Similarly, ketone bodies, which reflect the absence of calories, were

reasonably high in the fasting group, but in the second case, with adequate

energy, one might have expected low ketone bodies. The explanation of

the experiment is that the controlling factor was the level of insulin and

glucose. The study concluded that “these results demonstrate that restric¬

tion of dietary carbohydrate, not the general absence of energy intake itself,

is responsible for initiation of the metabolic response to short-term fast¬

ing.” The statement is undoubtedly something of an exaggeration—there

are other factors that might have modified the results—but the experiment

brings out one of the major themes in diet and metabolism: Carbohydrate

is a controlling element, whereas dietary fat plays a relatively passive role.

Circulating fat, body fat, and fatty acids do play a role in metabolism, but

it is wrong to assume that dietary fat equates to body lipids. This is a major

theme in this book: “You are what you eat” is not a good principle.

Looking Back, Are Carbohydrates Fattening?

Harpers Illustrated Biochemistry is one of the standard texts in medical and

graduate schools. Now in its twenty-ninth edition, it is a multi-authored

comprehensive view of the field. I am grateful to Adele Hite of the

University of North Carolina for pointing out that in the eighth edition

of the text, published in 1961 when it was called Review of Physiological

Chemistry and Harper himself was the sole author, the close connection

between carbohydrate and fat was evident. The chapter on metabolism of

carbohydrate began as follows: “In the average diet carbohydrate compro¬

mises more than half of the total caloric intake. However, only a limited

Basic Nutrition

Going Without Food

Human metabolism has two goals: to provide energy and to

maintain blood glucose at a relatively constant level. In the

eight hours or so after a meal—the fed state, or what nutri¬

tionists call the postprandial state—diet can provide a greater

or lesser amount of the material needed to meet these two

needs. If you go long enough without eating, food no longer

provides material for metabolism directly—this is referred to

as fasting, or the postabsorptive state. When you wake up in

the morning after an overnight fast, insulin is low and glucagon

is high, so fat is broken down through the process of lipolvsis

into fatty acids and glycerol (lipolysis is inhibited by insulin and

stimulated by glucagon.) The fatty acids are oxidized for energy,

and blood glucose is maintained by the processing of liver

glycogen. Muscle also stores glycogen that can be broken down.

This glucose is used by the muscle itself and is not exported.

In simpler terms, as noted before, we can think of muscle as a

consumer of glucose and the liver as a supplier, or more gener¬

ally, as a command center for metabolism.

Gluconeogenesis

Gluconeogenesis is frequently described as a last-ditch

source of energy when food isn’t avail able and glycogen is

depicted, but in fact, gluconeogenesis is happening all the

time. When you wake up in the morning, more than half of

the free glucose in the blood or produced from previously

stored glycogen comes from gluconeogenesis. Although there

is no form of protein formally defined as a storage site, as

there is for stored fat or glycogen, muscle can be thought of

as providing an internal source of protein, a source that must

be replenished from the diet, a dynamic store of amino acids

for metabolism.

Nutrition in Crisis

Is Starvation a Good Way to Lose Weight?

In the fasting state, adipocytes (fat cells) supply fatty acids from

the breakdown of fat. Most tissues, including the heart, oxidize

the fatty acids for energy. Some tissues—primarily the bxain and

central nervous system—require glucose, which can be provided

by tbe combination of glycogen breakdown and replenishment

from glueoneogenesis. The two goals of metabolism are thus

taken care of. Is starvation a good way to lose weight? Of course

it is not. The requirement for amino acids from protein for the

maintenance of blood glucose is the problem. In the absence of

dietary protein, your body will turn to its own sources. Fasting

has become popular, but the breakdown of protein after a day or

two might have physiologic consequences. The answer is not in!

so I would be cautious of a fast for more than twenty-four hours.

amount of this dietary carbohydrate can be stored as such. It is now known

that the un-stored portion of the ingested carbohydrate is converted to fat

by the metabolic processes of lipogenesis.” 9

In other words, the third sentence of the carbohydrate section of a

biochemistry text emphasized the closeness of carbohydrate and fat. In the

twentieth edition (1985), the chapter, now written by Peter Mayes, begins

similarly and continues: “It is possible that in humans the frequency of

taking meals and the extent to which carbohydrates are converted to fat

could have a bearing on disease states such as atherosclerosis, obesity and

diabetes mellitus.” 10

In the current edition, the process of conversion of carbohydrate to fat

now has a chapter of its own. The process is known as de novo lipogenesis,

new synthesis of fat, or more precisely de novo fatty-acid synthesis, since

the immediate product is a fatty acid, the saturated fatty acid palmitic acid

(C16:0). De novo lipogenesis appears to be the explanation of the counter¬

intuitive result, demonstrated in several studies, that dietary carbohydrate

leads to increases in saturated fatty acids in the blood. Chapter 9 describes

Basic Nutrition

an experiment from Jeff Volek, then at the University of Connecticut, where

such an increase in saturated fatty acids was greater in the blood of people

on a high-carbohydrate diet compared to those on a low-carbohydrate diet,

even though the latter had three times the amount of dietary saturated fat.

What About Protein?

In some classes that I teach, I ask the students for the definition of life. I

get different answers but I usually say “No, a one-word definition.” I try not

to drag it out too long or to overact, but the answer that I am looking for

is “protein.” Everything that goes on in life is controlled by protein, either

as the actual component or as the source for other things. Because of its

multiple roles in biology and the far more complicated chemistry, I will

present here only broad outlines in the context of an answer to an email.

I received the following question:

If one is on a very-low-carbohydrate/high-fat diet, what

happens to excess protein that is not needed for muscle repair

and growth, and gluconeogenesis? I see two alternatives:

1. Its excreted

2. More glucose is created.

Number 2 seems so unreasonable to me. Would your metabo¬

lism actually make more than the little bit it needs? Fm open to

a number 3 that I might be too unimaginative to think of Fm

interested in the theory. This isnt a request for diet advice. I like

to understand things at the cellular level.

The answer is that it depends on what else is going on, but protein, per

se, is not excreted (in the absence of some disease). Protein is a polymer.

Unlike glycogen, which is a homopolymer of identical glucose units, the

individual units, amino acids, are picked from about twenty different

choices. In digestion, protein is broken down to individual amino acids,

which are absorbed and reassembled into body protein. The sequence of

amino acids defines the biologic function, and this sequence is encoded

Nutrition in Crisis

in the genetic material. The genetic code is largely the code of amino

acid sequences.

After digestion and absorption, some of the amino acids that are not

used for protein synthesis may be trashed. The nitrogen is converted to

ammonia. Ammonia is converted to the compound urea and excreted. The

remaining carbon skeleton can be used for energy either directly in the

citric acid (TCA) cycle or by conversion to ketone bodies, especially on a

very low-carbohydrate diet. Some amino acids can be converted to glucose.

Much more than a little bit is needed. The carbon skeleton from amino

acids, directly or indirectly, can be converted to fat. So a practical answer is

that “excess” protein is recycled: used for energy or for synthesis of glucose.

Protein, as such, is not normally excreted. Proteinuria is an indica¬

tion of some abnormality, kidney malfunction, or other disease-related

nephropathy. Amino acids are excreted at some low level. High excretion

of particular amino acids is usually an indication of some metabolic distur¬

bance or inborn error of metabolism. It is important to understand that

everything that goes on in the body is mediated by proteins, which turn

over all the time, and whereas muscle “repair and growth” is important, it is

not the only thing. Body proteins, unlike glycogen or other homopolymers,

have specific amino acid sequences and so require a particular makeup.

Some amino acids are interconverted and some (essential amino acids) are

required from diet.

Current tendencies are to try to encourage vegetarianism or, at least,

to encourage reduction of meat consumption. Whatever the moral or

practical arguments are, the scientific case is highly questionable. The

proliferating studies trying to demonstrate an association between meat

consumption and one disease or another are largely bogus, and a couple

of these studies are deconstructed in chapter 14. The past few years have

also seen a proliferation of “experts” whose training and expertise in nutri¬

tion were still being questioned and who have now become authorities on

global warming and sustainability of life itself.

-CHAPTER 5-

An Introduction

to Metabolism

T he nineteenth century was a period of political, intellectual, and

scientific revolution. Although the politics made more noise,

there might be greater long-term impact from the turnarounds

in intellectual and scientific fields. Paris in the summer of 1848 was the

site of yet another French Revolution. People had taken to the streets and

were building barricades just as in Les Mis. There was dissatisfaction over

rights of assembly and the autocratic government, but rising food prices

were perhaps the larger driving force. Whatever the origin of the disor¬

der, faculty at the College de France complained that it had “slackened

the zeal for research among all of the chemists,” and that their time was

“absorbed by politics.” 1 The intellectual revolution at the College, however,

was unfolding in Claude Bernard's laboratory.

Bernard, generally considered the father of modern physiology, had

been studying digestion in dogs. He found sugar in a dog that he had been

dissecting, despite the fact that the animal had not been fed any sugar. The

finding was revolutionary because it was generally assumed at the time that

any sugar in an animal had to have come from the diet. Furthermore, it was

expected that even if an animal had consumed sugar, that sugar would, in

the end, be destroyed by oxidation. Antoine Lavoisier had shown more than

one hundred years before that animals eat sugar—or any food—for energy,

and that the energy comes from oxidation of the food, just as if you were

burning food in a furnace. So how did sugar wind up undigested in the

animal? Bernard s first thought was that there might be something wrong

with the reagent that he had used to detect the sugar, but the reagent was

okay—the dog really was making its own sugar. In fact, he soon found that

if he fed a dog only meat, there was as much sugar in that animal as there

Nutrition in Crisis

was in another dog that had been fed “sugary soup.” Strange as it seemed, he

concluded that the dog must have been making its own sugar.

The Discovery of Glycogen and Gluconeogenesis

Although Bernard's experimental findings were occasionally

at fault and at times influenced by preconceptions , . . his

strength appears to lie in his ability to discard a theory once its

experimental basis had been undermined. Even though he was

apt not to state frankly that he had been wrong, he neverthe¬

less did change his ideas.

—F. G. Young 2

It took some further work to show that the sugar was actually being

produced in the dogs liver, but by 1857, Claude Bernard had isolated

the matiere glycogene —that is, glycogen. We know now that glycogen is

a storage form of carbohydrate, a polymer of glucose. It is actually made

from glucose, although the glucose that goes into glycogen may come from

something else.

As I explained in chapter 4, glycogen is a highly branched, highly struc¬

tured polymer, and it is the key player in maintaining blood glucose at a

constant level. When cells use up the available glucose, the liver breaks

down glycogen to reestablish a constant level. Bernard emphasized this role

of glycogen as a supplier, providing glucose to the circulation, but he was

not quite right about how glycogen formed in the first place. He knew that

glycogen didn't come from fat, and he found through further experimenta¬

tion that dietary protein seemed to raise glycogen levels even more than

dietary sugar (he used fibrin, the blood coagulation protein for his tests).

The picture that evolved was that protein was the source of glycogen, which,

in turn, could produce glucose. This was not exactly right—if carbohydrate

intake is high, ingested sugar is converted to glycogen—but Bernard had

discovered something critical: the need for a process that converted other

things into glucose, a process now known as gluconeogenesis that would

not be fully understood for another hundred years.

Our current understanding focuses on glycogen synthesis and breakdown

as a control point in metabolism. Sugar in the diet or in the circulation can

An Introduction to Metabolism

be stored in glycogen and then made available when needed. However, it

is not only dietary glucose that shows up in glycogen. Bernard was right

that sugar could be made in the liver from protein—that is, amino acids

from proteins.The glucose synthesized in gluconeogenesis can be exported

to the blood, or alternatively, can be used to replenish previously used

glycogen, and the new glucose molecules might only appear in the blood

at a later time.

Although Bernard understood that glucose came from glycogen, he did

not realize that glycogen could also be made from ingested sugar. The key

mistake was Bernard’s inability to grasp that glucose is present in the blood

all the time and it is maintained at a constant level. He didn’t get it because

he had actually made an experimental error. A lucky one, it turns out.

Sometimes We Luck Out

Bernard’s reason for believing that glycogen was not made from glucose

came from his measurements of the inputs to and outputs from the liver.

Bernard determined the amount of sugar in the portal vein, which brings

blood from the digestive tract to the liver (its not a true vein), and he

also measured the amount of glucose leaving the liver in the hepatic veins.

His original observations showed that there was little or no glucose in the

inputs from the portal vein, but that there was sugar in the hepatic veins

leaving the liver. In other words, he could see glucose exiting the liver but

no glucose coming in. This observation pretty much made his case that the

liver was a sugar-producing organ and that glycogen was being made from

something else.

Some of Bernards scientific rivals said that he was wrong, claiming that

they had been able to detect sugar coming into the liver in the portal vein.

They were right, of course—we know that there is sugar throughout the

circulation and that glycogen is assembled from blood glucose. Bernard

had made the mistake of letting some of his preparations sit around too

long while the sugar in the portal vein to the liver was being metabolized.

Bernard was right to an extent: Usually, more sugar is leaving the liver than is

coming in (the fiver does produce glucose), but the differences are small and

the instruments available to Bernard might not have allowed him to detect

them. It was this error that allowed him to piece together a picture of a more

Nutrition in Crisis

complicated part of metabolism. Although he got things slightly wrong and

had to modify his ideas later, the error allowed him to identify glycogen and

to generate the idea of gluconeogenesis. Sometimes we luck out.

Today, we emphasize the need to keep blood glucose constant. We

understand that almost all cells in the body can use glucose as an energy

source, so having too little is not good. However, it turns out that too

much is also not good, because glucose will react with proteins in the blood

and tissues to form what are called advanced glycation end-products, or

AGEs. These modified proteins can be a factor in aging and degenerative

disease. In the popular press and on social media, the negative effects of

high glucose have led to the idea that glucose is a toxin, which is surely

exaggerated, or at least a mischaracterization of the AGEs.

An Evolving Understanding of Metabolism

Through the work of Bernard and others—particularly Louis Pasteur—the

end of the nineteenth century saw the evolution of a science of metabolism,

allowing greater understanding of the inner workings of the body and how

sources of energy are used. By that point, we knew that blood glucose came

from the liver as well as from the diet, and that dietary glucose came from

sugars and starches.

Looking ahead, under conditions where there is no food or there is low

dietary carbohydrate, there will be a continuing drain on body protein.

In the latter case, protein for the body can be supplied from the diet, but

ketone bodies provide an alternative source of energy to reduce the depen¬

dence on protein. Ketone body synthesis and utilization interact, as one

would expect, with glycogen metabolism.

To put this in context, the next section will provide an overview of

energy metabolism, illustrating the principle that, in metabolism, there are

two goals and two fuels.

The Black Box of Life

Metabolism—the conversion of food to energy and cell materials—is as

complicated as you would expect, but it is possible to get an idea of the

big picture. The approach here is called the “black box” strategy: getting

An Introduction to Metabolism

Figure 5.1. The black box approach to metabolism. A, The black box of life summarizes

how we eat food, take in oxygen, and excrete C0 2 and water. B, Inside the black box,

we see the oxidation of food is carried out by an intermediate coenzyme NAD+. The

product reduces oxygen to water. C, The main process. The big players.

as much information as possible by looking at the inputs and outputs

of a system without necessarily knowing the details of what’s going on

inside. As we uncover more details, we can nest black boxes inside each

other, thereby organizing the limited information we have. This method is

favored by engineers, who are the people most unhappy with the idea that

they don’t know anything at all.

You likely already have a basic idea of what we do in metabolism: We take

in food and oxygen, and put out CO 2 and water. Looking at the inputs and

outputs (see figure 5.1), even without knowing a great deal about chemis¬

try, you can figure out that oxidation is occurring within the box—like the

burning of fuel to generate heat or run a machine. Technically speaking,

it is an oxidation-reduction reaction (redox, for short). Oxidation, in this

context, means combination with oxygen. In metabolism, it is frequently

the hydrogen atom attached to a metabolite that is oxidized (to water).

Nutrition in Crisis

Chemical Energy

In physics, energy refers to the ability to do work. As compli¬

cated as systems can get, we are basically talking about lifting

a weight on a pulley In chemistry, energy is identified with the

progress of a reaction. If a chemical reaction proceeds by itself,

at any speed, without the addition of work, we know that we

can use it to lift a weight. If you have studied any chemistry,

you will remember the equilibrium constant, which tells you

how much product you have at the end of the reaction. If the

constant is favorable—that is, if you have a lot of product—then

the reaction is said to be exergonic, downhill and spontaneous.

You can get energy from it.

There are two parts to an oxidation-reduction reaction. In the oxidation of

a hydrogen atom (alone or in a compound), the oxygen is said to be reduced.

The generalization used in biochemistry is that a compound gets oxidized if

it combines with oxygen, and becomes reduced if it combines with hydrogen.

(Redox reactions are a fundamental part of all chemistry and the concepts

have been highly developed, but this simplification works well in biochem¬

istry.) Like combustion reactions, redox reactions produce energy that can

then be used to do work. Some energy is used for mechanical work—moving

muscles—but most goes toward chemical work: making body material,

keeping biological structures intact, generally keeping things running.

The Two Goals

To review, there are two major goals in human metabolism: first, provide

energy for life processes, and second, maintain more or less constant levels

of blood glucose. Too little glucose is not good because it is a major fuel,

but too much is also harmful because it will react with proteins to create

AGEs, as described earlier in this chapter.

An Introduction to Metabolism

Energy in biochemistry is described in terms of a particular chemi¬

cal reaction. When you study biochemistry, you first examine what the

compounds do precisely, but then you use them as abbreviations. Phosphate

ion (unattached to other atoms) is abbreviated Pi, where the “i” stands for

“inorganic.” Although it is now considered somewhat archaic, Pi is still

sometimes read as “inorganic phosphate.” So, the big energy system in

biology is:

ADP + Pi ATP + H 2 0 (1)

Energy storage occurs in living systems through the synthesis of

ATP from ADP (the other reactants are assumed). In metabolism, you

need “energy” from food to make ATP from ADP. Then, when ATP

is converted back to the low-energy form, ADP, through hydrolysis

(adding water)—that is, when the above equation moves from right to

left—energy is released. This energy can be used to do chemical work,

and make proteins, DNA, and other metabolites and cell material. The

reaction is favored in the reverse direction: It tends toward the release

of energy.

Textbooks frequently refer to ATP as a “high-energy molecule,” but it

is the reaction (synthesis and hydrolysis), rather than the compound itself,

that is high energy. For the moment, we can think of ATP as the “coin of

energy exchange in metabolism” and the ATP to ADP ratio as the energy

state of the system.

The Two Fuels

In order to fulfill the two goals, two kinds of fuels are used: glucose itself

and the two-carbon compound acetyl-coenzyme A (abbreviated acetyl-

CoA or acetyl-SCoA). Coenzyme A is a complicated molecule, but its

not important to know the details for our purposes. Coenzymes are small

molecules that take part in the metabolic changes in living systems. They

can be involved in energy metabolism (such as ATP to ADP) or other

reactions. The oxidation-reduction coenzymes are the NAD (nicotinamide

adenine dinucleotide) molecules, of which there are two forms: oxidized,

NAD+, and reduced, NADH.

Nutrition in Crisis

Most ATP in the cell comes from the oxidation of acetyl-CoA, but

glucose can be converted to acetyl-CoA. Acetyl-CoA also comes from

fat, and to a smaller extent, from protein. Glucose itself can be formed

from protein but not from acetyl-CoA. The significance of this, as we

have said, is that fat can be formed from glucose, but with a few minor

exceptions, glucose cannot be formed from fat. Historically, the challenge

for biochemistry has been to explain how the energy from an oxidation-

reduction reaction could be used to carry out the synthesis of ATP, which

has a different mechanism (phosphate transfer). The process is called

oxidative phosphorylation and was only figured out about fifty years ago.

Breaking into the black box, oxidation of food is separated into two

different processes. The food never sees the oxygen but instead there is an

intermediary player. The intermediate agent, the redox coenzyme NAD+,

does the oxidation of food, and the NADH (the product, the reduced

form) is re-oxidized by molecular oxygen. Why do we do it this way? In

general, biochemical reactions proceed in small steps to allow for control

and for capturing energy. Even if we could do it all in one big blast, like

an automobile engine, living tissues do not do well with explosive, high-

temperature reactions—we would have little control over them and we

would not be able to capture the energy in a usable chemical form.

Glycolysis

Glucose is at the center of metabolism. Glycolysis, the collection of early

steps in its processing, is common to almost all organisms. Glycolysis (sugar

splitting) ultimately provides two molecules of the three-carbon compound

known as pyruvic acid (pyruvate). In most cells, the pyruvate from glycolysis

is oxidized to acetyl-CoA, which is the input for aerobic (oxygen-based)

metabolism and the main source of energy for most mammalian cells. One

of the functions of glycolysis is to prepare glucose for oxidative metabolism—

that is, to provide acetyl-CoA. Some cells, however, can run on glycolysis

alone. Such cells are said to have a glycolytic metabolism and can convert

pyruvate to a number of different compounds, most commonly lactic acid.

Many microorganisms are glycolytic, and much of our understanding of

glycolysis comes from the study of bacteria and the process of fermentation.

The final product of glycolysis can be very different from one organism to

An Introduction to Metabolism

another. Alcoholic fermentation involves the conversion of pyruvate to a

two-carbon compound acetaldehyde, which in turn is converted to ethanol.

(When you ingest alcohol, your liver runs this reaction backward, converting

alcohol to acetaldehyde and then to acetyl-CoA, which is further oxidized.)

Other kinds of glycolytic bacteria, like those in yogurt, convert pyruvate to

lactic acid (lactate), accounting for the acidity of yogurt Mammalian cells can

also carry out this transformation. The brain, central nervous system, red blood

cells, and rapidly exercising muscle are the most common of the glycolytic

tissues that produce lactate. It was once believed that the lactate produced by

exercising muscle was the cause of delayed-onset muscle soreness (DOMS),

but this is not true. Although the cause of DOMS is not known, the lactic

acid is metabolized and long gone by the time soreness sets in.

Oxidative Metabolism

Acetyl-CoA is the main substrate for the oxidative process that produces

CO 2 , which is then released from the black box of life. This process is

frequently called the Krebs cycle, after Sir Hans Krebs, who was the

pioneer in assembling a coherent mechanism from the various observa¬

tions of where particular carbon atoms went when different foods were fed

to a tissue or organism. Oxidation of such a small molecule as acetyl-CoA

would have to take place in a small number of steps, and would not allow

the kind of control that it is necessary to keep a biological system respon¬

sive to different conditions, so the two carbons of acetyl-CoA are attached

to a carrier to form a six-carbon molecule, called citric acid, or citrate.

For this reason the cycle is frequendy referred to as the citric acid cycle.

Complicating things further, citric acid is a tricarboxylic acid, or TCA for

short, so the process might also be called the TCA cycle. All three names

are used: the Krebs cycle, the citric acid cycle, and the TCA cycle. Krebs

himself called it the TCA cycle, so we will try to stick with that.

The TCA cycle is complicated, but I will try to provide a rough descrip¬

tion: The substrate, acetyl-CoA, is bound to a carrier to form a compound,

citric acid, that is oxidized, stepwise, primarily by the redox coenzyme

NAD+.The products of the reaction are CO 2 and the reduced form of the

coenzyme, NADH. NADH is the ultimate reducing agent (transfers H)

that turns oxygen to water. This process, the electron transport chain, is a

100

Nutrition in Crisis

Sources of Acetyl-CoA

Glycolysis is not the only source of acetyl-C oA for the oxida¬

tive metabolism of glucose. The major source, for many cells,

is fatty acids from fat. The process of converting fatty acids is

called (3-oxidation. The long-chain fatty' acids are chopped two

at a time from the carboxyl end. It uses the Greek character (3

because the break is at the second carbon in the fatty acid chain.

sequence of reactions that, in effect, transfers electrons. The net effect is

the reduction of molecular oxygen to water. Somehow this converts ADP

to ATP. This is the mechanism for channeling the energy of burning food

into the conversion of ATP back to ADP—that is, the storing of chemical

energy. The process, still mysterious when I was in graduate school, is now

understood. The idea behind it is called the chemiosmotic theory—really

no longer a theory—which was largely the work of a single man, Peter

Mitchell. It would be a major digression to delve into the overall process

here, but I recommend doing some research into both chemiosmotic

theory and Peter Mitchell himself. Because it involves oxygen, obtaining

energy from the TCA cycle electron transport chain is referred to as oxida¬

tive metabolism, distinct from glycolytic metabolism, which might precede

it. Glycolysis provides acetyl-CoA for the TCA cycle. Looking ahead, a

major feature of cancer cells, referred to as the Warburg effect, appears

to be a greater reliance on glycolysis rather than oxidative metabolism, in

comparison to normal cells, even when oxygen is present. Explaining the

Warburg effect is a major focus of current cancer research.

Ketone Bodies

Ketone bodies have evolved as an alternative energy source under condi¬

tions of starvation or carbohydrate restriction. Synthesis and utilization

of ketone bodies provide a way of dealing with the biochemical principle

An Introduction to Metabolism

101

stated so many times throughout this book: You can make fat from glucose

but you cannot make glucose from fat. None of the energy stored in the

form of fat will help in providing glucose for the brain and the central

nervous system (CNS). In the absence of dietary carbohydrate or total

calories, protein becomes the main source of carbons for gluconeogen-

esis—and the risk here is that you will break down essential body protein

to meet this second goal of metabolism, maintaining blood glucose.

Created in the liver from acetyl-CoA, the ketone bodies,

p-hydroxybutyrate and acetoacetate, are four-carbon compounds. They

provide a way for acetyl-CoA units (from fat) to be transported from the

liver to other tissues, where they are turned back into acetyl-CoA and used

for energy. Through this process, protein no longer has to bear the full

pressure of providing glucose under extreme conditions.

The brain and CNS cannot use fatty acids for fuel. These and some

other tissues are dependent on glucose, at least under normal well-fed

conditions. The dilemma in a starvation state, or on a low-carbohydrate

diet, is how to supply energy to these tissues. Because you cannot make

glucose from fat, amino acids from protein must supply glucose via gluco-

neogenesis. In starvation, that protein must come from muscle and other

body proteins. It is the demand for glucose, rather than total energy, that

is the problem under extreme conditions. (On a low-carbohydrate diet, the

problem is avoided because of the availability of dietary protein. While

low-carbohydrate diets are not necessarily high in protein, a somewhat

higher level is needed to supply material for gluconeogenesis.)

The ketone bodies are derived from fatty acids, but unlike fatty acids them¬

selves, they can be used by the brain and CNS because they supply acetyl-CoA

directly. The evolutionary advantage of ketone bodies is that they provide a

way to avoid the breakdown of body muscle stores. For this reason, ketosis

(ketone bodies in the blood) is generally described as “protein-sparing/Tn the

beginning stages of ketosis, muscle tends to get most of the ketone bodies for

fuel, but as things proceed, more ketone bodies are diverted to the brain. The

ketone bodies can reduce the need for glucose by more than half.

A low-carbohydrate diet will supply the protein for glucose synthesis, but

the adaptive mechanisms that have evolved to spare protein in starvation

will remain operative—so the state of the body with reduced carbohydrate

intake is not unlike that in starvation. This is why George Cahill, who did

102

Nutrition in Crisis

the pioneering work in ketone bodies and starvation, described the Atkins

diet as a “high-calorie starvation diet.”

There is no requirement for dietary glucose, but what is the requirement

for glucose in the body as a fuel? The number that you see in the literature is

130 grams per day, and it’s a number with a strange history. In George Cahills

classic study on the response to starvation, 3 it was found that this was the

amount of glucose consumed by the brain under normal conditions—that

is, before the starvation phase of the experiment began. After several days

of starvation, however, the amount was found to be substantially less, in the

range of 50 grams per day (at this point, the glucose was obviously not coming

from diet, since it was measured under starvation conditions). Somehow,

nutritionists picked up on the baseline 130 grams per day figure and even

morphed this into a dietary requirement. Cahill told my colleague Eugene

Fine that by the time he realized this had happened it was too late to stop it.

The mistake has since spread throughout the literature, compounded by the

suggestion that not only do you need 130 grams of glucose (not always true),

but that you need to get that 130 grams specifically from diet (never tme).

In summary, the brain and CNS require about 50 grams per day of

glucose, and under conditions of low glucose, other fuels—namely, ketone

bodies—might become more important. Many questions remain unan¬

swered, however. Who is directing all this? How does the fat cell know when

to provide fatty acids to the liver? How does the liver know when to make

ketone bodies, and how many? What controls whether glucose is burned

for energy or stored as fat? And thus we see the real problem in the study

of metabolism. A metabolic map, like any map, only tells you about possible

routes; it doesn’t tell you where the traffic lights are, and it might or might

not tell you where the traffic cops are. What we do know is that the most

important of the traffic regulators, not surprisingly at this point, is insulin.

The Role of Insulin

Insulin is an anabolic, or building up, hormone. Of the numerous regula¬

tors of metabolism, insulin is the most important, targeting a number of

organs and processes. One of its primary effects is blocking the breakdown

of fat to fatty acids. Indirectly, insulin inhibits breakdown of fat and favors

fat storage, as well as glucose storage in the form of glycogen. Uptake into

An Introduction to Metabolism

103

muscle is also stimulated by recruitment of the glucose transporters, the

GLUT4 receptors. GLUT4 is described as an “insulin-dependent recep¬

tor,” and clearance through this receptor was once considered to be the

main effect of glucose, though it is now thought to be secondary. Overall,

insulin clears glucose and stores it, catalyzes fat storage, and builds new

proteins. Physiologically, it is a sign of good times.

Looking ahead, in the case of diabetes, where there may be an absence of

insulin (type 1) or poor response to circulating insulin (type 2), the effects

on lipid metabolism might be more important than the effects on carbohy¬

drate metabolism, even though the primary problem rests with inadequate

response to ingested glucose (contained in sugar or starch). The most imme¬

diate and salient feature of diabetes, however, is the hyperglycemia (high

blood glucose) due to a failure to prevent hepatic production from the break¬

down of glycogen and the synthesis of new glucose via gluconeogenesis. The

addition of dietary glucose on top of this will obviously make things worse.

Ketone Bodies, Ketoacidosis, and Insulin

Ketone bodies are acids, and as such, circulation must be protected against

acidosis in their presence. Ketosis can become very high in type 1 diabetes,

and ketoacidosis is one of the major threats of untreated type 1. So how is

ketoacidosis prevented in people without diabetes and why does it occur in

those with the disease? The idea is that, like most processes in metabolism,

ketone-body synthesis and utilization is controlled by feedback loops.

Here’s the process:

1. When glucose is low, insulin is low and glucagon is high (in cases of

starvation or low-carbohydrate diet). This leads to disinhibition of

lipolysis in the fat cells, the process of fat breakdown that is normally

repressed by insulin. The net effect is that fatty acid is increased and

exported in the blood. Fatty acids in the blood, again, are referred to as

free fatty acids (FFA) or nonesterified fatty acids (NEFA).

2. The fatty acids are oxidized in the liver. Fatty acids are degraded by

P-oxidation, which chops off two-carbon acetyl-CoA molecules stepwise.

3. Acetyl-Co A is the substrate for energy metabolism in liver and other

cells, but...

104

Nutrition in Crisis

4. As acetyl-CoA increases, it is converted to the four-carbon ketone

bodies (3-hydroxybutyrate and acetoacetate, and is transported to other

tissues where it regenerates acetyl-Co A. As before, ketone bodies are a

way of transporting acetyl-Co A.

5. Ketone bodies regulate their own production in several ways. One way

is to reduce the FFA that goes to the liver.

6. There are two major feedback inhibitors: Back at the adipocyte (fat

cell), the level of ketonemia (ketone bodies in the blood) is sensed by

receptors. When the concentration is high, the ketone bodies turn off

their own synthesis—that is, they inhibit lipolysis. In addition to this

direct effect, ketone bodies also stimulate secretion of insulin from the

pancreas. This, in turn, inhibits lipolysis. So you re left with lower fatty

acids, lower acetyl-CoA in the liver, and lower ketone bodies.

7. If glucose is still low, lipolysis is increased, and fatty acids “go back up.”

8. Which process happens first? Well, it all happens at once. The whole

system might not move because it is locked into the steady state of

interlocking effects of glucose, insulin, ketone bodies, fatty acids, and

everything else. The level of insulin biases the whole system in one

direction or the other, but everything controls the final state.

A person with type 1 diabetes, lacking insulin, cannot exert feedback

control and is therefore in danger of ketoacidosis—since the ketone bodies

are acids, high levels will increase acidity.

An electronic amplifier provides a good analogy. The input from your

sound system can be amplified greatly, in which case the amplifier is said

to have very-high open-loop gain. Such a large signal, however, will also

amplify the distortion: Small changes will cause large noise in the output.

The distortion is dealt with by feeding some of the output back into the

amplifier in the opposite direction of the incoming signal—so-called

negative feedback. The gain (amplification) is thus greatly reduced, but the

fidelity is increased. The process is similar in metabolism, where you don’t

want big fluctuations. In type 1 diabetes, however, the fatty-acid signal

cannot be adequately controlled. Its analogous to hooking your system up

to an electric guitar whose signal cant be controlled and is said to drive

your amplifier into saturation. You are left with an increase in distortion.

-CHAPTER 6-

Sugar, Fructose, and

Fructophobia

W hatever your degree of interest in nutrition, you are likely

familiar with the media s crusade against sugar. There are now

numerous scientific articles, picked up by the popular media,

with titles like “Consuming Fructose-Sweetened, Not Glucose-Sweetened,

Beverages Increases Visceral Adiposity and Lipids and Decreases Insulin

Sensitivity in Overweight/Obese Humans.” 1 The sudden and pervasive

spread of high-fructose corn syrup (HFCS) has made fructose a particular

object of fear and loathing, despite the fact that HFCS has about the same

proportions of fructose and glucose (55:45) as table sugar (50:50). The

strange bedfellows in this political movement include Michael Bloomberg,

the former mayor of New York, and Robert Lustig, a pediatric endocri¬

nologist. Exaggeration and alternative facts are the standard: Bloomberg

claimed that banning large bottles of soda would solve the problem of

obesity, and Lustig told us that sugar was as bad as alcohol or cocaine. How

do we make sense of all this?

We have always known that if you sit around all day eating candy, you

will get fat. Conversely, cutting down on sugar, which is a carbohydrate,

will contribute to weight loss and other benefits of a low-carbohydrate

diet. The extent to which sugar, that is, sucrose, or its component fructose

(sucrose is half glucose and half fructose), has a unique role in obesity and

other effects of carbohydrate is not well understood. However the observed

differences are not large, and as discussed below, on a low-carbohydrate

diet, any fructose that is ingested will be converted to glucose.

The problem with the numerous scientific papers claiming that fructose

is dangerous is that they are all carried out against a background of 55

percent total carbohydrate diets. If you are going to remove carbohydrates

106

Nutrition in Crisis

from your diet, you might want to know whether fructose or glucose

should be eliminated—that is, whether sugar is better, worse, or the same

as starch. What these studies show you, instead, is how bad it is to add

fructose on top of existing nutrients, or to replace glucose with fructose.

It is not reasonable to think that we will solve things by keeping a high-

carbohydrate diet and switching from orange juice to the whole-grain

bread that always seems slightly indigestible (at least to some of us).

The problem is that we dont really have an answer when it comes to

fructose. For people with diabetes or metabolic syndrome, or even those

who are overweight, the data are clear: Total carbohydrate restriction any

carbohydrate—is the most effective therapy. Yet we don't know the extent to

which removing fructose is a player in these therapeutic effects because of

the insulin effect, if you have diabetes, it is almost always better to remove

starch than to remove sugar. The sugars are interconvertible and effects differ

depending on conditions. You dont need credentials to do science, but what

professionals do know is that you cant simply make guesses. Because of the

complexities—depending on conditions more than half of ingested fruc¬

tose is turned to glucose—you cant extrapolate from isolated experiments.

Sometimes, you need the data. Biochemists are rare in nutrition because we

simply dont know enough. I continue to admit that I am just about the only

biochemist dumb enough to get involved here.

In Defense of Sugar (and Fructose)

Writing a “defense” of sugar seems very odd. In some ways, it’s in the same

vein as the defenses of saturated fat that I have written in the past. In those

cases, I stressed the need for perspective. People have been making and

eating cheese for as long as they knew how. What am I defending? Eggs

Benedict? Bearnaise sauce? Soppressata sausage? These have been part of

our culture for generations. Were we really supposed to believe that some

recent earth-shattering scientific discovery meant we’d been consuming

poison all along? It didn’t seem to make sense, and in fact, the science was

very poor, sometimes embarrassingly poor. It’s the same case with fructose.

Sugar is a food, and while nobody denies that many in the population are

plagued by overconsumption due to excessive availability, it is still a feature

of la cuisine, haute and otherwise. Chocolate mousse is as much a part of

Sugar, Fructose, and Fructophobia

107

our culture as Wagner. Some of us have to limit mousse to very small doses,

but for many, that is also true of Wagner.

The Threat of Fructophobia

Rob Lustig is a nice guy. Everybody says this before criticizing his increas¬

ingly unrestrained crusade against sugar. He is almost as ubiquitous as

HFCS itself. His YouTube and 60 Minutes performances, among others,

have raised him to the standard of scientific spokesperson against fructose.

One of Lustig’s YouTube videos begins with the question, “What do the

Atkins diet and Japanese food have in common?” The answer is supposed

to be the absence of sugar but, of course, that isn’t true.

Sugar is an easy target. These days, if you say “sugar,” people think of oft-

vilified foods such as Pop-Tarts or Twinkies, rather than red wine poached

pears or tamagoyaki, the sweet omelet that is a staple in Bento Boxes.

Here’s a question, though: If you look on the ingredients list for Pop-Tarts,

what is the first ingredient, the one in largest amount? It s not sugar; it s

enriched flour. When I posted this on my blog, one person claimed that

although flour is the first ingredient, if you add up the high-fructose corn

syrup, dextrose, and other types of sugar, their sum is larger. Some have

indeed suggested that this is a strategy to hide total sugar content. It’s an

interesting idea, but easily disproven: The label clearly says that there are

38 grams of total carbohydrate and only 17 grams of sugar.

Whether or not they think that carbohydrates are inherently fattening,

most people agree that by focusing on fat, the nutritional establishment

gave people license to overconsume carbohydrates, thus contributing to

the obesity epidemic. Now, by focusing specifically on fructose, the AHA,

USDA, and other organizations are giving implicit license to overconsume

starch—its almost guaranteed since these agencies are still down on fat

and protein. The additional threat is that in an environment of fructopho¬

bia, the only studies on fructose that will be funded are those with subjects

consuming high levels of carbohydrate. In such studies the two sugars,

glucose and fructose, interact metabolically and sometimes synergistically,

so deleterious effects of fructose are likely to be found. The results will

likely be generalized to all conditions, and as with lipophobia, there will be

no null hypothesis.

108

Nutrition in Crisis

The barrier to introducing some common sense into the discussion is,

again, that we really don't have the answers. We don’t have enough data.The

idea that fructose is a unique agent in increasing triglycerides (fat in the blood)

is greatly exaggerated. High-carbohydrate diets lead to high triglycerides,

and there are indeed conditions where sugar has a worse effect than starch,

but the differences between the effects of fructose specifically and those of

carbohydrates in general are small. The greatest threat of fructophobia is that

we wont find out what the real effect of fructose is. Furthermore, the two

regimes being compared in current studies are both high in carbohydrate,

and because the absolute level of the triglycerides is high in both cases, it

is likely that the biggest difference would be seen if either of the outcomes

were compared to the results of a low-carbohydrate diet.

The single most important question to ask with regard to sugar is this:

For a person on the so-called standard, high-carbohydrate American diet,

is it better to replace carbohydrate (any carbohydrate) with fat (any fat

except trans-fat), or to replace fructose with glucose? Ignoring subtleties,

to a first approximation, which is better?

We might not know the answer to that question in every case, but the

studies that have been done clearly indicate that replacing carbohydrate

with fat is more beneficial. We also know that ethanol is not processed like

fructose despite what many claim on the internet. Although the pathways

converge, like almost all substrates that are used for energy at the entry to

the TCA cycle, they are different. It's just not true that there is a similar¬

ity. We also know that although some claim glycogen is not formed from

fructose, and Lustig shows a metabolic pathway from which glycogen is

absent, fructose does actually give rise to glycogen. Under most condi¬

tions, fructose is a preferred substrate for glycogen synthesis. (In fact, for

some period in the history of chemistry, fructose was primarily considered

a “glycogenic substrate.”) Of course, fructose must first be converted to

glucose. Claude Bernard knew that fructose could give rise to glycogen,

but he couldn't understand why he wasn't able to find any fructose within

the glycogen itself.

To return to the comparison that is often made between sugar and

ethanol, it is possible that sugar and ethanol have behavioral effects in

common, but this is not due to similarities in metabolism. Moreover, the

behavioral effects are not even settled within the psychology community:

Sugar, Fructose, and Fructophobia

109

Alcoholism is far different from sugar addiction, if there is such a thing.

While there is no definite agreement, addictive has formal definitions in

behavioral psychology, and polishing off the whole bag of chocolate chip

cookies might not technically qualify as addictive behavior. Some restraint

is necessary. Alcohol-associated liver disease is a well-characterized, life-

threatening condition, and many people do die from it. The idea that

fructose can have the same effect, either physically or psychologically, has

no basis in science and is deeply offensive to people who have had personal

experience with alcohol-related illness and death.

In the end, nobody has ever been admitted to a hospital for an overdose

of fructose. People might say that your diabetes was caused by fructose

consumption, but you cant be admitted to a hospital on somebody's opinion

of what you did wrong. There is no diagnosis called “fructose poisoning.” If

you are admitted to the hospital for type 2 diabetes, that is the diagnosis.

The Threat of Policy

The numerous scientific papers that find something wrong with fructose

may have had little impact on people s behavior, but possibly for that

reason, the fructophobes have taken new steps. Convinced of the correct¬

ness of their position, they have taken their case to politicians who are

always eager to tax and regulate. There is an obvious sense of deja vu as

another group of experts tries to use the American population as guinea

pigs for a massive population experiment, along the lines of the low-fat

fiasco under which we still suffer (not to mention the historical example of

alcohol prohibition). It is not just that the lipophobe movement had unin¬

tended consequences (think margarine and trans-fats) but rather that, as

numerous people have pointed out, the science was never there for low-fat

to begin with. In other words, before science is turned into policy, we have

to address the question of whether the science is any good to begin with.

Fructose in Perspective

Contrary to popular myth, the “Twinkie Defense” of Harvey Milk's

murderer Dan White did not argue that the defendant was possessed by

some kind of sugar rush. Rather, the defense claimed that he was propelled

Nutrition in Crisis

to homicide by his depressed state, and that this deranged mental condi¬

tion was indicated by his consumption of junk food. (He had previously

been a health nut.) It was a strange defense, because most of us think of

depressives as going for suicide rather than homicide but in any case,

sugar did not make him do it.

Fructose is a normal nutrient and metabolite. It is a carbohydrate and

it is metabolized in the metabolic pathways of carbohydrate processing. If

nothing else, your body makes a certain amount of fructose. Fructose, not

music (the food of love according to Shakespeare), is the preferred fuel of

sperm cells. Fructose formed in the eye can be a risk but its cause is gener¬

ally very high glucose. The polyol reaction involves sequential conversion

of glucose to sorbitol and then to fructose.

One truly bizarre twist in the campaign against fructose was the study

by Miguel Lanaspa and his colleagues 2 that attempted to show that the

deleterious effects of glucose were due to its conversion to fructose via the

polyol reaction. The paper was technical, but many people reading it in

Nature don’t think that the conclusion really followed from the data—it

would be a major change in metabolic thinking, and if it were true* it

would have been accepted. If the idea has not actually been excluded, then

it must be some expression of the need to say that, somehow or another,

everything bad is caused by fructose.

My review with Eugene Fine, ^Fructose in Perspective, was published

in 2013 in Nutrition and Metabolism with the following conclusion:

We all agree that reducing sugar intake as a way of reducing

calories or limiting carbohydrates, is a good thing, especially

for children, but it is important to remember that fructose

and sucrose are carbohydrates. We know well the benefits of

reducing total carbohydrate. How much of such benefits can be

attributed to removing sugar is unknown.The major point is that

these sugars are rapidly incorporated into normal carbohydrate

metabolism. In some sense, any unique effects of sucrose that are

observed are due to fructose acting as a kind of super-glucose. 3

Again, the threat is thinking that fructose is sufficiently different from

glucose that substituting glucose for fructose is guaranteed to be better. It’s

Sugar, Fructose, and Fructophobia

111

The Evolutionary Argument

It is still frequently said that our metabolism is not designed to

handle the high input of fructose that was absent in our evolu¬

tion and that is present in our current environment. Although

fructose might not have been widely available in Paleolithic

times, that doesn’t mean that high consumption was uncom¬

mon. It is likely that, for our ancestors, finding the rare berry

bush was like finding a coupon for Haagen-Dazs. Moderation

was not the key word. While few would argue that wide avail¬

ability or high consumption is without risk, it is important to

hew to the science. There is a big diffeience between saying that

continued ingestion of high sugar (or high carbohydrates, or

high anything) is not good, and saying that we do not have the

metabolic machinery to deal with high intake, or that it is a

foreign, toxic substance.

not. Our review was a technical analysis of the literature, summarized in a

rather complicated figure showing what is known of the two sugars. That

figure might not be easily accessible to everybody, but figure 6.1, a simplified

version of the one in the study, explains some of the key points. Fructose

and glucose follow separate paths initially, but both six-carbon sugars are

broken down to three-carbon fragments, the triose phosphates. These triose

ph osphates are the intermediates in the effects of high carbohydrate inges¬

tion: high plasma triglycerides and low HDL (“good cholesterol”).

The question for anybody who wants to attribute special properties to

fructose is how an atom in a triose phosphate knows whether it came from

fructose or from glucose. Of course, it doesn’t, and the outstanding feature

of fructose metabolism is that it is part of glucose metabolism. Figure 6.1

shows the path by which fructose can be turned into glucose, and under the

appropriate conditions, into glycogen. As much as 60 percent of ingested

fructose can be converted to glucose via gluconeogenesis. So, is there

112

Nutrition in Crisis

// \

GLYCER0L-3-P PYRUVATE * T LACTATE

TAG ETC TCA CYCLE

Figure 6.1. A model of hepatic fructose metabolism. Key notes include: (1] Fructose

stimulates glycogen storage; (2) fructose and glucose share common intermediates;

(3) fructose can be converted to glucose; [4] ingested fructose calls for more glucose;

and [5] generally, the liver expects glucose and fructose to come in together.

really no difference between fructose and glucose? It turns out that the

differences are small, and those that exist are due to kinetic (rate) effects

rather than overall pattern of processing. Fructose is rapidly processed.

Technically, fructokinase, the first enzyme to process fructose, has a low

Km, which refers to the fact that it doesn’t take much fructose to get its

metabolism going at a high rate. The carbons from fructose appear in the

triose phosphates very quickly. So, if one wanted to make a grand state¬

ment, it might be that fructose is, as we called it, a kind of super-glucose.

This is not an academic conclusion, however. We still need to know

whether sugar acts primarily as a carbohydrate. Again, if we remove sugar

and replace it with starch, how will the results compare with the many trials

that show benefits of removing carbohydrate across the board and replac¬

ing it with fat? Nobody has directly made the comparison. The experiment

has never been done. Experiments with real low-carbohydrate diets give

far more dramatic results than those that replace fructose with glucose.

Sugar, Fructose, and Fructophobia

113

One important phenomenon to note: The amount of glucose in the

human diet is always greater than the amount of fructose (see figure 6.2).

It is unlikely that anybody eats only sugar and no starch (which, again, is

all glucose), and even HFCS has only slightly more fructose than glucose.

Pure glucose is used for various effects in commercial food preparation.

You will therefore almost never see pure fructose in the absence of glucose

in the human diet. Moreover, ingested fructose triggers an increase in the

glucokinase, the first glucose-processing enzyme in the liver, leading to

further uptake of glucose. In other words, fructose calls for glucose—the

liver expects the two sugars together. This is important, in that adminis¬

tering fructose alone is clearly detrimental. When it was discovered that

fructose does not stimulate insulin secretion, some thought that fructose

might be a good replacement for glucose for people with diabetes—again

reflecting the idea that everybody needs sugar to be happy. It turned out

that consuming fructose without glucose is dangerous. It is important to

recognize, however, that the effects of fructose alone are not comparable to

those of fructose in combination with glucose.

In demonizing fructose across the board, we have to ask whether we

are making the same rush to judgment that we did with fat. We said

that dietary fat would give you heart disease, and a cascade of changes in

medical practice flowed from that, but it has never been shown that fat

gives you heart disease. A whole human population went all out to replace

fat with carbohydrate, and this led to an increase in obesity and diabetes.

Carbohydrate restriction remains the best strategy for obesity, diabetes,

and metabolic syndrome. The specific contribution of removal of fructose

or sucrose to this effect remains unknown. Unknown is the key word.

Sweetener Consumption

What about sweetener consumption? Surprisingly, it hasn’t gone up as

much as you might have thought—about 15 percent, according to the

USDA. One question is whether this increase is disproportionately due

to fructose. The data show that, in fact, the ratio of fructose to glucose

has remained constant over the last forty years. While sugar or HFCS is

the main sweetener, pure glucose is sometimes used in the food industry

and has remained at a constant 20 percent, explaining the deviation from

Per Capita Availability (Ibs/person/year) Pounds per Year

114

Nutrition in Crisis

100

Total HFCS and Sugar

1970 1975 1980 1985 1990 1995 2000 2005

Year

140

q-u

1970 1975 1980 1985 1990 1995 2000 2005

Year

Figure 6.2. Sweetener consumption data, 1970-2005. Data from the USDA Sweetener

Yearbook tables and the USDA Food Availability Data System.

1:1, which would be expected (see figure 6.2). There is, however, more

glucose than fructose in the food supply. One might argue that despite

the constant ratio, the absolute increase in fructose has a more perni¬

cious effect than the increased glucose—but, of course, you would have

Sugar, Fructose, and Fructophobia

115

to prove that. Figures 6.1 and 6.2 suggest that you would have to be

careful in determining whether the effect of increased sweetener is due

to fructose or to glucose, or whether it is the effect of one on the other,

or the effect of insulin and other hormones on both.

Scapegoats

For a good example of the hyperbole surrounding discussion of fructose,

consider an article in Mother Jones by Gary Taubes and Cristin Kearns

Couzens called “Big Sugar’s Sweet Little Lies.” 4 The article is a fascinating,

well-researched story of how the sugar industry has pushed its products,

but the connection the authors make to the tobacco industry’s attempts

to bury information and maintain profits is an overreach. I suspect one

does not have to look hard for industries that try to sell their product

by underhanded means, but the ethical problems in the case of tobacco

were related to the product, not its promotion. Sugar does not functionally

resemble tobacco in any way. The goal, of course, is to tax or otherwise

impose punishment for sugar consumption—the first reaction of politi¬

cians is to punish and tax. The decline in cigarette smoking is cited as the

effect of taxation. The question of whether reduction in cigarette consump¬

tion was a response to financial pressure or education is testable, however:

If the former case is true, reduction in smoking should be greater in

lower-income economic groups, and it should be opposite in the latter case

(sources such as the Gallup polls show the latter case is true). The Mother

Jones’s piece described a large rise in consumption of sugar sweeteners that

was accompanied, in turn, by “a surge in the chronic diseases increasingly

linked to sugar.” You can’t really expect a popular article to stand up to

academic analysis, but the sentence is tricky—really illogical. It says that

“the increase” in sugar is linked to diseases “linked to sugar.’’You can’t use

“linked” twice. The link of chronic disease to sugar is exactly what’s in

question. The graphics in the article tried to bring out the statistics, but

the presentation is misleading. Mother Jones is not Annalen der Physik, of

course, but some degree of precision is still required. In fact, the increase in

the past thirty years is quite small. Using the numbers in the article:

% Increase in sugar = 12 / 120 = 10%

116

Nutrition in Crisis

Other data show that the increase was 15 percent, so we can even

go with that number. For comparison, in a comparable time period, the

increase in the United States for total carbohydrate was 23.4 percent for

men and 38.4 percent for women. The idea, however, is that sugar has a

very powerful, almost catalytic effect—in other words, a little increase in

sugar is supposed to bring about big changes in health. So what happened

in the thirty years that is presumed to have been a consequence of the 9

percent increase in added sugars? Again, from the figure in the paper:

% Increase in diabetes

% Increase US children who are obese

% Increase US adults who are obese

4.3

2.5 =

172%

11.4

5.5 =

207.2%

20.7

15 =

138%

Is fructose that powerful? If it is, the recommended reduction would

certainly be a good thing, but how likely is it that fructose was truly the

root of these increases, and if it is that powerful, could it ever be adequately

reduced? A 9 percent or even 15 percent increase in sugar consumption is

supposed to have caused a major increase in obesity and diabetes; the abso¬

lute changes are small, but still significant since these statistics represent

the entire population. The experiment to test whether fructose is actually

that powerful would be easy to conduct: Compare removal of fructose with

removal of glucose and the results should be evident. Again, that experi¬

ment hasn’t been done. Why not? We have good experiments showing that

if you take out carbohydrates and put in fat, you get significant benefit. If

fructose is the most harmful of the carbohydrates, it should not be hard to

show that it was fructose that was the controlling player in these compari¬

sons. It is surprising, and perhaps even suspicious, that the experiment

hasn’t been done.

-CHAPTER 7-

Saturated Fat

On Your Plate or in Your Blood?

A high-fat diet in the presence of carbohydrate is different than

a high-fat diet in the presence of low-carbohydrate. Failure to

understand this principle, and the resulting failure to adequately

test the effect exerted by carbohydrate, stands as a major source of confu¬

sion. Numerous reports in the medical and popular literature describe the

effect of a “high-fat diet” or even “a single high-fat meal.”The source of the

high fat, however, might be a slice of carrot cake, a Big Mac, or something

else that is also very high in carbohydrate.

On the subject of saturated fat, the studies from Jeff Volek’s laboratory,

then at the University of Connecticut, provide the most telling evidence.

Science does not run on majority rule. The total number of experiments

is less important than the scientific design of the individual trials and

whether it is easy to interpret their results. A study from Volek’s lab on

forty volunteers with metabolic syndrome provides a classic case, carefully

controlled and unambiguous.

Particularly striking, the study found that when the blood of volun¬

teers was assayed for saturated fatty acids, those who had been on a

low-carbohydrate diet had lower levels than those on an isocaloric

low-fat diet—this despite the fact that the low-carbohydrate diet had

three times the amount of dietary saturated fat as the low-fat diet. How

was this possible? Well, that’s what metabolism does. For those on the

low-carbohydrate diet, the saturated fat was oxidized, while the low-

fat group was making new saturated fatty acid. Volek’s former student,

Cassandra Forsythe, extended the idea by showing how, even under

eucaloric conditions (no weight loss), dietary fat has relatively little

impact on fat in the blood. 1

118

Nutrition in Crisis

A barrier to understanding the role of saturated fat rests with the

emphasis on “diets,” where it is impossible to come to an agreement on

definitions and where an accidental or individual response might happen

to work for an individual or small group of people (e.g., the generic ad hoc

grapefruit diet). We would do better by speaking, instead, of basic prin¬

ciples. The key principle is that dietary carbohydrate, directly or indirectly,

through insulin and other hormones, controls what happens to ingested

and stored fatty acids. Carbohydrate has a catalytic effect—it controls what

happens to other nutrients. The fat in the Big Mac will not constitute any

risk if you chuck the bun. You are not what you eat. You are what you do

with what you eat.

The question of saturated fat is critical. The scientific evidence shows

that dietary saturated fat, in general, has no effect on cardiovascular disease,

obesity, or probably anything else—but plasma saturated fatty acids do. In

particular, plasma saturated fatty acids can provide a cellular signal, and they

exacerbate insulin resistance. If you study dietary saturated fatty acids under

conditions where carbohydrate is high, or, more importantly, if your study

effects are in rodents, where plasma fat better correlates with dietary fat,

then you will confuse plasma fat with dietary fat. There is a real difference.

To understand the problem, recall that strictly speaking, there are only

saturated fatty acids (SFAs). What is called “saturated fat” simply means

those fats that have a high percentage of SFAs. Things that we identify as

“saturated fats,” such as butter, usually contain only 50 percent saturated

fatty acids. Coconut oil is probably the only fat that is almost entirely

saturated fatty acids, but because those acids are medium-chain length,

they are usually considered a special case.

In Volek’s experiment, forty overweight subjects were randomly assigned

to one of two diets 2 : a very low-carbohydrate ketogenic diet (VLCKD),

which provided a macronutrient distribution of about 12 percent carbo¬

hydrate, 59 percent fat, and 28 percent protein; or a low-fat diet (LFD)

composed of 56 percent carbohydrate, 24 percent fat, and 20 percent

protein. The group was unusual in that they were all overweight and would

all be characterized as having metabolic syndrome. All subjects demon¬

strated the features of atherogenic dyslipidemia, a subset of metabolic

syndrome markers that describes a poor lipid profile: high triacylglycerol

(TAG), low HDL-C, and high small-dense LDL (so-called pattern B).

Saturated Fat

119

What’s striking in Volek’s work is the difference in weight loss between

the two diet regimens. Participants in the study were not specifically coun¬

seled to reduce calories but both groups spontaneously reduced caloric

intake. (Apparently, people in diet studies tend to automatically reduce

calories.) However the level of weight loss between the two groups, due

to the macronutrient composition of the plans, was dramatically different.

People on the VLCKD lost twice as much weight on average as those on

the LFD despite the similar caloric intake. Although there was substantial

individual variation (see figure 9.2 on page 134), nine of twenty subjects

in the VLCKD group lost 10 percent of their starting body weight, more

weight than was lost by any of the subjects in the LFD group. In fact,

nobody following the LFD lost as much weight as the average for the low-

carbohydrate group. The major differences between the VLCKD and LFD

groups appeared in the changes in whole body fat mass: 5.7 kilograms and

3.7 kilograms, respectively.

It is generally believed that deposition of fat in the abdominal region

is more undesirable than subcutaneous fat. Abdominal fat was found to

be reduced more in subjects on the VLCKD than in subjects following

the LFD (-828 grams versus -506 grams). Volek’s study thus provides

one of the more dramatic demonstrations of the benefits of carbohydrate

restriction for weight loss. Similar results had preceded it, though. Those

preceding studies had frequently been disparaged for increasing the

amount of saturated fat as a sort of automatic criticism (whether or not

any particular study actually increased saturated fat). Although the original

“concern”was increased plasma cholesterol, eventually saturated fat became

a generalized villain, and insofar as the science was concerned, the effects

of plasma saturated fat were assumed to be due to dietary saturated fat.

The surprising outcome of Volek’s study was that there was, in fact, inverse

correlation between dietary and plasma SFA. It was surprising because

the effect was so clear-cut that no statistics were needed, and because an

underlying mechanism could explain the results.

On Your Plate or in Your Blood?

In Volek’s study the dietary intake of saturated fat for the people on the

VLCKD averaged out to 36 grams per day, threefold higher than that of

120

Nutrition in Crisis

the people on the LFD (12 grams per day). When the relative proportions

of circulating SFAs in the triglyceride and cholesterol ester fractions were

determined, however, they were actually lower in the low-carbohydrate

group. Seventeen of twenty subjects on the VLCKD showed a decrease

in total saturates, while the other three already had low values at baseline.

In distinction, only half of the subjects consuming the LFD showed a

decrease in SFA. When the absolute fasting TAG levels are taken into

account (low-carbohydrate diets reliably reduce TAG), the absolute

concentration of total saturates in plasma TAG was reduced by 57 percent

in the low-carbohydrate group compared to only 24 percent reduction

in the low-fat group—again, this is despite the fact that LFD group had

reduced their dietary saturated fat intake. How could this happen? The

low-fat group reduced their SFA intake by one-third, yet had more SFA in

their blood than the low-carbohydrate group who had actually increased

intake. Metabolism is about change. Chemistry is about transformation.

De Novo Lipogenesis

To review the major features of metabolism bearing on fat, there are

roughly two kinds of fuel: glucose and acetyl-CoA. The big principle is

that you can make acetyl-CoA from glucose, but in most cases you cant

make glucose from acetyl-Co A—or, more generally, you can make fat

from glucose but you cant make glucose from fat. So, how do you make

fat from glucose? Part of the picture is the process of making new fatty

acids, known as de novo lipogenesis (DNL), or more accurately de novo

fatty acid synthesis. The mechanism for making new fatty acids is, in a

rough sort of way, the reverse of breaking them down. You successively

patch together two-carbon acetyl-Co A units until you reach the chain

length of sixteen carbons: palmitic acid, the saturated fatty acid that was

found in higher concentrations in the subjects on low-fat diets in Volek’s

experiment. Palmitic acid can then be elongated to stearic acid (18:0) or

desaturated to the unsaturated fatty acid, palmitoleic acid (16:l-n7, 16

carbons, one unsaturation at carbon 7).

The critical part is getting the process going. The first step is the formation

of a three-carbon compound called malonyl-CoA, a process under the control

of insulin. Malonyl-CoA enters into the synthetic process and simultaneously

Saturated Fat

121

prevents transport of fatty acid into the mitochondrion where it would be

oxidized. If you are making new fatty acid, you don’t want to burn it. New

fatty acid is a reasonable explanation for the increased SFA in the low-fat

group. The low-carbohydrate group, on the other hand, would be expected to

have higher insulin levels on average, encouraging diversion of calories into

fatty acid synthesis and repressing oxidation. How could this be tested?

It turns out that the unsaturated fatty acid, palmitoleic acid (16:l-n7), is

not common in the diet and therefore stands as a good indicator of synthesis.

The same enzyme that catalyzes conversion of palmitic acid also catalyzes

conversion of stearic acid (18:0) to the unsaturated fatty acid oleic acid

(18:ln-7), as in olive oil. The enzyme in question is named for the second

reaction, stearoyl-CoA desaturase-1 (SCD-1). SCD-1 is membrane-bound,

which means that it is not swimming around the cell looking for fatty acids,

but instead is closely tied to DNL (waiting at the end of the assembly line

so to speak), and preferentially desatufates newly formed fatty acids: palmitic

acid to palmitoleic acid and stearic to oleic. It is unsurprising, then, that the

data from Volek’s experiment show a 31 percent decrease in palmitoleic acid

(16:ln-7) in the blood of subjects on the low-carbohydrate group with litde

overall change in the average response in the low-fat group. The low-fat group

was making saturated fatty acid more than the low-carbohydrate group.

Saturated Fat in Your Blood

Cassandra Forsythe, Volek’s student at the time, extended this work to

an experiment on weight maintenance. It is commonly claimed that

the physiologic effects of low-carbohydrate diets are linked to weight

loss rather than to an inherent response to reduction in carbohydrate,

so it is important to tackle this objection. In her experiment, men were

assigned to one of two different weight-maintaining diets, both low in

carbohydrate, for six weeks. The first of the diets was designed to be high

in SFA (dairy fat and eggs), and the other was designed to be higher in

unsaturated fat from both polyunsaturated (PUFA) and monounsaturated

(MUFA) fatty acids (fish, nuts, omega-3 enriched eggs, and olive oil). For

the SFA-carbohydrate-restricted diet, the relative percentages of SFA,

MUFA, and PUFA were 31,21, and 5, respectively. For the UFA diet, the

percentages were 17,25, and 15.The results as stated in Forsythes study:

122

Nutrition in Crisis

The most striking finding was the lack of association between

dietary SFA intake and plasma SFA concentrations. Compared

to baseline, a doubling of saturated fat intake on the CRD-SFA

(carbohydrate-restricted diet with high saturated fatty acid) did

not increase plasma SFA in any of the lipid fractions, and when

saturated fat was only moderately increased on the CRD-UFA,

the proportion of SFA in plasma TAG was reduced from 31.06%

to 27.48 mol%. Since plasma TAG was also reduced, the total

SFA concentration in plasma TAG was decreased by 47% after

the CRD-UFA, similar to the 57% decrease we observed in over¬

weight men and women after 12 week of a hypocaloric CRD. 3

The bottom line: Dietary carbohydrate, rather than dietary SFA, controls

plasma SFA. Therefore, while it is widely held that the type of fat is more

important than the amount, this is not a universal principle, and it becomes

less important if carbohydrates are low. But what about the amount? A

widely cited paper by Raatz et al. 4 suggested, as indicated by the title, that

“Total Fat Intake Modifies Plasma Fatty Acid Composition in Humans.”

The data in the paper, however, show that the differences between high-fat

and low-fat were, in fact, minimal. How can you say one thing when your

data show something else? One doesn’t know what was on the authors

minds, and maybe they interpreted things differently, but in general it seems

that the literature takes an approach similar to that of lawyers: If the jury

buys it, it doesnt matter whether or not its true. In scientific publishing, the

jurors are the reviewers and the editors. If they are already convinced of the

conclusion, if there is no voir dire, you will surely win the case.

The bottom line is that distribution of types of fatty acid in plasma

is more dependent on the level of dietary carbohydrate consumption

than the level or type of dietary fat consumption. The work of Volek and

Forsythe provides a good reason to focus on the carbohydrate content of

your diet. What about the type of carbohydrate, though? In other words, is

glycemic index important? Is fructose as bad as they say? Consistent with

the small perturbation caused by fructose compared to glucose, as shown

in the previous chapter, we have a good general principle: No change in the

type of macro nutrient—carbohydrate or fat—will ever have the same kind

of effect as replacing carbohydrate across the board with fat.

-CHAPTER 8-

Hunger

What It Is, What to Do About It

T he reporter from Mens Health asked me: “You finish dinner, even

a satisfying low-carb dinner”—he is a low-carbohydrate person

himself—“you are sure you ate enough but you are still hungry.

What do you do?” I gave him good advice: “Think of a perfectly broiled

steak or steamed lobster with butter—some high protein, relatively high-

fat meal that you usually like. If that doesn’t sound good, you are not

hungry. You might want to keep eating. You might want something sweet.

You might want to feel something rolling around in your mouth, but you

are not hungry. Find something else to do—push-ups are good. If the steak

does sound good, you might want to eat. Practically speaking, you might

want to keep hard boiled eggs, kielbasa, something filling, around (and,

of course, you don’t want cookies in the house).” I think this was good

practical advice. My recommendation was based on the satiating effects of

protein food sources, or perhaps the nonsatiating, or binge-inducing effect

of carbohydrate. All this raises a larger question, though: What is hunger?

We grow up thinking that hunger is our body’s way of telling us that we

need food, but for most of us that is not the case. Few of us are so fit, or have

so little body fat, or are so active, that our bodies start calling for energy if

we miss lunch. Conversely, those of us who really like food generally hold

to the philosophy that “any fool can eat when they’re hungry.” Passing up

a really good chocolate mousse just because you are not hungry is like ...

well, I don’t know what it’s like. Of course, if you are on a low-carbohydrate

diet, you might pass it up for dietary reasons, or at least restrain yourself

from eating too much.

Getting to the point here, if I presented you with a multiple-choice

question that asked what hunger is, the answer would be “all of the above.”

124

Nutrition in Crisis

We feel hunger when we haven’t eaten for a while; we feel hunger if the

food looks good; we feel hunger if we are in a social situation in which

eating is going on (the spread of petits fours that were in the lobby at the

break in an obesity conference, the congressional prayer breakfast, or the

Pavlovian lunch bell).

Or we might eat because we think it is time to eat. This point was made

by the Restoration poet and rake, John Wilmot, Earl of Rochester. Wilmot

is famous for his bawdy poetry, raunchy even by today’s standards, but his

“Satyr against Reason and Mankind,” which is more commonly included

in texts on eighteenth-century literature, makes fun of dumb rules and

phony reason:

My reason is my friend, yours is a cheat;

Hunger calls out, my reason bids me eat;

Perversely, yours your appetite does mock:

This asks for food, that answers, “Whats o’clock?”

This plain distinction, sir, your doubt secures:

Tis not true reason I despise, but yours.

Americans have not conquered this problem, and we may in fact have

made it worse. A diet experiment invariably includes a snack as if it has

the same standing as breakfast, lunch, and dinner (anecdotally, the number

of people who prefer to go without breakfast suggests that that meal is at

least not for everyone). Visitors remark on how Americans are eating all

the time, not just at meals. If you maintain such a habit, it doesn’t take long

until you are hungry all the time.

Different people have different responses to external cues. In experi¬

ments in which subjects are interrogated, but incidentally have snacks

available—a bowl of crackers on the table, for example—it is not surprising

that thin people regulate their intake by the clock on the wall. Overweight

people, in distinction, are less sensitive to the clock and dip into the snacks

even if “it’s almost dinner time.” Similarly, at Union Theological Seminary

in New York, the school for training rabbis, it is the overweight students

who adhere better to fasting on high holy days. Consumption is less

connected to internal (physiologic) cues and external (religious) reasons

can have control.

Hunger

125

The psychologist B. F. Skinner 1 described the problem in a characteristi¬

cally dense way:

“I am hungry” may be equivalent to “I have hunger pangs,”

and if the verbal community had some means of observing the

contractions of the stomach associated with pangs, it could pin

the response to these stimuli alone. It may also be equivalent to

“I am eating actively.” A person who observes that he is eating

voraciously may say, “I really am hungry,” or, in retrospect, “I was

hungrier than I thought,” dismissing other evidence as unreli¬

able. “I am hungry” may also be equivalent to “It has been a long

time since I have had anything to eat,” although the expression

is most likely to be used in describing future behavior: “If I miss

my dinner, I shall be hungry.”

What Skinners saying is that whatever the actual causes of eating

behavior, the behavior itself may precede the description of the “motivation

to eat.” In other words, we tend to identify a feeling that is associated with

eating behavior as the cause of the behavior. “I am hungry” may also be

equivalent to “I feel like eating” in the sense of “I have felt this way before

when I have started to eat.”The point is that “hungry” only means you are

in a situation where you are used to eating. It doesn't mean that feeling

hungry will make you eat, or, more importantly, that you have to eat.

Lessons from Vagotomy

The vagus nerve contains many nerve fibers that facilitate communication

between the brain and other parts of the body (a nerve is a collection of

nerve cells or neurons whose long extensions or axons are referred to as

fibers). Cells that send signals from the brain to distant organs are called

efferent. Efferent fibers in the vagus nerve regulate the digestive tract in

various ways: enlargement of the stomach, for example, or secretions from

the pancreas to deal with larger volumes of food (known to doctors as

accommodation). Most of the fibers in the vagus nerve are sensory affer-

ents (afferents carry information from the body to the brain) providing

sensations of satiety and hunger as well as a feeling of discomfort when

126

Nutrition in Crisis

we are full. (Efferent is usually pronounced “EE-ferent” to distinguish

from afferents.)

Vagotomy, cutting the vagus nerve, was historically practiced as a means

of controlling ulcers, and is still a target, at least experimentally, for treating

obesity. Dr. John Krai at the Department of Surgery at Downstate, who

has performed such operations, described to me how patients complained

that they had lost their appetite. He had to explain to them that you do not

have to eat all the time, and that nothing bad will happen if you miss a few

meals. Hunger is a signal that you are used to eating in a particular time or

situation. You are not required to answer that signal.

“You Eat Because You Are Fat”

In trying to go beyond energy balance there is a tendency to think of

hunger in terms of hormones, emphasizing regulation by the hypothala¬

mus, analogous to temperature regulation. The hormones in question are

referred to as either orexigenic, increasing appetite (from the Greek: the

Greek equivalent of bon appetite is kali orexi ), or anorexigenic, depressing

appetite. While this is part of the picture, it leads to some confusion because

the endocrine approach emphasizes hormonal output from the fat cell, and

in some sense bypasses the question of how the fat cell got fat in the first

place—that is, how it bypasses metabolism. More importantly, for animals

and humans outside of a laboratory setup, behavior overrides hormones.

The analogy is not entirely accurate in that animals (and humans) regulate

their temperature hormonally only to a small extent. The major control of

temperature is behavioral: We put on clothes and we hide in caves.

An important aspect of this problem is the need to understand the error

in “a calorie is a calorie.” One critique of the energy balance model runs

something like this:

dietary (other increased greater

, , : -> insulin x

carbohydrate hormones) appetite consumption

In the extreme case, the explanation might boil down to: “You don’t get

fat because you eat; you eat because you got fat.”This doesn’t make much

sense. It sounds like one of the seemingly profound academic aphorism

Hunger

127

that Woody Allen was so good at parodying: “All of literature is just a

footnote to Faust.” I understand that it implies that the hormonal secre¬

tion from adipose tissue encourages eating—but again, it does not tell you

why you got fat in the first place. It mixes up metabolism with behavior

and implicitly accepts the idea that calories are what count—that it is the

total energy you consume that matters, rather than how that energy is

processed. Although macronutrients clearly differ in satiety, regardless of

your hormonal state, if there is no food, you will not increase consumption.

Also, the effects of insulin are not so clear-cut. Whereas metabolically

insulin is anabolic, at the level of behavior it is probably anorexigenic in

most cases.

Why do we get fat? We get fat because we eat too much of the stuff that

encourages excessive weight gain. We don’t know what that stuff is, but

we know that it is not fat per se. Given the unambiguous effectiveness of

carbohydrate restriction in reducing excess weight, it would be surprising

if carbohydrate weren’t a big part of the picture. The so-called metabolic

advantage (less weight gain per calorie), where it exists, is a metabolic

effect. The most likely mechanism appears to be due to the effect of insulin

on rates of reaction: Anabolic (storage) steps might increase accumulation

before competing feedback (breakdown) can catch up. As I will explain in

chapter 9, it rests on nonequilibrium thermodynamics, 2 which recognizes

the importance of rates as well as energy.

What Can You Do About It?

The suggestion at the beginning of this section was to make sure you know

what kind of hunger you are talking about. Behavioral psychology stresses

the difference between “tastes good,” and the technical term, “reinforcing,”

which means that the food increases the probability that you will keep

eating. Anecdotally, we all have the experience of saying, “I don’t know why

I ate that. It wasn’t very good.”

However little you eat to answer feelings of hunger, it is certainly bad

advice to eat if you are not hungry Professional nutritionists, even the

Atkins website, are always telling you to have a good breakfast. Why you

would want to have any “good” meal if you are trying to lose weight is not

easy to answer. The nutritionists claim that without breakfast, you will eat

128

Nutrition in Crisis

too much at the next meal, as if in the morning you can make the rational

decision to eat breakfast despite no desire for food, while at noon you are

suddenly under the inexorable influence of urges beyond your control. A

more reasonable way to put it would be: “If you find that you eat too much

at lunch when you don't eat breakfast, then ... ,”but that is not the style of

traditional nutrition.

More on Behavior

The principle in behavioral psychology is that how soon a behavior is

reinforced is more important than the quality of the reinforcer. It is

important to reemphasize that reinforcement is not always equivalent

to reward. Reinforcement simply means increasing the likelihood that a

behavior will reappear, even if that behavior doesn't feel totally good Junk

food may be reinforcing in that it will make you eat more even though

you realize afterward it didn't taste that good. Taste and mouth feel are so

immediately reinforcing that, in all likelihood, only aversive stimulation

can work well to discourage eating. There are positive ways to look at

hunger. For example: Feeling hungry may mean that your diet is working,

that you really are losing weight, and therefore you might stop eating

before satiation. However encouraging that notion might be, it usually

cant compete well with even the smell of food. You need something

strongly negative.

One of the more effective regimens is a diet strategy from Dr. Allen Fay,

a psychiatrist in New York. It works like this:

1. You pick an amount of weight you want to lose in the next week; you

can pick zero but, of course, you cant go up.

2. You write a check for $2,000 to an organization that you don't like

(Republican National Committee, in my case) and give it to Dr. Fay.

(Note: This is a personal choice. This is not a political book.)

3. If, at the end of the week, you haven't hit the weight target, he mails

the check.

In some cases, Dr. Fay said, you don't even need money. One patient

wrote a letter, poised to be sent to a right-to-life organization, telling

Hunger

129

Exercise

Hie only thing that people in nutrition agree on is the value of

exercise. While it is not as important as diet for weight loss, it

does positively interact with diet and has obvious benefit. One

question that comes up is when to have meals in relation to

exercise. Although outside my area of expertise, and likely to be

an individual thing, there is some good guidance in the follow¬

ing old joke:

The couple go to the doctor and don't want to have

any more children but they don't want any artificial

methods of birth control. The doctor recommends

exercise.

Husband: Before or after?

Doctor: Instead of.

them what a great job they were doing. The thought that she would get

on their mailing list as a supporter was sufficiently unpleasant to keep

her on target.

You pick your own threat, of course—it is not a political method. Dr.

Fay suggested the American Nazi Party, but I thought that they would only

buy those shabby uniforms, whereas from my perspective, the Republicans

would do real damage. My relation to Dr. Fay is partly professional

(although he admits there are people who are beyond psychotherapy) and

partly friend. For the technique to work, you must be distant enough from

the person holding the check so that they will actually mail it, but close

enough that you will not consider physical violence if they do.

Imagery can help, but only up to a point. A major problem situation

for dieters is that they have eaten what they want and feel satisfied, but

130

Nutrition in Crisis

there is still food on their plate, which they pick at until they feel sick.

My approach when I felt full was to imagine spiders coming out of the

food. This worked for a while, but over time I noticed that I was losing my

distaste of spiders. Eating has a very strong Pavlovian component. It is not

nice to fool Mother Nature.

-CHAPTER 9-

Beyond

“A Calorie Is a Calorie”

An Introduction to Thermodynamics

A rnold Sommerfeld was one of the great physicists in the develop¬

ment of quantum mechanics (theory of atomic structure). He was

also generally considered to be an expert on most areas of physics.

His take on thermodynamics:

The first time I studied it, I thought that I understood it

except for a few minor details.

The second time I studied it, I thought that I didn’t under¬

stand it except for a few minor details.

The third time I studied it, I knew I didnt understand it but it

didn’t matter because I already knew how to use it. 1

Can you really lose more weight on one diet than another if you

consume the same number of calories? The question usually comes in

response to a low-carbohydrate diet, where the so-called metabolic

advantage promises you that cutting out carbohydrate will lead to

reduced efficiency in storing fat. Folks go crazy when you suggest its

true. Whenever a scientific paper presents data showing that such a thing

really happens—that one'diet, usually low-carbohydrate, is more effec¬

tive than another—somebody always jumps in to say that it is impossible,

and that it would violate the laws of thermodynamics. Like the cartoon

characters who run over the cliff, remain suspended in thin air, and fall

only when they realize that they aren’t on solid ground, somehow the data

are expected to go away once thermodynamics is invoked. Of course, the

132

Nutrition in Crisis

data can't violate the laws of thermodynamics. The real possibility is that

the data are accurate and that the critic doesn't get it. Thermodynamics—

the physics of heat, work, and energy—is a tough subject, and it takes real

chutzpah to jump in where many physicists fear to tread.

Although it may be difficult to grasp, thermodynamics is interesting. It

has been described as the first revolutionary science. You probably don't

really need it to study nutrition, but if you catch on to the basics, you will

understand something that seems counterintuitive to many people. It will

explain how you get more bang for your nutritional buck—that is, how you

lose more weight per calorie.

When you consider that the fundamental unit in nutrition is the calorie,

a unit of energy, it seems likely that it would be worth knowing something

about the physics of energy exchange.

The Bottom Line on Metabolic Advantage

Here are the four main principles of metabolic advantage that will guide our

discussion. The rest of the chapter will explain and justify these conclusions:

1., Metabolic advantage, or a better term, energy efficiency, is not contra¬

dicted by any physical law.Thermodynamics, in fact, more or less predicts

variable energy efficiency. The way it has been discussed in nutrition is

incorrect and does not conform to the way chemical thermodynamics is

normally used.

2. Arguments against metabolic advantage often rely on practical consider¬

ations: how small the effect is. Yet the same critics espouse the value of

cumulative small effects, operative in diets where you explicitly control

calories, where 50 calories a day is supposed to add up over a year—and

it doesn't. Metabolism doesn’t work like that. Homeostatic (stabilizing)

mechanisms compensate for simple changes in calories unless they are the

right kind of calories, and in fact, the effects of different macronutrients

can be dramatic. In any case, if there really is any change at all, that should

be a call to figure out how to maximize that change, rather than ignore it.

3. Even if you aren’t sure that metabolic advantage has been effectively

demonstrated, it makes sense to try to make it work for you, since there

is a great deal of potential scientific payoff.

Beyond “A Calorie Is a Calorie

133

4. Several mechanisms, particularly substrate cycling and gluconeogenesis,

are involved. Experimentally, inefficiencies in digestion and metabolic

processing (the so-called thermic effect of feeding) contribute as well.

Many people find metabolic advantage counterintuitive because the

idea of energy conservation has been so deeply ingrained in their minds,

I will explain the fallacy. I will present some of the data and then explain

how the process plays out in terms of biochemical mechanisms.

The Data

There are basically two kinds of diet experiments. Some clearly show

energy balance and some clearly do not. Here's an example of the former:

If you take a normal person, keep them in a hospital room, and feed them

constant calories (figure 9.1), you will find that it doesn't matter much how

the calories are distributed among different foods. “Wide variations in the

ratio of carbohydrate to fat do not alter total 24-h energy need” 2 —their

weight will stay roughly constant. In some experimental cases, like this, a

calorie is indeed just a calorie. Two people who are roughly similar in age

and health will respond similarly to two isocaloric (same caloric value)

diets regardless of the diet composition (amount of fat, carbohydrate, and

protein).This means that, yes, calories count. Yet there are many exceptions.

The energy balance shown in figure 9.1 is achieved by biologic mecha¬

nisms, not the laws of thermodynamics. In those cases, where everything

balances out, it isn't physical laws but rather the unique characteristics of

Days

10 24 38 52 66 80 94 55

< -► 10% Fat--70% Fat-►

Figure 9.1. A thirteen-week study of a subject first on 10 percent (75 percent carbo¬

hydrate) of energy intake as fat and then on 70 percent (15 percent carbohydrate)

of energy intake as fat. From J. Hirsch et al., “Diet Composition and Energy Balance in

Humans,” American Journal of Clinical Nutrition 67, no. 3 (1998): 551-55S.

134

Nutrition in Crisis

Figure 9.2. Comparison of low-carbohydrate ketogenic diet and low-fat diet. From

J. S. Volek et al., "Carbohydrate Restriction Has a More Favorable Impact on the Metabolic

Syndrome than a Low Fat Diet" Lipids 44, no. 4 [2009): 297-309.

Beyond “A Calorie Is a Calorie”

135

living systems that keep things constant through the process of homeosta¬

sis. Big rule: In biology almost everything is connected in feedback, and

homeostatic mechanisms compensate for chemical changes. If you reduce

your dietary intake of cholesterol, for example, your body will compensate

by synthesizing new cholesterol. In view of this, the question we should

be asking is, “How is energy balance possible when it is not predicted

by thermodynamics?”; not, “How could there be different weight gain or

loss for the same number of calories?” So let s look at the exceptions—the

cases where the energy does not balance out—the second type of dietary

experiment seen in the literature.

The exceptions can be dramatic. The experiment from the laboratory

of Jeff Volek, then at the University of Connecticut, described previously

in chapter 7, studied forty overweight men and women with metabolic

syndrome (high triglycerides, low HDL, or at least two other factors).They

were assigned to one of two ad libitum diets: a very low-carbohydrate keto-

genic diet (VLCKD) (% carbohydrate [CHO]:fat:protein = 12:59:28), or a

low-fat diet (%CHO:fat:protein = 56:24:20.) The experiment lasted twelve

weeks. Although neither group was specifically told to reduce calories, both

groups did show a spontaneous decrease in energy intake—it seems that if

you sign up for a diet experiment, you automatically eat less.

Figure 9.2 shows the dramatic difference in performance between the

VLCKD and the LFD. Figure 9.2 A indicates the average effect: Weight loss

in the VLCKD group was dramatically better. In reading the medical litera¬

ture, however, it is important to ask about individual performance. People are

different. Nobody loses an average amount of weight. People in both groups

lost weight, but what is remarkable is the number of people on the VLCKD

that lost a lot of weight (Figure 9.2J5). Half of the people on the VLCKD lost

more weight than the single most successful subject on the LFD.

Can You Trust Dietary Records?

Once again, there is the problem of patient dietary records. These records

are known to be inaccurate, but not wildly so. Values can be off by 20

percent, but rarely by 50 percent, and are unlikely to be sole cause of the

differences. Ketone bodies were measured, as well, so at least the VLCKD

group did what they were told. It is important to reiterate that if the

136

Nutrition in Crisis

differences were due to inaccurate reporting of dietary intake (that is, if

the diets were truly different in caloric intake), the people on the VLCKD

must have overreported what they ate and the people on LFD must have

underreported what they ate, or both. Bottom line: Food records have very

high error rates. They are meaningful if the differences in the observed

experimental differences are large, or if there are indicators of compliance

such as the presence of ketone bodies.

Low-carbohydrate diets almost always win in a face-off with low-fat

diets. Establishment nutritionists will write off the ties as wns, but they

can only do so for so long. So if it is not just about the calories, where

does thermodynamics really fit in? With the disclaimer of Sommerfeld's

comments at the top of this chapter, I will give you a rough idea about how

it's done in real biochemistry.

Thermodynamics

Thermodynamics is the physics of heat, work, and energy. The subject is

simultaneously theoretical and mathematic, and fundamentally down to

earth. Its roots are in the attempt to find out just how efficient a steam engine

you can build. Thermodynamics came about during the industrial revolution,

when the efficiency of steam engines was a big deal. In weight loss, we are

really asking how efficiently food is utilized—much like the efficiency of a

steam engine's fuel. The difficult side of thermodynamics is that its methods

are highly mathematical and arcane, even for scientists. Thermodynamics

has been described as “the science of partial differential equations,” which is

not to everybody's taste. The results, however, give you very simple equations

for predicting things. Its a combination of heavy-duty theory and practical

application. This is what those of us who are interested in thermodynamics

actually like about it, and its also the main theme of this book: Science with

direct applicability. You get an equation that tells you whether you have a

good steam engine, or, in fact, whether your food is fattening.

Real Thermodynamics

When people evoke the laws of thermodynamics, they're usually just talk¬

ing about the first law, the law of conservation of energy. Unfortunately,

Beyond “A Calorie Is a Calorie 1

137

"conservation of energy” has become a sound bite, at the level of "Einstein

said that everything is relative.” You have to know exactly what is being

conserved, and more important, in thermodynamics, what is not. What

follows is how thermodynamics is taught and used in science. The math

is simple, but understanding depends on precise definitions and careful

logic. What I’m about to cover many not be necessary for nutrition, but

since thermodynamics is so often invoked, it is worth going through at

least once.

The first law says precisely that there is a parameter called the internal

energy and the change (A) in the internal energy of a system is equal to

the heat, q f added to the system minus the work, w, that the system does

on the environment. (The internal energy is usually written as U so as not

to confuse it with the electrical potential.) In other words, the energy of a

system is not determined directly, but rather by the difference between the

heat and work that changes when the system changes.

A u = q - w

This is how thermodynamics is taught. To get to the next step you

need to understand the idea of a state variable. A state variable is a vari¬

able where any change is path-independent. For example, the familiar

temperature, T, and pressure, P, are state variables: It doesn't matter

whether you change the temperature quickly or slowly because the

effect on the system is controlled only by the difference between the

temperature after the change minus the temperature before the change

(AT). The usual analogy is the as-the-crow-flies geographical distance,

say, between New York and San Francisco. This is a state variable: It

doesn't matter whether you fly directly or go through Memphis and Salt

Lake City (like the flights I wind up on)—all that matters is the final

state you arrive in.

The energy U is a state variable. Any process that you carry out will have

a change in U that depends only on the initial and final states. However, q

and w are NOT state variables (which is why they are written in lower case

or with some other marking to distinguish them from state variables). How

you design your machine will determine how much work you can get out

of it and how much of the energy change will be wasted as heat. Looking

138

Nutrition in Crisis

at the biological case: Two different metabolic changes, such as responses

to different meals, might have the same total energy change, but the work

(both physical and chemical) and heat you generate are likely to differ

if the two meals have very different inefficiencies. There is no theoretical

reason why they couldn’t have the same efficiency just by coincidence, but

it is highly unlikely

A preview of the second law is contained within the first law itself. The

second law is what thermodynamics is really about—it is why thermo¬

dynamics took so long to evolve and why it is so hard to understand. The

second law pretty much guarantees, contrary to popular misconception,

that “a calorie is not a calorie.”

Lets take a step back. There are four laws of thermodynamics, and the

first law only operates in concert with the others. The zeroth law and the

third law are pretty much theoretical, defining thermal equilibrium and the

condition of absolute zero of temperature. It is the second law that embod¬

ies the special character of thermodynamics. Described by Ilya Prigogine,

the Nobel prize-winning chemist and philosopher of thermodynamics, as

the first revolutionary science, the second law explains how one diet can be

more or less efficient that the other. The essential feature of the second law

is the existence of a thing called entropy

Entropy

Entropy is a measure of how disorganized a system is—that is, what its

possibilities are. As described below, it is a measure of information. A gas

filling a box completely is said to have higher entropy than a gas that is

confined to only one side of the box. Another description: You would not

need a very good GPS to find out if the molecule is in New York City

versus whether it is in Yankee Stadium.

Entropy is traditionally defined with respect to a classic imaginary

experiment. Although theoretical, it is clearly related to practical things.

The experiment involves a creature referred to as the Maxwells demon,

who is capable of doing things perfectly smoothly and slowly without

exerting any energy or creating any heat from friction. The experimental

apparatus is a box with a partition—a membrane that separates the box

into two compartments, one that is filled with a gas and one that is empty.

Beyond “A Calorie Is a Calorie”

139

Maxwells demon very carefully and slowly removes the membrane so

that no work is done and no heat is generated. According to the first law

of thermodynamics, which specifies the need to conserve energy and heat,

nothing has really happened. In a real experiment, you could make an

electronic device to open the partition so efficiently that it would hardly

raise your electric bill—so effectively that energy is conserved and nothing

should happen. Of course, you know that something will happen. The gas

will fill up the whole box. Why? Because thats the way the world works

according to the second law: Entropy will always increase and the gas will

always spread as much as it can.

The gas fills up the whole box because it is the most probable way for

the molecules to be distributed. The second law is about probability. Before

the nineteenth century, physics held a view of a universe that was mechani¬

cal, a universe standing alone in time and controllable. The second law

suddenly brought out the irreversible and statistical nature of things. It is

not that nobody knew that time passed before that point, but the idea of

physical processes running down like a clock—the notion of irreversible

processes—was revolutionary. The history of physics shows how hard it is

for people, even very smart people, to understand and cope with the idea

of probability. As it evolved, however, the second law became very practical

and explained how energy can be used to do work and how chemical reac¬

tions occur. It is the key to understanding energy transformation in life.

Entropy and Information

Entropy is about information. It is frequently said that increased entropy

means increased disorder. The more precise way to put it would be that

increased entropy means a less demanding way of arranging a system,

and in turn, higher probability. In poker, a straight flush has much lower

entropy than two of a kind because there are fewer ways to get a straight

flush. Here’s where it gets more complicated: Entropy—that is, informa¬

tion—can overpower energy. A receiver in American football can catch

a pass even though he is double-teamed (his energy is compromised)

because he has the information to know where the ball will be thrown.

(The term entropy is also used in communications, where it indicates the

extent to which a message has been corrupted during transmission.)

140

Nutrition in Crisis

There are many ways to state the second law. One formal way says that

it is impossible to carry out an operation where the sole effect is to transfer

heat from a cold body to a warm body. This is obvious enough but it might

not be easy to explain why. After all, the ocean is cold, but it has a huge

amount of heat because it is so large. Why couldn’t a ship extract a little bit

of that heat to run its engines? That’s the question Maxwell was really asking.

The original creature proposed by Maxwell was more sophisticated

than the demon described above. This original demon sat on a membrane

separating two compartments. One compartment had a hot gas, and the

other, a cold gas. The partition had a little door—a very well-machined,

well-oiled door, whose openings and closings did not require any real

amount of work and did not generate any heat. The demon would look

into the cold gas and if he saw a fast-moving particle, he would open the

door and let the particle into the hot compartment. He would, similarly,

let slow-moving particles move from the hot gas to the cold gas. The net

effect was to transfer heat from a cold body to a warm body, a clear no-no

according to the second law. It was a thought experiment, but you really

could make a super efficient door, so why couldn’t you make something

at least close to this setup? The ocean is cold but it is so far from absolute

zero and it is so big that there is so much heat, why couldn’t you make

something like a Maxwell’s demon to get some of it to power your ocean

liner? Was the second law not universal?

This paradox completely stumped physicists of the nineteenth century.

It’s because it’s about probability and randomness and they were not used

to thinking like that. The answer to Maxwell’s paradox is that it is not really

a paradox so much as another way of stating the second law. That’s how

the world is. Information costs. You can’t make such a setup. If you could

use some fancy electronic device to set up a sensor to distinguish between

fast and slow molecules, it would still have to do work. The reason it gets

confusing is that in the physical (real) world, we are only used to dealing

with gross collections of things and averages. If we want to get down to

single molecules, we have to run a machine to do it. The simpler way of

saying it is that the reason that we can’t make anything perfectly efficient

is because we can’t control individual molecules.

The point of all this is to say that, if you’re going to carry out all the

work, physical and chemical (that is, metabolic), using the energy from

Beyond “A Calorie Is a Calorie”

141

food, then you can’t do it perfectly efficiently. Some part of the energy must

be wasted. Of course, unlike an engineer, if you are trying to lose weight,

the job of synthesizing and storing fat might be one that you want to run

as inefficiently as possible.

The Second Law and Metabolic Advantage

There are many different ways of stating the second law, but one version

emphasizes the fact that all real engines—in fact, all processes in the real

world—are inherently inefficient—not just practically, not because you

cant construct them so carefully that theres no friction, but theoretically,

absolutely. There is no escape from inefficiency. The second law says that

a perfect engine, living or otherwise, is not possible (unless you could get

one to run at the mysterious temperature called absolute zero; the third

law does give you that).

Inefficiency depends on where you stand and what you are trying to

accomplish. In human beings, keeping warm (that is, using food for heat)

might not be considered inefficient but from the point of view of a machine

that is trying to manufacture protein and other cell material, it is energy that

is wasted. The heat generated in the processing of food is called either ther¬

mic effect of feeding (TEF) or diet-induced thermogenesis (DIT). (The old

name was “specific dynamic action.”) These are a measure of energy wasted

as heat. They are an expression of the inefficiency of the human machine.

There are other ways to waste energy: for example, the so-called NEAT,

nonexercise activity thermogenesis, which is the scientific name for fidget¬

ing. The measured TEF of different macronutrients is different: Protein is

much greater than carbohydrate, which is greater than fat. In other words,

metabolic advantage is a well-documented fact, and the extent to which

small changes add up is only a question of how you do the experiment. If

you have to make glucose through the process of gluconeogenesis, rather

than getting it from the diet, you are going to waste energy.

Beyond “Calories In = Calories Out”

In chemical thermodynamics we want to know whether a chemical reaction

can be made to produce energy, or whether we have to put in energy to make

142

Nutrition in Crisis

the reaction go. As in the industrial revolution, we want to make an efficient

chemical machine—we re thinking, of course, of a machine where you put in

food and out comes enough energy to lift heavy weights, or more to the point,

synthesize complicated chemicals like proteins. The key idea rests with iden¬

tifying the likelihood of a reaction going forward with the reactions ability to

require or to give off energy. With the discussion of the second law behind

us, we can guess that it is not about conservation. It is about dissipation of

energy. The 4 kcal/g that we assign to the energy in carbohydrate is the energy

exported from the reaction of oxidation to the environment. That’s what you

do in chemical thermodynamics. Otherwise, all food would have zero calo¬

ries: The heat lost in oxidation is gained by the calorimeter. Calories in equals

calories out, sure, but that literally leaves you with absolutely nothing.

What comes next shows you how real thermodynamics works. It is not

particularly mathematical. All you need is high school algebra. You can see

the beauty of thermodynamics and how it can tell you, right off, that its

not just calories in, calories out (commonly abbreviated CICO on social

media). You may like it, but if you prefer to avoid the math, you can skip to

the summary at the end.

The second law says, in essence, that all (real) systems are inefficient.

In practice, the law can be used to tell you whether a chemical reaction

actually produces energy or, as we have said, whether you will have to put

in energy to make it go. This is the key in chemistry (and living systems in

general): Does the reaction proceed as written?

When we write A B, we want to know whether the reaction will

proceed from left to right spontaneously. “Spontaneously” means without

the addition of energy. It does not mean fast, which is a separate question.

The 4 kcal of energy that you measure in the calorimeter is both a measure

of the tendency of the reaction to occur (oxidations generally go easily by

themselves and produce energy) and also the maximum energy available to

do work. They are very closely related, because although living systems do

mechanical work, the main use of the energy is in chemical work: synthe¬

sizing metabolites and cell material. The second law leads to the definition

of a number of different forms of the energy that are used under different

conditions. The particular form of the energy that is used under conditions

of constant temperature and pressure—the conditions in which biochem¬

ists typically work—is called the Gibbs free energy, which is almost always

Beyond “A Calorie Is a Calorie”

143

abbreviated with the letter G, and whose change is written with the Greek

delta, AG. The Gibbs free energy for a chemical is precisely defined as the

maximum work you can get from running the reaction at constant tempera¬

ture and pressure, and it is identified with the tendency of the reaction to go

in the forward direction. The 4 kcal produced by the oxidation of glucose

tells you that is the most you could get out of it in terms of work or driving

other chemical reactions. In practice, some energy might be wasted as heat

or other unproductive processes.

To summarize what we have just discussed: Chemical thermodynamics

emphasizes the reaction of the system, not the whole universe. We want

to know about the energy exchange when we burn food. The complete

oxidation of glucose in the calorimeter produces 4 kcal. It is not about

conservation. It is about dissipation of energy

The word thermodynamics is thrown around a lot in nutrition, mostly by

people who have no idea what it s about. Again, you don t need thermody¬

namics to do nutrition, but if youre going to bring thermodynamics into

the picture, you have to do it right. So, in case you want to see what people

really do in chemical thermodynamics, I will present a good example.

Let s begin with energy change, meaning Gibbs free energy Here we have

a grand rule: If the free energy change is negative (AG < 0) for the reaction,

the reaction is downhill, will go by itself, and will give off energy (which you

might be able to capture by coupling it to another chemical reaction or to

some mechanical, electrical, or heat machine).The Gibbs free energy has two

components, the heat of reaction, called enthalpy (A#) and the entropy (AS).

Here's a simple example of what you might do in real thermodynamics,

which will also illustrate why the calorie equivalence idea in nutrition is

not right. Suppose that you wanted to know about the formation of carbon

monoxide (CO)—if carbon is oxidized to CO, how much energy is given

off in the process. Generally thought of in the context of a poison, CO has

other uses—small amounts are actually produced in the human body. So,

we want to know: Is the oxidation of carbon to CO uphill or downhill and

by how much? To keep it simple, well take the enthalpy (heat of reaction,

A H). We can do the experiment so that the entropy is not an important

player (low temperature). The heat of reaction is easily measured.

In the case of oxidizing carbon, then, if heat is given off (AH = (-)),

the reaction will be spontaneous and go by itself. The problem with trying

144

Nutrition in Crisis

to figure out how much energy you can get by burning carbon to carbon

monoxide is that you cant really measure it. If you try to carry out the

reaction, you always get some CO 2 . So, what can you do? Here’s how to

deal with it: We can’t measure the heat of reaction for oxidation of carbon

to CO directly (again, because it is always complicated by some CO 2 being

formed), but we can measure the enthalpy of burning of carbon to CO 2 (a

minus sign means heat is given off):

C + O 2 CO 2 AH = -94 kcal

We also know the energy of burning CO to CO 2 :

CO + V 2 0 2 -> C0 2 AH = -68 kcal

Another great simplifying feature of thermodynamics is that if we know

the energy for a chemical reaction (or any process), the energy for going

the other way is the same numerically, only with the opposite sign, so:

C0 2 -> CO + % 0 2 AH = +68 kcal

The energy functions, G and H, are state functions. Remember from

earlier in this chapter that means they are path-independent, or independent

of the conditions under which a particular reaction is carried. State functions

can be added just as in simple algebra. The energies add up (figure 9.3). Here

you see the beauty of thermodynamics. The attraction to those of us who like

it is that you can manipulate the results with elementary algebra. The great

simplicity in this kind of calculation reflects its highly predictive power.

What did we do? We had two different paths from carbon to carbon

monoxide: one (two-step) path that we could calculate, and one that we are

trying to find out. We knew they must be equal. The principle that allows

you to add up heats of reaction is called Hess’s Law.

Hess’s Law Shows That a Calorie Is Not a Calorie

The idea that “a calorie is a calorie,” means that the energy yield for

metabolism will be path-independent—that is, the same for all diets and

Beyond “A Calorie Is a Calorie

145

C + 0 2

AH = ?

> CO

c + o 2 -► co + y 2 o 2

AH = -110.5 kj

Protein

■> Carbohydrate

Figure 9.3. Hess’s Law. A, Calculating heat of reaction for formation of CO. Path of the

arrows (measured] must equal the direct conversion to CO, so we just add them up.

B, According to Hess’s Law (adding up energies), the energy for path 1, AG, should be

equal to the energy for path 3 followed by path 2, A G^ - AG 2 . Using calorimeter values

and the principle that a calorie is a calorie leads to a contradiction. Note: Energies in

figure in kj = 4.28 kcal.

proportional to the calorimeter values. To show that this is not true in

general we use the simple additive property of state variables as in the

carbon monoxide problem above. We will look at oxidation of protein by

146

Nutrition in Crisis

two pathways, either directly or indirectly by first converting the protein to

glucose through gluconeogenesis (GNG) and then oxidizing the glucose

to CO 2 . The laws of thermodynamics say that these should be the same.

The calorimeter values say that energy yield for carbohydrate and for

protein are equivalent: AG (oxidation) = -4 kcal/mol (figure 9.2). The

(-) sign, again, means energy is given off and the process is spontaneous,

Here’s the plan. We make multiple paths for oxidizing protein: path 1

(direct), and paths 2 and 3 (indirect):

Path 1: Protein + 02 “^ CO 2 + H 2 O AGi = 4 kcal/g

Path 2: Carbohydrate + O 2 ^ CO 2 + H 2 O AG 2 = 4 kcal/g

Path 3 : Protein—GNC -> Carbohydrate + O 2 ^ CO 2 + H 2 0 AG 3 + AG 2 = ?

In path 1, we burn protein directly to CO 2 . Because free energy is a

state variable, the free energy AGi must be equal to the sum of AG 2 , the

energy for path 2, plus AG 3 , the energy for path 3. This means that AG 3

for path 3 must be about zero. However, path 3 includes the process of

gluconeogenesis. Chemistry students work very hard in order to learn that

gluconeogenesis is an expensive, endergonic (energy requiring) process: It

costs energy (about 6 ATP) to turn a mole of protein into a mole of carbo¬

hydrate. It is the failure to take this into account—the assumption that

only the calories measured in a calorimeter are important—that leads to a

contradiction. Once you take into account the real metabolic processes, it

becomes clear that “a calorie is a calorie” is not a valid principle. The exam¬

ple bears on the real behavior of living systems, and is likely a contributor

to the benefits of dietary carbohydrate restriction. More generally, if one

diet is reported to be more effective for weight loss than another, there is

no reason not to take it seriously. Like any scientific result, it should be

evaluated carefully, but if it is your body mass that you are concerned about,

the stakes are high.

The Nonequilibrium Picture

There is one more level of sophistication to address. To some extent, it is

not really about thermodynamics at all, or at least not equilibrium ther¬

modynamics (the energy required to go between states in equilibrium).

Beyond “A Calorie Is a Calorie”

147

Breakfast Lunch Dinner

Meal

Figure 9.4. Hypothetical model for the effect of insulin on efficiency of storage. The

lower line, indicating response under conditions of weight maintenance, fluctuates

but averages to no change. The upper line shows the effect of added insulin on

hormone-sensitive lipase activity. The important point is that the energy differences

are very small compared to the equilibrium value, which would be very far below the

figure. The system is not controlled by energy but by rates.

Equilibrium thermodynamics is what is usually studied, and we are taught

that rates of reaction are considered separately from energy Equilibrium

thermodynamics tell, for example, that amino acids are more stable than

proteins. The rate of breakdown is very slow. If you could keep the bacteria

off your steak it would last for months, or even years, but at the end of

time it would be all amino acids, (or even simpler things). In biochem¬

istry, however, rates become important because living systems are not at

equilibrium until they die. Things are moving forward. All the reactions in

biology are run by catalysts—that is, enzymes that control the speed of a

reaction, rather than the energetics. A better way to put it might be to say

that the key players in all this are hormones and hormones generally affect

enzymes, which in turn affect rates, not energy.

Living systems are not at equilibrium. Living systems, in fact, main¬

tain themselves very far from equilibrium. They are characterized by an

in-and-out flux of material and energy. In a dietary intervention, material

fluctuates around a level far from equilibrium. In other words, changes

148

Nutrition in Crisis

with time become important and changes might be controlled by the pres¬

ence of catalysts—enzymes or other factors that affect the rate of reaction.

Figure 9.4 shows the theoretical fluctuations of fat within a fat cell.

The key idea is that the reactions, breakdown, and resynthesis of the fat

molecule are very far from equilibrium (at equilibrium, you would have

very little fat, mostly fatty acid, and free glycerol). Looking at fat gain and

loss (figure 9.4), adipocytes cycle between states of greater or lesser net

breakdown of fat (lipolysis and reformation) depending on the hormonal

state, which in turn is dependent on the macronutrient composition of

the diet. A hypothetical scheme for changes in adipocyte TAG and a

proposal for how TAG gain or loss could be different for isocaloric diets

with different levels of insulin is shown in the figure. The basic idea is that

fat fluctuates, but if you slowly store enough so that the fat molecules you

form don't have a chance to break down before you consume another meal,

fat will accumulate regardless of caloric input.

Under normal control conditions of weight maintenance, the break¬

down and utilization of TAG follows a pattern of lipolysis (fasting), and

intake plus resynthesis (meals). To make it simple, let’s assume a sudden,

instantaneous spike in food at meals. Then the curves represent the net

flow of material within the adipocyte. This averages out to a stable weight

maintenance (lower dotted line in figure 9.4). If we keep each meal at

constant calories but increase the percentage of carbohydrate or otherwise

generate a higher insulin level, the hormone-sensitive lipases, the enzymes

that catalyze the breakdown of fat, will be inhibited (the solid line in figure

9.4). Resynthesis of TAG is thus less affected by the elevated insulin and

might actually slow down. The net effects are changes in the direction

of accumulation of TAG. It is theoretical, but the model shows you how

kinetics (how fast things happen) might be more important than thermo¬

dynamics (how stable they are at the end of the reaction).

PART 3

The Low-

Carbohydrate Diet

for Disease

-CHAPTER 10-

Diabetes

At the end of our clinic day, we go home thinking, “The

clinical improvements are so large and obvious, why don't

other doctors understand?” Carbohydrate-restriction is easily

grasped by patients: because carbohydrates in the diet raise

the blood glucose, and as diabetes is defined by high blood

glucose, it makes sense to lower the carbohydrate in the diet .

By reducing the carbohydrate in the diet, we have been able to

taper patients off as much as 150 units of insulin per day in

8 days, with marked improvement in glycemic control—even

normalization of glycemicparameters .

—Eric Westman 1

T he scene is lunch in the cafeteria at SUNY Downstate Medical

Center. I am going on about how strange it is that carbohydrate

is recommended for people with diabetes, a disease whose most

salient symptom is high blood glucose. Overhearing my story, several

clinicians at the table ask incredulously, almost in unison, “Who gives

carbohydrates to diabetics?” I say, “The American Diabetes Association

(ADA). They recommend 55-60 percent carbs (or whatever their values

were at the time).” Their response? I believe the cliche is “deafening

silence.” Its true. The ADA recommends high-carbohydrate diets to be

compensated for with medication. Sugar is okay as long as you “cover with

insulin.”That was really their recommendation, at least in 2010: “Sucrose-

containing foods can be substituted for other carbohydrates in the meal

plan or, if added to the meal plan, covered with insulin or other glucose

lowering medications.”In 2013 they finally, and quite quietly, dropped this

bizarre and ultimately harmful recommendation, but it remains a testa¬

ment to their willingness to recommend a dietary treatment that will make

152

Nutrition in Crisis

things worse so that more drugs can be taken, or more accurately, because

more drugs can be taken.

The folks at the table were not endocrinologists and they were of my

generation. They grew up knowing that you don’t give carbohydrates to

people with diabetes. Simple. After all, it will increase their blood sugar,

which is precisely what you are trying to prevent in those who have the

disease. Because it was not their medical specialization, the clinicians were

unaware that recommendations had evolved at official agencies, or even

that it is now politically incorrect to refer to patients as “diabetics” (unless

you yourself are a person with diabetes). They believed in the old common-

sense idea that because diabetes was a disease of carbohydrate intolerance,

cutting out carbohydrate should be the first line of attack—and they prob¬

ably knew that it worked. Why wouldn’t it work?

Diabetes: Type 1 and Type 2

Diabetes is a disease—several diseases, really—of carbohydrate intolerance.

In type 1 the intolerance is due to the inability of the pancreas to produce

insulin in response to carbohydrate, and in type 2 it is due to poor cellular

response to the insulin that is produced, accompanied by deterioration of

the insulin-producing cells of the pancreas. The most salient symptom of

diabetes (and a major contributor to the pathology) is high blood sugar,

which, not surprisingly, is most effectively treated by reducing dietary

carbohydrates. The common clinical measurements are: (1) fasting blood

glucose (sometimes written FBG), usually given in units of milligrams per

deciliter (mg/dL) and normal considered to be around 100 mg/dL; (2) an

oral glucose tolerance test (sometimes abbreviated OGTT), the response

in the blood to a dose of glucose; and (3) the percent of hemoglobin Ale

(HbAlc), a modified form of hemoglobin (that has reacted with blood

glucose) that is a measure of the cumulative effect of the high blood sugar.

Normal levels for the latter are about 5 percent, but people with diabetes

can have values of 20 percent. In type 1, patients are required to inject insu¬

lin, whereas people with type 2 are advised to improve their lifestyle “with

diet and exercise.” If those with type 2 cannot lower their blood sugar with

diet—the diet that their doctor recommends is unlikely to adequately do

so—they will also require insulin or other drugs for reduction of glucose,

Diabetes

153

of which there are many. The insulin level of 150 units per day cited in the

introductory quotation to this chapter is a very high dose. Coming off such

a high dose is a major accomplishment, suggesting that Eric Westmans

diet might be better than the standard, and of course, reduction in medica¬

tion is considered improvement in all the diseases that I know of.

Why Diabetes?

Diabetes stands at the forefront of the nutrition problem because it’s

so clearly linked to control of metabolism and the glucose-insulin axis.

Correspondingly, it shows exactly what carbohydrate restriction can do

for you. Diabetes is, in a real sense, the extreme version of the nutritional

problem in obesity and possibly heart disease. Although Atkins or other

low-carbohydrate diets are used as a therapy for overweight or obesity, they

are not just about weight loss. In fact, they are not even primarily about

weight loss, despite the fact that they are more frequently used by people

who are overweight than by those with diabetes. Many diets work for weight

loss, and calorie restriction may or may not be part of the reason that a

low-carbohydrate diet makes you lose weight, but to bring diabetes under

control, we know that you have to reduce carbohydrate regardless of whether

or not calories go down, and regardless of whether or not you lose weight.

The intuitive idea that people with diabetes should not consume much

sugar or starch is a good principle. Nothing in the science contradicts this.

However effective a diet is for treatment, nobody really knows what is

required for prevention because it is often a kind of hidden disease. Early

symptoms might be low-level, such as simple irritability and fatigue, and

type 2, in particular, can have a very slow onset. As a treatment, however,

a low-carbohydrate diet is better than drugs for most people. This is

knowledge we cannot afford to ignore, considering how terrible a disease

diabetes is: It is the major cause of acquired blindness and the major cause

of amputations after accidents.

Where Does It Come From?

In chapter 5,1 described Claude Bernard’s revolutionary discovery of the

liver’s control of blood glucose and its regulation of glycogen turnover and

154

Nutrition in Crisis

gluconeogenesis (GNG). Bernard voiced astonishment at finding sugar

in a dog that hadn’t consumed any sugar. Historians suspect, however,

that writing about it later he might have exaggerated how much the

original observations really took him by surprise. He wouldn’t be the first

and undoubtedly will not be the last scientist to revise the history of his

discoveries for dramatic effect. It makes for a great story to have the answer

fall from heaven, especially if you can describe the intervention of your

prepared mind, as Pasteur put it.

It is likely that Claude Bernard had suspected for a long time that

animals could make their own sugar. He guessed that it could be made

from fat—which was not true—or from protein—which was true. One clue

that would have led him in this direction was his observation that people

with diabetes had more glucose in their blood than could be accounted for

by dietary intake alone. He must have suspected that those people with

diabetes were making their own sugar. In fact, gluconeogenesis goes on all

the time in the body, whether or not you have diabetes. GNG is not, as the

textbooks sometimes imply, a last-ditch resource during starvation, after

glycogen is depleted.

GNG goes on all the time. As described in chapter 5, when you wake

up in the morning, half of your blood glucose might come from GNG.

The difference for people with diabetes is that they have lost the ability

to turn it off at the right time. This is due to the reduction or absence of

insulin, as in type 1, or the poor response to the insulin that occurs in

type 2. In healthy individuals, the presence of glucose coming in from

the diet causes insulin secretion, which then inhibits gluconeogenesis and

glycogen breakdown. Many people think that the high blood sugar seen

in diabetes is due to a failure in clearance because the cells cannot take

up the glucose in the blood for fuel. Even the textbooks make this claim.

While the number of GLUT4 receptors in people with diabetes does

not increase in response to dietary glucose as much as it does in healthy

people, it seems that this is not the major cause of hyperglycemia—people

with diabetes still have enough of these glucose-transporting receptors

under most conditions.

The major problem, in fact, appears to be the persistence of glucose

production from the liver. Under normal conditions, release of glucose

from liver glycogen and gluconeogenesis are both regulated by insulin and

Diabetes

155

glucagon. As glucose goes up, insulin goes up and glucagon goes down.

In response, the liver stops producing additional glucose. A key feature of

diabetes is the loss of this stimulus-response control over glucose produc¬

tion. With that in mind, it does not make sense to recommend dietary

glucose to people with diabetes, as it will simply sit on top of the unregu¬

lated level of built-in production.

Can You Treat It?

Dietary carbohydrate restriction has been a therapy for diabetes since

before the discovery of insulin. 2 It is not a new idea that blood glucose

control has positive effects on all of the downstream effects of the

disease, including lipid markers for, and incidence of, cardiovascular

disease. For many diabetes sufferers, dietary carbohydrate restriction is

all that is needed to improve or eliminate symptoms. Normal glucose

control, stable weight, and normal lipids can be attained, and the

benefits persist as long as carbohydrate intake is low. It then becomes a

semantic question whether someone can be called “cured” if they have

to stay on a prescribed diet indefinitely. The real question is whether a

low-carbohydrate diet, or the high-carbohydrate diet recommended by

health agencies, the one that was associated with onset of the diabetes,

is the more extreme.

A key issue is the established association between body weight and type

2 diabetes, and whether the association is causal. Studies by researchers at

the University of Minnesota provide strong evidence against the idea that

there is a causal link between the two. Nuttall and Gannon have produced

a series of well-designed, well-controlled experiments demonstrating the

value of carbohydrate restriction in treating diabetes under conditions

where no weight is lost. They found that carbohydrate restriction has a

consistent and dramatic ability to reduce high blood glucose, even when

there is no change in body mass. It is not easy to lose weight, so the abil¬

ity to treat diabetes without concern for weight loss provides an obvious

advantage. From a theoretical standpoint, Nuttall and Gannon’s studies

support the idea that both obesity and diabetes are reflections of a central

cause—disruptions in the glucose-insulin axis, most likely. In other words,

obesity and diabetes might stand in a parallel, rather than serial, relation to

156

Nutrition in Crisis

each other. The most obvious evidence to support this idea is the fact that

there are people with diabetes, both types 1 and 2, who are not fat, and of

course, most fat people do not have diabetes.

The obstinate refusal of endocrinologists and diabetes educators to face

these ideas, and hardest to understand, their reluctance to face Nuttall

and Gannons experimental results, remains a major obstacle in dealing

with the current epidemic. Almost every statement, whether from health

agencies or individual experts, continues to emphasize weight loss as the

prime goal of diabetes treatment, and endocrinologists continue to make

this their recommendation. The problem is that the scientific evidence on

the subject is just not on their radar. Why not? Endocrinologists already

have to retain so much medical knowledge that its not entirely surprising

that they dont know much about nutrition. There is no excuse, however,

for the endocrinologists who act as if they know nutrition but refuse to

consider low-carbohydrate diets. I doubt that they love their patients any

less, but perhaps they love hating Dr. Atkins more.

Low-GI versus

Real Low-Carbohydrate

If low-GI isgood y how about no-GI?

—Eric Westman

In 2008, David Jenkins compared a diet high in cereal with a low-

glycemic index diet. 3 As I explained in chapter 2, the glycemic index

is a measure of the actual effect of dietary glucose on blood glucose.

Pioneered by Jenkins and coworkers, a low-GI diet is based on the same

rationale as a low-carbohydrate diet: that glycemic and insulin fluctua¬

tions pose a metabolic risk. GI emphasizes “the type of carbohydrate,”

thus offering a politically correct form of low-carbohydrate diet. As

stated in the 2008 study:

We selected a high cereal fiber diet treatment for its suggested health

benefits for the comparison so that the potential value of carbohy¬

drate foods could be emphasized equally for both high cereal fiber

and low—glycemic index interventions. (Emphasis added)

Diabetes

157

After the completion of the twenty-four-week studyjenkins concluded:

“In patients with type 2 diabetes, 6-month treatment with a low-glycemic

index diet resulted in moderately lower HbAlc levels compared with a

high-cereal fiber diet.” Coincidentally, on almost the same day that David

Jenkins’s study came out, Eric Westman’s group published a study that

compared a low-GI diet with a true low-carbohydrate diet. 4 The stud¬

ies were comparable in duration and number of subjects, and a direct

comparison (figure 10.1) shows that the true low-carbohydrate has much

greater benefits—hence the quotation from Dr. Westman at the head of

this section.

Figure 10.1 by itself constitutes the best evidence for a low-carbohydrate

diet as the “default diet,” the one to try first, for diabetes and metabolic

Weight

(kg)

HbAlc

[% x 10]

Glucose

[mg/dL]

Total-C

[mg/dL]

LDL

[mg/dL]

HDL

[mg/dL]

-30

-40

-50

-60

Jenkins High-Cereal Fiber Diet

Jenkins Low-Glycemic Index Diet

| Westman Low-Glycemic Index Diet

| Westman Low-Carbohydrate,

Ketogenic Diet

-70

Figure 10.1. Effect of high-cereal fiber or low-glycemic index diets on body weight and

hHbAlc. Data from D. J. Jenkins et al., "Effect of a Low-Glycemic Index or a High-Cereal Fiber

Diet on Type 2 Diabetes: A Randomized Trial "Journo! of the American Medical Association

300, no. 23 [2008]: 2742-2753; E. C. Westman et al., “The Effect of a Low-Carbohydrate,

Ketogenic Diet Versus a Low-Glycemic Index Diet on Glycemic Control in Type 2 Diabetes

Mellitus" Nutrition and Metabolism 5, no. 36 (2008).

158

Nutrition in Crisis

syndrome. There are hundreds of studies about all aspects of diabetes but

none contradict the benefits of carbohydrate reduction shown in figure

10.1. It is worth noting, however, that one additional benefit wasn’t

addressed by Westman’s study: Low-carbohydrate diet reliably reduces

the dependence on drugs. In William Yancy’s classic study 5 of twenty-one

patients on low-carbohydrate ketogenic diets, all but four reduced their

level of medication and seven patients discontinued medication altogether.

Reducing medication, in and of itself, is a huge advance. Is that it? That’s

it. The hundreds of papers published provide a variety of details about this

very complicated disease and its response to the numerous drugs that are

used as therapy, but the basic principles described here have never been

refuted in any fundamental way.

Diabetes is a central battleground in the new low-carbohydrate revolu¬

tion and it is likely to provide a major victory. Scientifically, the burden of

proof is on those who continue to claim that it is a good idea for people with

diabetes to consume any significant amount of carbohydrate. Impressive

as the comparison shown in figure 10.1 is, it has made little impact.

Establishment medicine has arbitrarily decided that long-term randomized

controlled trials (RCTs) are a kind of “gold standard.” Whatever their value

for particular types of experiments, this principle is used to ignore any other

type of study, especially if it challenges the party line. RCT studies can be

useful, but the people who insist on using them for diabetes sit on panels

that would never fund an RCT study if it included low-carbohydrate diets.

Moreover, an RCT is not the best thing for everything. It is one kind of

experiment, and all possibilities are not up for grabs in science. Since diabetes

is primarily about carbohydrate, there is no better treatment than reducing

carbohydrate. There is nothing in the outcome of low-carbohydrate trials of

whatever length to suggest harm from such a therapeutic regimen—only

benefit. Absolute dependence on arbitrary rules and “gold standards” is

what continues to cause the most harm.

What About Cardiovascular Disease?

The “concerns” about low-carbohydrate diets still revolve around the

imagined risk of cardiovascular disease from fat in the diet despite the

Diabetes

159

continued failure to show any such risk. Carbohydrate restriction actually

improves the usual markers of CVD—notably HDL (“good cholesterol”)

and triglycerides. Although there haven't yet been long-term trials to show

that carbohydrate reduction prevents CVD, there are plenty of long-term

trials on the effect of reducing fat, and they all fail to show any clinically

significant effect.

There is a strong association between diabetes and the incidence of

CVD, but the largely discredited, or at least greatly exaggerated, diet-heart

hypothesis—and its proscriptions against dietary fat—has caused us to

ignore the carbohydrate elephant in the room. It turns out, however, that

the best predictor of microvascular complications (blindness, amputa¬

tions) and, to a lesser extent, macrovascular complications (heart attack,

stroke) in patients with type 2 diabetes is hemoglobin Ale, which is under

the control of chronic dietary carbohydrate consumption. Data from

the United Kingdom Prevention of Diabetes Study (UKPDS) 6 provide

support for this idea. In some sense, increased risk of CVD for people

with diabetes is due simply to the diabetes itself. More precisely, the same

conditions that gave rise to the diabetic state—disruptions in the glucose-

insulin axis—increase the risk of vascular disease.

Summary

Diabetes is a disease of carbohydrate intolerance. Type 1 is character¬

ized by an inability to produce insulin in response to carbohydrate, and

in type 2, there is peripheral insulin resistance along with deterioration

of the beta cells of the pancreas. The most salient symptom of diabetes

(and a major contributor to the pathology) is high blood sugar, which,

not surprisingly, is most effectively treated by reducing dietary carbo¬

hydrates. The clinical measurements are: (1) fasting blood glucose; (2)

an oral glucose tolerance test, the response in the blood to a dose of

glucose; and (3) the percent of modified hemoglobin, hemoglobin Ale

(HbAlc). It is not hard to guess the best treatment for a disease char¬

acterized by poor response to ingested carbohydrates, but you do need

to know if the theory will hold up in practice. A personal story might

bring this into perspective.

160

Nutrition in Crisis

Wendy's Story

The Uses of Metabolic Adversity

By Wendy Pogozelski

Wendy Pogozelski is professor and chairman of biochem¬

istry at SUNY Genesee in upstate New York. One of

the people who has used low-carbohydrate diets to teach

metabolism , she developed type i diabetes as an adult . In

2012 , ASBMB Today, the house organ ofthe American

Society of Biochemistry and Molecular Biology* started a

series about challenges to biochemists, Wendys story was

their first publication in this series:

Blurred vision was the first sign that something was wrong. The

front row of the freshman chemistry class I was teaching looked

strangely fuzzy. Then, over the next few days, I was gripped by

an unquenchable thirst and was constandy fatigued. Seemingly

overnight I lost eight pounds. I recognized the symptoms of

diabetes, but I was young(ish), slim(ish) and an avid kickboxer.

Mine was not the typical diabetic profile.

Despite my suspicion that I was experiencing raging hyper¬

glycemia, the diagnosis—“You have diabetes 7 ’—was devastating.

It marked the beginning of a lifestyle that is an enormous

challenge. However, the journey has led me to an increased

understanding of biochemistry, has enhanced my teaching

and ultimately has cast me in a new role of helping others. It

turned out that I had developed latent autoimmune diabetes in

adults, or LADA, a subset of type 1 diabetes. LADA is due to

an autoimmune reaction to pancreatic glutamate decarboxylase,

or GAD65. While LADA has a slower onset than classic type

1, formerly known as juvenile diabetes, the two diseases follow

a similar course and require injections of insulin.

Diabetes

161

Fortunately, I felt equipped to manage my condition. I teach

metabolism to undergraduates using an approach that empha¬

sizes insulin-dependent pathways as a unifying theme and one

that offers an everyday context. I knew that carbohydrates,

whether whole grain or highly processed, could raise my blood

glucose to dangerous levels, so my response to the diagnosis was

to reduce greatly carbohydrates in my diet. In addition, I was

careful to monitor my blood sugar levels and insulin doses. Hie

result was that my hemoglobin Ale (glycosylated hemoglobin,

a measure of blood sugar control) was 5.4 percent, within the

normal 4 percent to 5 percent range. My doctor said that I was

his “best patient ever 7 * and that I was achieving the blood sugars

of a nondiabetic person.

Despite satisfaction with my glycemic control, my physician

wanted me to see a dietitian. To my surprise, the dietitian was

appalled by my diet. She said, “You have to eat a minimum of

130 grams of carbohydrates a day/T protested, but she recruited

the rest of the medical team to endorse her position: “We all say

you have to eat more earbs.The American Diabetes Association

gives us these guidelines/ 5 One member of the team said, “I

want you to eat chocolate. I want you to enjoy life.”

As someone raised to be cooperative, and because I found it

easy to embrace medical advice to eat chocolate, I agreed to eat

more carbohydrates. The result was that my HbAlc rose above

7 percent. My blood sugar levels were frequently in the 200 to

300 mg/dL range (far above the normal level of about 85 mg/

dL), even when I supplemented with extra insulin. My former

dose of seven units of insulin per day increased to 30 units per

day. The loss of control was immensely frustrating. My physi¬

cian attributed my initial success to what is called the diabetes

honeymoon. Often, when someone first begins taking insulin,

there is a short-lived period during which beta cells seem to

recover a bit and secrete insulin. Regardless, it was clear that the

162

Nutrition in Crisis

dietitians approach was not yielding the success I desired. I felt

confused and uncertain as to what to do.

I decided to investigate for myself what my best diet should

be. I studied the literature, I sought out researchers and physi¬

cians, and I attended countless metabolism-related talks. In

addition, I connected with hundreds of people with diabetes.

The most important contribution to my achieving clarity,

however, was evaluating literature based on a molecular under¬

standing of how metabolism works.

In my quest for answers, I found to my surprise that many

dietitians and physicians were unable to explain the basis for the

dietary recommendations they endorsed. Some did, however,

express a desire for a better understanding or review of what

they'd once learned. And in the general public, I encountered

scores of diabetics and nondiabetics who also wanted tools to

make sense of conflicting nutritional information.

I began to use what I had learned not only to expand and

improve my teaching and research but also to step into the role

of a nutrition explainer. First I was determined to see that none

of my students would lack understanding of processes such as

gluconeogenesis and the many pathways affected by insulin. I

created new lecture topics and problem sets based on diabetes

and nutrition applications. My students responded positively

and appreciatively. There was a palpable increase in attention

in class.

Students came to my office to chat about things that they had

read. My class evaluations praised the use of nutritional context

and often said, a This material could have been rather dull withi

out 'all these great applications.” I even heard (frequently) “I

love metabolism!”

Beyond my student population, I engaged a world of

bloggers, physicians and other people with diabetes, many of

whom were eager to understand more deeply how things work

Diabetes

163

metabolically. I now find myself being interviewed, quoted

in papers, and invited to speak to groups of people, includ¬

ing physicians, who want to deepen their understanding of

metabolic pathways. I am asked to share my nutrition-based

teaching applications with other professors and with textbook

publishers. In these efforts, I try to avoid dispensing nutritional

advice; instead, I attempt to show how nutrient composition

affects metabolic pathways so that my audience feels better able

to evaluate nutritional recommendations.

Five years later, diabetes is still an immense mental and

physical challenge, but I am grateful for the insight and tools

that my education and training have provided me. Most

importantly, if I am able to further the use of molecular

science to help others find optimal dietary strategies, and if I

can help the next generation, then my adversity will have had

a positive outcome.

Epilogue

The story in ASBMB Today was written for a series on how

scientists overcome hard times rather than as a treatise on how

to manage diabetes. However, the essay was read by many folks

who were interested from a standpoint of their own health

concerns. I received many requests for an update, as people

wondered if, after my foray into inclusion of carbohydrates, I

returned to my low-carbohydrate style of eating. The answer is

a resounding “yes.” (With one small caveat, as you 11 see below.)

As I studied the scientific literature, I became more and more

convinced that my absolute primary concern should be keeping

my blood sugar levels as close to normal as possible. I saw that

I achieve the flattest blood sugars when I keep my carbohydrate

low and my insulin low. Dr. Richard Bernstein, a type 1 diabetic

and engineer-turned-physician who pioneered the use of

glucose meters, calls this principle “the law of small numbers.”

164

Nutrition in Crisis

Carbohydrate restriction results in minimization of errors.

Hyperglycemia and hypoglycemia come from mis-estimation

of carbohydrate amount, rate of carbohydrate absorption and

insulin absorption and activity. These factors are nearly impos¬

sible to predict accurately. Many people who use large amounts

of insulin to cover large amounts of carbohydrate in their diet

frequently find themselves in dangerous situations (passing out,

etc.) when the peak activity of the insulin occurs earlier than

the peak absorption of the dietary carbohydrate. I have never

passed out and my health care team has been astounded by my

lack of hypoglycemic episodes. Hyperglycemia also needs to be

avoided though. It is these high blood sugars that correlate with

the long-term deleterious effects of diabetes. Observing the

diabetic amputees at the endocrinologists office and watching

people painfully shuffle into the dialysis center has been strong

motivation to avoid these high blood sugars.

What is the evidence for the effectiveness of a low-carb

dietary approach with me? I was used as a test subject for

“sensor"’ technology. This device consists of a needle, sensor

and a transmitter worn on the body and it communicates with

an insulin pump. When my results were printed out after the

experiment, my health care team was astonished at how level

my glucose readings were. I was the last appointment of the day

and all the workers gathered around to admire the printout of

the “beautiful” blood sugars and ask me how in the world I did

that. (This was at the office where the team had insisted that I

needed to eat carbohydrate. Yes, it was very satisfying.)

It is worth noting too how well having level blood sugar

makes me feel. When my blood sugar goes over 200 mg/dL,

I start to feel melodramatic and have an elevated emotional

response to even minor difficulties. Low blood sugar induces

glucagon and epinephrine hormone release, which result in

sweating and a feeling of panic as well as low energy Both

Diabetes

165

high and low blood sugar make my brain sluggish. Also, as a

headache-prone person for much of my life, I found that my

formerly frequent headaches nearly disappeared when I began

to practice carbohydrate restriction.

When I keep my meal at less than 10 g of carbohydrate,

my blood sugars rise by no more than 40 mg/dL; sometimes

they rise as little as 10 mg/dL, but regardless, they return to

normal within two hours—like a nondiabetic, I find that too

much protein will elevate my blood sugar so I keep that amount

moderate but I don’t consciously restrict it, or my fat.

Despite my knowledge that carbohydrate restriction is best

for me, I sometimes have deviated from this plan of attack.

Particularly, since I became a mom, there has been a lot more

carbohydrate in the kitchen tempting me. Couple that with

the chronic exhaustion of motherhood and you’ve got a situa¬

tion that makes reaching for carbohydrate much more likely. I

rediscovered that M&Ms- (used effectively for potty-training a

toddler) are delicious, and my plans to eat no more than three

have consistently been shown to be no match for whatever else

is at work. (My theory is that there is an evil force activated

when one eats M&Ms and he is only appeased when the bag

is empty'.) Every time I have indulged, however, I have always

concluded “It wasn’t worth it” when I saw my high blood sugars

or had to compensate for my over-estimation of my insulin.

I also decided that I really, really like dark chocolate and

given that it is full of antioxidants, I do allow myself some. I

save my carbs for it. The 85% variety I eat has 4 g/square and

is much more satisfying than milk chocolate, so it’s possible to

enjoy without over-indulging. There are days too when I decide

that it’s a “Garb Day.” For example, one very' hot day in the

summer warrants my once-a-year small ice cream or custard

cone. Or if some kind person makes me a birthday cake I’ll eat a

bit. I call these excursions “experimental error.”I am not perfect,

166

Nutrition in Crisis

but I try to keep my carbohydrate intake under 50 g/day and

ideally around 30 g/day. For me, the lower my carbohydrate, the

far better my control.

Wendy Knapp Pogozelski earned a BS from Chatham University and

a PhD in chemistry from Johns Hopkins University under the direction

of Thomas Tullius She spent two years as an Office of Naval Research

postdoctoral fellow working at various sites in radiation biology. She is a

professor of chemistry at the State University of New York [SUNY) College

at Geneseo, where she has been since 1996 She teaches biochemistry,

emphasizing medical and nutrition-based applications. Her current

research focuses on radiation effects on mitochondrial function and

mitochondrial DNA as well as on understanding how dietary strategies

affect biochemical pathways

-CHAPTER 11-

Metabolic Syndrome

The Big Pitch

T hrough the last two chapters, we’ve learned that diabetes repre¬

sents the most clear-cut example of how the glucose-insulin axis

affects health. Diabetes rests at the center of nutritional thinking,

whether or not you are a patient yourself. If you are primarily interested in

weight loss or cardiovascular disease (CVD)—or even if your concern is

general good health—insulin metabolism is always in the foreground. The

focal point of so many of the issues we’ve discussed is metabolic syndrome

(MetS). The idea is generally credited to Gerald Reaven, an endocrinolo¬

gist who died in 2018 at age eighty-nine. His original observation, which

sparked the idea of MetS, was that overweight, high blood glucose, high

blood pressure, and the lipid markers assumed to indicate cardiovascu¬

lar risk commonly appeared together in the same patients (table 11.1). 1

MetS is not a disease but rather a collection of physiological markers that

represent risk of disease. The separate components of the syndrome are not

just superficially related, either. Reaven’s insight was to suggest that the

markers arose from some common central cause, which he, and most of us,

see as disruption in the glucose-insulin axis.

The concept has now been extended to include several other physiologic

markers, including inflammation and LDL particle size, that also seem

to be tied together. Insulin is credited with control of the syndrome, and

fittingly, several years before his death, Reaven insisted that MetS should

be called “insulin-resistance syndrome.” 2

The importance of metabolic syndrome came through to me a few years

ago. I was listening to a presentation at a seminar on metabolic syndrome

and its underlying cell biology. I don’t remember the details of the presen¬

tation, but I thought it was quite good. The speaker was also a doctor—in

168

Nutrition in Crisis

Table 11.1. National Cholesterol Education Program Adult Treatment Panel III Definition

of Metabolic Syndrome-Subjects have three of the following criteria

Men Women

Abdominal obesity,

waist circumference

> 40 inches

> 35 inches

Hypertriglyceridemia

> 150 mg/dL

>150 mg/dL

Low HDL-cholesterol

< 40 mg/dL

< 50 mg/dL

High blood pressure

> 130/85 mm Hg

High fasting glucose

> 110 mg/dL

Source NCEP Expert Panel on Detection, Evaluation, and Treatment of High Blood

Cholesterol in Adults (Bethesda, MD National Institutes of Health, 2001]

fact, he was Mike Huckabee’s doctor. After the seminar, I asked him about

low-carbohydrate diets and he unexpectedly went ballistic. “Go to the

Atkins website. You can eat all the bacon you want. That's what it says.”

I was somewhat taken aback. Did I say Atkins? I didn't really know what

to say. I noticed that he still had on the screen his last slide showing the

criteria for metabolic syndrome (something like table 11.1).

Pointing to the screen, I said, “You know, all of those markers are exactly

the things that are improved by low-carb diets.” He said, “Well, they're also

improved by low-calorie diets,” which is simply not true. Low-fat diets, or at

least high-carbohydrate diets, will not improve triglycerides and other mark¬

ers, and will more likely make them worse. Plus the level of blood glucose is

determined by carbohydrate in the diet. Everybody agrees on that.

The incident stuck in my head. A little later that day it occurred to me

that I might have said something smart. The features of MetS were known

to be improved by carbohydrate restriction, but this was not generally stated

explicitly, or maybe its full import was not realized. Turning it around in

your mind, you might say that the response to a low-carbohydrate diet

could actually be the essential feature of metabolic syndrome, a kind of

operational definition.

The existence of a syndrome—that is, the simultaneous appearance of

the markers—is no longer in question. It is also widely accepted that the

syndrome might be a reflection of insulin resistance. Yet the full impact of

MetS has not yet been appreciated. Its clinical importance continues to be

Metabolic Syndrome

169

questioned by some critics who claim that MetS is not a useful idea. These

critics hold that saying your patient has the markers of MetS provides no

more information than the primary observation that the markers do in fact

often show up together. What they mean is that, whatever the underlying

causes, the only way to treat markers A, B, and C (for example, obesity,

diabetes, and high cholesterol) is with a drug for A, a drug for B, and a

drug for C. In other words, they don’t have a single drug for MetS—only

for each of the individual markers.

The fact that A, B, and C can all be treated with a single interven¬

tion, namely, a low-carbohydrate diet, suggests that all the markers do in

fact arise from a common cause: a disruption in the glucose—insulin axis,

roughly described as insulin resistance. Low-carbohydrate diets could

thus provide a working definition of the syndrome. If your patient had

the markers traditionally associated with MetS and they improved with

a low-carbohydrate diet, it would confirm that the patient did indeed

have metabolic syndrome. I knew this was a good idea because other

people had brought it up before. I had even written a paper myself titled

“Metabolic Syndrome and Low-Carbohydrate Ketogenic Diets in the

Medical School Biochemistry Curriculum,” but I hadn’t really seen the

impact. There is a step in the evolution of ideas when you realize that a

comment you made in passing has important implications and has to be

restated as a law.

If all the markers of metabolic syndrome could be improved by a low-

carbohydrate diet, possibly even in the absence of calorie restriction, than

what would that mean for the millions of people facing the risk predicted

by these markers? What would it mean for the drugs that treat each indi¬

vidual condition but ignore the root cause? Lastly, what would it mean for

those national authorities on health that were, and are still, recommend¬

ing a low-fat, high-carbohydrate diet? If the experimental data are there,

would the nutritional establishment embrace them, even though they

contain the words “low-carbohydrate”? It was 2005 and we really thought

that progress could be made. Talk about the naivete of youth.

Jeff Volek and I went through the literature and tabulated the responses

to low-carbohydrate and low-fat diets with respect to the markers of

MetS. The results, which we published in Nutrition and Metabolism? were

as we expected (figure 11.1). I was the editor of Nutrition and Metabolism

170

Nutrition in Crisis

at the time and thought that the paper would improve the standing of

our journal. We recognized that we were dealing with modern science,

where people don’t even have time to read an abstract, so we put the

whole story in the title: “Carbohydrate Restriction Improves the Features

of Metabolic Syndrome. Metabolic Syndrome May Be Defined by the

Response to Carbohydrate Restriction ”The data showed that except for a

couple of measurements (insulin, fasting glucose), the markers of MetS are

improved by a low-carbohydrate diet, sometimes dramatically. Probably

the best indicator of CVD risk based on commonly measured parameters

is the ratio of triglycerides :HDL where the reduction is typically three to

ten times greater in carbohydrate reduction.

Volek’s Test of the Theory

In chapter 7, we looked at the experiments in Jeff Volek’s laboratory show¬

ing that saturated fat in the blood was reduced by a low-carbohydrate diet

with high saturated fat, as compared to low-fat diet with low saturated

fat. The results on control of plasma saturated fat are critical because the

presence of dietary saturated fat is still held up as an objection against

low-carbohydrate diets. It was also the magnitude of the effect in Voleks

experiment that was surprising. The total SFA fraction in the low-carbo-

hydrate group was reduced by more than half, and this reduction was more

than three times the average change in the low-fat group.

This study had a larger overall significance, however. The real power

of Voleks experiments was that the participants all fit the definition of

metabolic syndrome, and that a wide variety of lipid and physiologic

parameters were measured. Figure 11.1 shows that everything got better:

HDL, insulin, leptin, and most dramatically, triglycerides.

The Pitch

This chapter reinforces the big pitch: If the coincidence of metabolic

markers that defines MetS indicates a unified mechanism, if seemingly

different physiologic effects—overweight, high blood pressure, athero¬

genic dyslipidemia, high triglycerides, low HDL, high blood glucose, high

insulin—are all a reflection of a common underlying stimulus (proposed to

Metabolic Syndrome

171

Body Abdominal

Mass Fat TG

ApoB/

Small

Total

TG/HDL ApoA-1

LDL Glucose Insulin

SFA

Figure 11.1. Summary of responses of forty people with metabolic syndrome to a very

low-carbohydrate ketogenic diet [VLCKD] or a low-fat diet (LFD). Data from J. S. Volek

et al., "Carbohydrate Restriction Has a More Favorable Impact on the Metabolic Syndrome

than a Low Fat Diet" Lipids 44, no. 4 (2009]: 297-309.

be disruption in insulin metabolism), then if we can treat any one of those

features, we can treat them all.

Nothing is better than low-carbohydrate for weight loss, but other

diets do work. Its harder for women and its harder as you get older, but

there are lots of ways to get thinner. We don’t really have the answer on

C VD. We know a lot of things that might be relevant but we don’t really

know the fundamental cause. However, what we do know with some

certainty is about diabetes: Cutting back on carbohydrate is the most

effective treatment. For many it is a virtual cure. In the long term, it is

better than drugs. So if MetS is really a clue to an underlying mechanism

for all of the disparate markers, then effectively treating hyperglycemia

will improve all of the features of MetS—that is to say, we have a

prescription for general health.

172

Nutrition in Crisis

The Head-and-Shoulders Effect

In addition to metabolic effects, a low-carbohydrate diet -will

cure irritable bowel syndrome and related disorders in many

people. (Cancer is waiting in the wings, too.) In some sense, the

problem with convincing people of the benefits of a reduced

carbohydrate strategy is that it appears to be good for every¬

thing, good for what ails you. You can sound like a hard-sell

pitchman. I call this the Head-and-Shoulders effect. I don’t

know whether it is true but a rep from Procter 6c Gamble once

told me that when they first brought out the shampoo of that

name, they advertised that it would cure your dandruff in three

days. What their tests actually showed was that it would cure

it in one day, but they didn’t think anybody would believe that.

It is probably not that low-carbohydrate is so good but that

high-carbohydrate is so bad.

PART 4

The Mess in

Nutritional Science

-CHAPTER 12-

The Medical Literature

A Guide to Flawed Studies

N utrition is in crisis. Almost every day a new study shows that

you are at risk for diabetes, cardiovascular disease, or all-cause

mortality brought on by a newly identified toxin that turns out

to be something that you just had for lunch. It is not clear that any of these

studies are subject to serious critical peer review, and for the curious, blog¬

gers usually do a good job of dismembering them. The continuous cycle

of weak studies and their deconstruction goes beyond mere time wasting.

People are hurt because bad recommendations are left out there, even when

research shows that they are inappropriate and, in the process, science

takes a big hit. The editors and reviewers of technical and medical journals

who conduct peer review are supposed to act as the gatekeepers of scientific

evidence, but the journals continue to publish papers showing very weak

associations, and even some that are grossly misleading or contradictory.

The media, which might be expected to help our case, make it worse. It’s

not really their fault. A science reporter cannot reasonably have the time to

read the original study in detail, and must instead accept the conclusions

in the abstract, and so that message is transmitted through mass media.

When you do explain to the reporter how misleading these reports are,

and how people will be hurt, they are truly concerned and sympathetic,

but they don’t always have complete editorial control. In any case they

would like to help, but tomorrow they have to cover a story that may be

even worse. It is really hard for the consumer. This post from Facebook

probably tells the story: “So epically confused about diet. Everything I

read is contradictory on epic proportions. About the only consistencies are

low-sugar raw veggies and water. How in the world is a girl to sort it out,

other than try everything and see what works for me?”

176

Nutrition in Crisis

She went on to ask why it isn’t “possible to come up with a system

that takes inputs—body stats and genetic history—and outputs a general

reasonable diet to follow?” Chances are that the population at large is no

more comfortable than this woman on Facebook when it comes to matters

of nutrition. I wrote to her through private messages and reiterated the first

three rules: (1) If you are okay, you are okay; (2) if you want to lose weight,

don’t eat, and if you have to eat, don’t eat carbs; and (3) if you have diabetes

or metabolic syndrome, you have to try a low-carbohydrate diet first.

It is likely that many people wind up believing nothing at all, however,

and simply assume that everything is exaggerated, save for the most

ingrained popular notion that “maybe fat is bad and maybe I should not

put so much salt on my food.” Then there is the progression of articles on

raspberry-ketones, resveratrol, trans-fats, and methylglyoxal, each of which

will either kill you or save you, depending on whom you ask.

Most discouraging are the health agencies. The American Diabetes

Association (ADA) wants people with diabetes to consume a lot of carbo¬

hydrates. They keep saying that they don’t have a diet, and that they’re

not opposed to low-carbohydrate diets (for weight loss)—but instead they

stress “individualization” without any indication as to which individuals

benefit from which intervention. Despite the disclaimers and ever-shifting

language, there is no doubt that the ADA is perceived as opposing low-

carbohydrate—and it seems clear they are the only ones responsible for

that perception.

The evidence that weight loss is not required for improvement in diabe¬

tes, from the work of Nuttall and Gannon, 1 for example, is not mentioned

by the ADA. They know about that evidence. I’m sure of it because I have

told members of the committee personally, and they should already be

aware because some of the work was funded by the ADA itself. The ADA

guidelines do not cite important scientific work showing that weight loss

is not required for improvement in diabetes. People who are not scientists

ask me, understandably incredulous, “Can you do that? Are you allowed to

make recommendations without citing other people’s work?”

There is a daily progression of sweeping statements that go way beyond

the published data. At the same time there remains an inability or an unwill¬

ingness to zero in on real factors. The low-carbohydrate diet has attained

the status of the name of God in Hebrew: It must never be said out loud.

The Medical Literature

177

Part of the problem is that the literature, especially major medical

publications, is still predominantly subscription-based. Most people

cannot access the information, so the results are then fed downstream to

the media, who take anything they are fed at face value, and then pass it on

further to the general public.

The rigid dogma of the literature has reached Galilean proportions.

Fructose and sugar are bad (unless you try to lump them in with all carbo¬

hydrates). If you want your paper on fructose to be published, begin with:

“Because of the deleterious effects of dietary fructose, we hypothesized

that . . .” Never start with: “Whether dietary fructose has a deleterious

effect. . .” Our paper on fructose was published with “whether ..as the

opening sentence, 2 but only after a hard-fought rebuttal of reviewers’ criti¬

cisms that turned out to be fifteen pages long. Worse, if you even mention

low-carbohydrate, you are guaranteed real grief. When the Journal of

the American Medical Association published George Bray’s “calorie-is-a-

calorie” 3 and I pointed out that the study more accurately supported the

importance of carbohydrate as a controlling variable, the editor refused to

publish my letter. Thankfully, blogs have performed a valuable service by

providing an alternative point of view, but if unreliability is a problem in

the scientific literature, that problem is multiplied exponentially in internet

sources. In the end, consumers might feel that they are pretty much on

their own.

It does take some confidence, especially for the layperson, to feel

that their intuitive understanding is correct: that the difference between

white rice and brown rice is so small that it really doesn’t matter what

Harvard’s computer says. Most researchers are very much disinclined to

get into a shooting match, or worse, whistle-blowing. The long blue line

is a strong force in repressing investigation, not because the authorities

think corruption is okay, but because scandal reflects badly on everybody.

Whistle-blowing in this field is especially weird because sometimes the

transgressions are right out in the open.

Statistics: Death of the Medical Literature

Many scientists believe that if you do a good experiment, you don’t need

statistics. David Colquohon, a well-known neuroscientist and critic of

178

Nutrition in Crisis

The Golden Rule of Statistics

Here’s the Golden Rule for reading a scientific paper, from the

book PDQ Statistics by Norman and Streiner: “The important

point ... is that the onus is on the author to convey to the

reader an accurate impression of what the data look like, using

graphs or standard measures, before beginning the statistical

shenanigans. Any paper that doesn't do this should be viewed

from the outset with considerable suspicion.” 4

In other words, explain things clearly to the reader. There are

complicated ideas in science but the often quoted statement

(only once before in this book) from Einstein, that you should

make it simple but not too simple, is a reasonable demand for

you, the reader, to make of the scientific literature.

poor scientific method, agrees. Colquohon is the author of an excellent,

if technical, statistics book, which is now freely available online. 5 The

introduction to his book points out: “The snag, of course, is that doing

good experiments is difficult. Most people need all the help they can get

to prevent them making fools of themselves by claiming that their favorite

theory is substantiated by observations that do nothing of the sort.”

This kind of circumspection is, unfortunately, more common among

people who write the statistics books than those who use them. A good

statistics book will have an introduction that says something like “In

statistics, we try to put a number on our intuition.” In other words, it

is not really, by itself, a science. It is, or should be, a tool for the experi¬

menter's use, and like any tool, you have to know how to use it. Because

statistics offers so many tools, there is not always agreement on which

hammer goes with which nail. All of statistics is interpretation. The

problem is that many authors of papers in the medical literature allow

statistics to become their master rather than their servant: -Numbers

are plugged into a statistical program and the results are interpreted in

The Medical Literature

179

a cut-and-dried fashion. Statistical significance (that two sets of data

are not from the same population) is confused with clinical significance

(that differences are sufficiently large to have a biological effect). Misuse

of statistics is the subject of numerous papers and books, but this has

had little effect.

Statistical Shenanigans

versus Common Sense

How do you deal with the reports in the press that tell you that white

rice will give you diabetes but brown rice wont? The saving principle is

that, despite its subtleties and reliance on mathematics, biochemistry

uses the same basic rules of logic as daily life. So the first question you

have to ask is whether a given study makes sense: How is it possible

that white rice is so different from brown rice? A moment s thought

suggests that the major difference between the two is in the stuff that

isn’t even digested, the fiber. The nutritional establishment is always

pushing fiber—whole grain and all that—so you might want to hear

their case, but then there is the possibility that the entire fiber thing

itself is questionable or at least exaggerated—and what of the Asian

societies that are always invoked to tell us how much we need grain, yet

never ate brown rice. Never. So, it can be confusing, and common sense

makes you suspicious.

Habeas Corpus Datorum

Science is an extension of common sense, but that doesnt mean there

aren’t revolutionary ideas. You want to be suspicious of a revolutionary

idea if it violates common sense, but you can’t throw out the idea just for

that reason. The solution is simple, if not always easy to implement. If it is

a reasonable conclusion, you can cut the author some slack. If the idea is

far out, you need to see the data—all the data—not just the average of the

data or conclusions from the computer. My new grand principle of doing

science: habeas corpus datorum. Let’s see the body of the data. If the conclu¬

sion is nonintuitive and goes against previous work or common sense, then

the data must be strong and clearly presented.

180

Nutrition in Crisis

So, how should you read a scientific paper? I usually want to see the

pictures, or figures, first. It s not just about saving a thousand words, as the

saying goes. Its about presentation of the data, and its about the Golden Rule

of Statistics: “Convey to the reader an accurate impression of what the data

look like, using graphs or standard measures, before beginning the statistical

shenanigans.” On the topic of tables: Figures are so much better than tables

that a whole book, Medical Illuminations , was written about the idea. 6

Many of us write scientific papers in the same way we read them. We

make the figures first and then try to figure out what they say. The prin¬

ciple: A scientific paper is supposed to explain. I tell graduate students

that if you do an experiment and you don’t explain it well, it’s as if you

never did it at all. In teaching students how to present their work, I ask

them: “Describe what you are supposed to do in a scientific seminar or

other presentation.” Sometimes they begin to say something, but I usually

make it worse by adding: “No. In one word. What are you supposed to

do? One word.” Having reached an appropriate level of annoyance where

they will be relieved to hear the answer, I give it to them: “Teach.” You

want to explain things to your audience. The same is true of a scientific

paper. Again, the Golden Rule of Statistics: The onus is on the author.

Caveat Lector

Presenting a scientific paper is also a bit like selling something. Scientific

papers are rarely just data. They have an idea that they are trying to sell.

Teaching and selling are the two things you do in science. The reader has

to be an educated consumer, however, and as an educated consumer she or

he should be suspicious of overselling. One good indicator of overselling

is the use of value judgments as if they were scientific terms. “Healthy” (or

“healthful”) is not a scientific term.

If a study describes a diet as “healthy,” it is almost guaranteed to be a

flawed study. If we knew which diets were “healthy,” we wouldn’t have an

obesity epidemic. A good example is the paper by Appel on the DASH

diet, which concluded: “In the setting of a healthful diet, partial substi¬

tution of carbohydrate with either protein or monounsaturated fat can

further lower blood pressure, improve lipid levels, and reduce estimated

cardiovascular risk.” 7

The Medical Literature

181

It’s hard to know how healthful the original diet could have been if

removing carbohydrate improved everything. In addition, not only was

this about a “carbohydrate-rich diet used in the DASH trials” but it is

“currently advocated in several scientific reports.” Another red flag is when

the authors tell you how widely accepted their idea is.

Understatement is good. One of the more famous is from the classic

Watson and Crick paper of 1953, in which the two researchers proposed

the DNA double helix structure. They said, “It has not escaped our notice

that the specific pairing we have postulated immediately suggests a possible

copying mechanism for the genetic material.”

What Is Wrong with the Literature

and What Is Right

Let’s start with what’s right. There are many published papers in the

nutritional literature that are informative, creative, and generally conform

to the standards of good science. Naturally, like any scientific literature,

most papers are fairly routine. They are of specialized interest, or as

described in one of the choices on the checklist for referees who review

manuscripts—“of interest to other workers in the field.”

What’s wrong is not the mediocre papers but rather the surprising number

of really objectionable papers. The medical literature is full of papers border¬

ing on fraud, or at least, guilty of misrepresentation. There are many papers

that are full of fundamental errors and total lack of judgment in interpreta¬

tion. Worse, there is a possibility that somebody is going to get hurt because

of bad medical advice that follows from these misinterpretations.

Most work in most fields, more or less by definition, is mediocre. What

makes papers in medical nutrition different is their drastic claims about

saving hundreds of thousands of lives by scaling up a result that had hardly

any effect to begin with to the entire population.

It is difficult to face the fact that so much of the medical literature is

published by people who are not trained in science, which means they

don’t know the game, they haven’t seen much of it, and they wouldn’t know

good science if they saw it. There is no real conceptual training in science.

You can learn techniques, but it’s not about cyclotrons, it s about ideas.

There is no reason why a physician can’t do real scientific thinking, but at

182

Nutrition in Crisis

the same time, there is no reason why he or she can. An MD degree is not

a guarantee of any expertise outside the practitioner’s area of specialty.

The irony is that the practice of medicine can be highly scientific.

Differential diagnosis and the experience in recommending the right drug

are the kinds of things that are part of scientific disciplines. The same

physician who will intuitively solve a medical mystery, though, will assume

that, for scientific research, things are different—that there are somehow

arbitrary rules and that brute-force application of statistics will tell you

whether what you did is true.

The next six chapters will detail the failures of nutritional literature,

ranging from the slightly inaccurate—“association does not imply causal¬

ity” (sometimes it does and sometimes it doesn’t)-—to the idiotic—“you

must do intention-to-treat” (if you assign subjects to take a drug and they

don’t take it, you have to include their data with those who did). I will also

cover “levels of evidence”: arbitrary rules that get incorporated into tables,

the top of which is always some kind of “gold standard.” The odd thing

about levels of scientific evidence is that nobody in any physical science

would recognize them. They are, in fact, the creation of people who are

trying to do science but wouldn’t know science if they saw it—fundamen¬

tally amateurs who have arbitrary rules along the lines of the apocryphal

story about Mozart:

A man comes to Mozart and wants to become a composer.

Mozart says that he has to study theory for a couple of years,

that he should study orchestration and become proficient at the

piano, and goes on like this. Finally the man says, “But you

wrote your first symphony when you were eight years old.”

Mozart says, “Yes, but I didn’t ask anybody.”

The Bottom Line

To understand research in nutrition, or really, any science, one has to be

prepared to question “the experts.” If a paper does not adhere to the Golden

Rule and is too quick to begin “the statistical shenanigans,” it “should be

viewed from the outset with considerable suspicion.”

The Medical Literature

183

The bottom line is that you have to expect real communication from

the authors of a scientific paper. The problem, for many people, lies in the

difficulty of believing that the best and the brightest are at fault. But it is

not hard to find examples of experts making mistakes. I will try to provide

you with all the help I can in the following chapters.

-CHAPTER 13-

Observational Studies,

Association, Causality

789 deaths were reported in Doll and Hills original cohort.

Thirty-six of these were attributed to lung cancer. When

these lung cancer deaths were counted in smokers versus non-

smokers, the correlation virtually sprang out: all thirty-six of

the deaths had occurred in smokers. The difference between the

two groups was so significant that Doll and Hill did not even

need to apply complex statistical metrics to discern it. The trial

designed to bring the most rigorous statistical analysis to the

cause of lung cancer barely required elementary mathematics

to prove his point.

—SlDDHARTHA MuKHERJEE,

The Emperor of All Maladies

S cientists don’t like philosophy of science. It is not just that pomp

ous phrases like hypothetico-deductive systems are such a turnoff;

it’s that we rarely recognize descriptions of science in philosophy

articles as accurate reflections of what we actually do. In the end, there is

no definition of science any more than there is a definition for music or for

literature. Because scientists have different styles, it is hard to generalize

about actual scientific behavior. Research is a human activity, and precisely

because it puts a premium on creativity, it defies categorization. As the

physicist Steven Weinberg put it, echoing Justice Stewart on pornography:

“There is no logical formula that establishes a sharp dividing line between

a beautiful explanatory theory and a mere list of data, but we know the

difference when we see it—we demand a simplicity and rigidity in our

principles before we are willing to take them seriously. 1

186

Nutrition in Crisis

We know, however, that what we see in the current state of nutrition is

not “a beautiful explanatory theory” or anything of the sort. This forces us

to consider what it is that makes nutritional medical literature so bad. If

we can identify some principles, maybe we can penetrate the mess and see

how it could be fixed.

One frequently stated principle is that “observational studies only

generate hypotheses.” There is the related principle that “association does

not imply causality,” usually cited in a backhanded way by those authors

who want you to believe that the association they found does in fact imply

causality. These two principles are not exactly right. They fail to recognize

that scientific experiments are not so easily wedged into categories like

“observational studies.” Principles like “observational studies only generate

hypotheses” are also widely invoked by bloggers and critics to discredit

the continuing stream of observational studies that make an association

between their favored targets—eggs, red meat, sugar-sweetened soda—and

prevalence of some metabolic disease or cancer. In most cases, the original

studies are getting what they deserve, but the bills of indictment are not

accurate, and it would be better not to cite absolute statements of scientific

principles. It is not simply that these studies are observational studies, but

rather that they are bad observational studies, and in many cases, the asso¬

ciations that they find are so weak that the study—if anything—constitutes

an argument against causality. On the assumption that good experimental

practice and interpretation could be roughly defined, I laid out a few prin¬

ciples that I thought were a better representation—if you can even make

such generalization—of what actually goes on in science:

* Observations generate hypotheses. Observational studies test hypoth¬

eses. Associations do not necessarily imply causality. In some sense, all

science is associations.

* Only mathematics is axiomatic (starts from absolute assumptions).

* If you notice that kids who eat a lot of candy seem to be fat, or even if you

notice that you yourself get fat eating candy, that is an observation. From

this observation, you might come up with the hypothesis that sugar

causes obesity. Thus, an observation generates hypotheses. A test of your

hypothesis would be to carry out an observational study. For example, you

might try to see if there is an association between sugar consumption and

Observational Studies, Association, Causality 187

incidence of obesity. There are different ways of doing this—the simplest

epidemiologic approach is simply to compare the history of the eating

behavior of individuals (insofar as you can get it) with how fat they are.

When you do this comparison you are testing your hypothesis.

For the final point on the list, you must remember that there are an

infinite number of other things—meat consumption, TV hours, distance

from a French bakery, grandfather’s waist circumference—that you could

have measured as an independent variable. Your hypothesis, however, is

that the variable is candy. What about all the others? Mike Eades, author

of the influential Protein Power, described falling asleep as a child by trying

to think of everything in the world. You just can’t test them all. As Einstein

put it, “Your theory determines the measurement you make.” If you found

associations with everything, would anything be causal?

Association Can Predict Causality

In fact, association can provide strong evidence for causation. If we didnt ask

about the extent to which the tacos we ate were the cause of our stomach

upset, we would not function well in our daily lives. Single observations might

generate a hypothesis or a guess that we can test. We can determine, for

example, whether a particular restaurant is good through continuous “test¬

ing.” Single observations generate hypotheses. Hypotheses, in turn, generate

observational studies, not the other way around. A correct statement is that

association does not necessarily imply causation. In some sense, all science is

observation and association. Even thermodynamics, the most mathematical

and absolute of sciences, rests on observation. We have never observed a true

perpetual motion machine, so we have generalized that observation into a

physical law, but as soon as somebody makes one that works, it s all over.

Biological mechanisms, or perhaps all scientific theories, are never

proved. By legal analogy: You cannot be found innocent, only not guilty.

That is why excluding a theory is stronger than showing consistency. The

grand epidemiological study of macronutrient intake the association of

what Americans ate in the last forty years and the incidence of diabetes and

obesity—shows that increased carbohydrate is associated with increased

calories, even under conditions where fruits and vegetables also go up

188

Nutrition in Crisis

and where fat, if anything, goes down. The data on dietary consumption

and disease in the whole population can be described as an observational

study, but it is strong because it gives support to a lack of causal effect of

decreased fat on positive outcome—it excludes a theory

Again, in science, finding a contradiction has greater impact than

merely finding a consistent result. The multitude of prospective experi¬

ments (where you pick the population first and see how people do on your

variable of interest) have shown in the past, and will undoubtedly continue

to show, the same lack of relation between fat intake and disease. But will

anybody give up on saturated fat? In a court of law, if you are found not

guilty of child abuse, people may still not let you move into their neighbor¬

hood. An association will tell you about causality if: (1) the association is

strong; (2) there is a plausible underlying mechanism; and (3) there is not

a more plausible explanation.

The often-cited coincidental correlation between cardiovascular disease

and number of TV sets does not imply causality because, although the

first principle of observational studies is observed, there is no logical direct

underlying mechanism. TV does not cause CVD. Interestingly, in the case

of CVD, where so many associations have been published, and so many

learned societies have told us how to prevent or cure the disease, there

remains little in the way of an agreed upon mechanism.

Re-Inventing the Wheel: Me and Bradford Hill

This chapter is a reworking of a blog post that I published in 2013. The post

included the principles I laid out at the top of the chapter for dealing with

the kind of observational studies that you see in the scientific literature. I

was speaking off the top of my head, trying to describe the logic that scien¬

tists use in interpreting data. It was an obvious description of what is done

in practice. I didn’t think it was particularly original, and again, I don’t think

that there are any hard and fast principles in science. When I described

what I had written to my colleague, Dr. Eugene Fine, his response was,

“Aren’t you re-inventing the wheel?” He meant that Bradford Hill, pretty

much the inventor of modern epidemiology, had already established these

and a couple of others as principles. In our conversation, Eugene cited The

Emperor of All Maladies? an outstanding book on the history of cancer. I

Observational Studies, Association, Causality 189

had, in fact, read Emperor on his recommendation. I remembered Bradford

Hill and the description of the evolution of the ideas of epidemiology,

population studies, and randomized controlled trials. The story is also told

in James LeFanu’s The Rise and Fall of Modern Medicine , 3 another captivat¬

ing history of medicine.

I thought of these as general philosophical ideas, rather than as absolute

scientific principles. Perhaps it is that we’re just used to it, but saying that

an association has to be very strong to imply causality is common sense,

and not in the same ballpark with the Pythagorean theorem. It’s some¬

thing that you might say over coffee or in response to somebody’s blog.

Being explicit about it turns out to be very important, but like much in

philosophy of science, it struck me as not of great intellectual import. It all

reminded me of learning, in grade school, that the Earl of Sandwich had

invented the sandwich. At which time I thought, “This is an invention?

Woody Allen thought the same thing years later and wrote the history

of the sandwich. He recorded the Earl’s early failures: “In 1741, he places

bread on bread with turkey on top. This fails. In 1745, he exhibits bread

with turkey on either side. Everyone rejects this except David Hume.”

In fact, Hill’s principles remain important even if seemingly obvious. The

pervasive violation of these principles in the medical literature constitutes

their real significance. The concept of the randomized controlled trial

(RCT)—randomly assigning people to a drug or behavior that you’re test¬

ing, or to a group that is the control—while obvious to us now, was hard

won. Likewise, proving that any particular environmental factor—diet,

smoking, pollution, or toxic chemicals—was the cause of a disease, and that

by reducing that factor, the disease could be prevented, turned out to be a

very hard sell, especially to physicians whose view of disease might have

been strongly colored by the idea of an infective agent, bacterium, or virus.

The Rise and Fall of Modern Medicine describes Bradford Hill’s two

important contributions 4 : He demonstrated that tuberculosis could

be cured by a combination of two drugs, streptomycin and PAS (para-

aminosalicylic acid), and even more importantly, he showed that tobacco

causes lung cancer. Hill was a professor of medical statistics at the London

School of Hygiene and Tropical Medicine, but was not formally trained in

statistics, and like many of us, thought of proper statistics simply as applied

common sense. Ironically, an early near-fatal case of tuberculosis prevented

190

Nutrition in Crisis

formal medical education. His first monumental accomplishment was, in

fact, to demonstrate how tuberculosis was cured by the streptomycin-PAS

combination. In 1941, Hill and his coworker Richard Doll undertook a

systematic investigation of the risk factors for lung cancer. His eventual

success was accompanied by a description of the principles that allow you

to say when association can be taken as causation.

Association and Causality: The Nine Criteria

Bradford Hill described the factors that might lead you to believe that an

association supports a causal role. Hill s criteria are still perfectly reasonable

today, though in the current medical literature, they are probably much

more widely practiced in the breach than the observance. The diligent

application of Hills principles to the medical literature would substantially

reduce the size of the literature and improve the quality of what remained:

1. STRENGTH. “First upon my list I would put the strength of the

association,” Hill wrote. This, of course, is exactly what is missing in the

continued epidemiological scare stories whose measures of relative risk are

so small. Hill describes:

Prospective inquiries into smoking have shown that the death rate

from cancer of the lung in cigarette smokers is nine to ten times

the rate in non-smokers and the rate in heavy cigarette smokers

is twenty to thirty times as great.... On the other hand the death

rate from coronary thrombosis in smokers is no more than twice,

possibly less, than the death rate in non-smokers. Though there is

good evidence to support causation it is surely much easier in this

case to think of some features of life that may go hand-in-hand

with smoking—features that might conceivably be the real under¬

lying cause or, at the least, an important contributor, whether it be

lack of exercise, nature of diet or other factors.

Hill expressed doubts about a relative risk of 2 or less. Criticized else¬

where in this book, relative risk (RR) is what it sounds like: the ratio of the

risks from two outcomes. Risk is the probability of an outcome; relative

risk is the ratio of the individual risks of two events. For example, if you

Observational Studies, Association, Causality

191

were to compare a group of factory workers in a chemical plant, say, to

the general population and you found that for every 1,000 workers, 26

developed cancer, then the probability, the risk of cancer is 26 out of 1,000

or 0.026 or 2.6 percent. If you found in the general population that there

were only 13 cases of cancer for every 1,000 people, then the risk for the

general population is 13 out of 1,000 or 0.013 or 1.3 percent. You can then

calculate relative risk as follows:

RR = (risk for workers) / (risk for population) = (26/1,000) / (13/1,000)=2:1

This might be considered evidence for environmental hazard, although as

Hill says, it would be stronger if the RR were 10. Still 2 is considered grounds

for taking the results seriously (or taking the factory owner to court). RR,

however, as its name suggests, is relative. It hides information. The RR would

still be 2 if there were 26 out of a million workers getting sick and 13 out of a

million in the general population. The absolute risk is the difference between

the probabilities. Absolute risk in this example is small (2.6 percent-1.3

percent = 1.3 percent). In other words, not even a full 2 percent more factory

workers got cancer than the general population. If you did take the factory

owner to court, the judge might ask for additional evidence.

This is what are we up against. Even within the limitations of rela¬

tive risk, the way in which the results are presented can significantly

affect a reader's perception. The abstract from a study on eating breakfast

found that: “Men who skipped breakfast had a 27% higher risk of CHD

compared with men who did not.” 5 The number 27 percent sounds a great

deal more impressive than the relative risk of 1.27, even though the latter

might paint a better picture of the limited effects. For every 227 people

with cancer, 100 will be regular breakfast eaters and 127 will have passed.

By Hills standards, or by common sense, a relative risk of 1.27 would not

make you force yourself to eat breakfast if you weren't hungry. Describing

the results as “27% higher risk,” on the other hand, might influence your

behavior more. Reporting relative risk so as to make the effect larger is the

single most prevalent mistake in the medical literature and is even worse

in the media that report on that literature. It must be noted, however, that

while a high relative risk is no guarantee of causality, a low relative risk is

definitely suspicious.

192

Nutrition in Crisis

2. CONSISTENCY- Hill listed the repetition of the results of studies,

under different circumstances, as a criterion for considering the extent to

which an association implied causality. We expect results to be reproduc¬

ible, and a weak association might gain some strength if the observation is

reproduced. Consistency of strong results, however, is what we want to see.

Criterion 2 is not, however, independent of criterion 1. Although Hill

himself did not mention this, it is of great importance. Consistently weak

associations do not generally add up to a strong association. If there is a

single practice in modern medicine that is completely out of whack with

respect to careful consideration of causality, it is the use of meta-analyses

where studies with marginal strengths are averaged to create a conclusion

that is stronger than the majority of its components. In fact, many meta¬

analyses include studies that have not shown any association at all; the

authors average them with a couple that have and then report the average

as significant. Averaging studies without significant outcomes and expect¬

ing to get an effect is as foolish as it sounds, but it is widely practiced.

3. SPECIFICITY. Hill was circumspect on this point, recognizing that

we should have an open mind on what causes what. On specificity in the

study of cancer and cigarettes, Hill "noted that the two sites where he had

showed a cause-and-effect relationship were the lungs and the nose.

4. TEMPORALITY- Obviously we expect the cause to precede the

effect. Hill recognized that temporality was not so clear for diseases that

developed slowly. “Does a particular diet lead to disease or do the early

stages of the disease lead to those peculiar dietetic habits?” Of current

interest are the epidemiologic studies that show a correlation between diet

soda and obesity. These studies are quick to assert a causal link, but there is

always a question of which way causation proceeds. One might reasonably

ask, “What kind of people drink diet sodas?”The kind of people who want

to lose weight. Most important from the temporal perspective, carbohy¬

drate restriction has immediate and dramatic effects on a disease, diabetes,

that has severe acute symptoms. The fact that the “concerns” about long¬

term effects have never actually materialized pretty much defines the crisis

in nutrition.

5. BIOLOGICAL GRADIENT. Association should show a dose-

response curve. In the case of cigarettes, the death rate from cancer of the

lung increases linearly with the number of cigarettes smoked. A subset

Observational Studies, Association, Causality

193

of the first principle—that the association should be strong—is that the

dose-response curve should have a meaningful slope—that is, the differ¬

ence between the numbers at the beginning and the end of the scale

should be substantial, and of course, should not show large fluctuations.

Numerous studies in the literature can avoid putting their results to the

test by lumping data into quartile. When fifty points are lumped into one

quartile, it might show slope, but it would be nice to know the gradient

within each quartile.

6. PLAUSIBILITY. Hill said, “What is biologically plausible depends

upon the biological knowledge of the day.” Here, Hills emphasis on effect

size is important. If the association is far-fetched, unexpected, or derived

from a less-than-obvious idea, there is a greater burden of proof and any

association should be strong.

7. COHERENCE. Data, according to Hill, “should not seriously conflict

with the generally known facts of the natural history and biology of the

disease.”The natural history of diabetes is the effect of carbohydrate—not

fat, not red meat, not any protein. Any number of other factors might be

involved, but if you dont put carbohydrate first, you are giving up on the

common sense that Hill acknowledged was the basis of his criteria.

8. EXPERIMENT. It was another age. It is hard to believe that it was

in my lifetime that Hill proposed this principle: “Occasionally it is possible

to appeal to experimental, or semi-experimental, evidence. For example,

because of an observed association some preventive action is taken, does

it, in fact, prevent?” The inventor of the randomized controlled trial would

be amazed how many try to take preventative action, and how many, in

fact, dont prevent—and most of all, he would have been astounded that it

doesn’t seem to affect the opinion of the medical community. The progres¬

sion of failures, from the Framingham Study to the Womens Health

Initiative, and the lack of association among low-fat, low saturated fat, and

CVD, is strong evidence for the absence of causation.

9. ANALOGY. “In some circumstances it would be fair to judge by

analogy. With the effects of thalidomide and rubella before us, we would

surely be ready to accept slighter but similar evidence with another drug or

another viral disease in pregnancy.”

With the list concluded, this is Hills final word on the criteria for

determining causation:

194

Nutrition in Crisis

Here then are nine different viewpoints from all of which we

should study association before we cry causation. What I do not

believe—and this has been suggested—is that we can usefully lay

down some hard-and-fast rules of evidence that must be obeyed

before we accept cause and effect. None of my nine viewpoints

can bring indisputable evidence for or against the cause-and-effect

hypothesis and none can be required as a sine qua non. What they

can do, with greater or less strength, is to help us to make up our

minds on the fundamental question— is there any other way of

explaining the set of facts before us, is there any other answer equally,

or more, likely than cause and effect? (Emphasis added)

This might be the first critique of the still-to-be-invented evidence-

based medicine.

Nutritional Epidemiology

There are many critics of current nutritional epidemiology and most of us

don't understand how the field could persist with so many weak results. It

is simply the low standards and the failure to understand that statistics is

not data, and it is not a science as such. Statistics comes from a particular

persons opinion on how the data should be interpreted. Conflicts then arise

from differing opinions. You must adjudicate between two principles: one,

“The risk is small but when you scale it up to the whole population, you will

save thousands of lives”; and two, to which I am partial, “When risk is small,

there is low predictability of outcome. You cant scale up bad data.”

The real impact of Hill's criteria is that they provide standards for

interpreting epidemiological studies. They are standards that still have a

good deal of subjectivity, but they are standards nonetheless. Yet they are

ignored in nutrition. They are ignored by authors, and most importantly,

they are ignored by the reviewers and editors who are expected to be the

gatekeepers on solid science. In the end, the decision that an observational

study implies causation is another way of saying that it is meaningful—

that it is not an outcome of mathematical juggling—that it is, you know,

science. The Emperor of All Maladies describes Hills criteria as principles

“which have remained in use by epidemiologists to date.” But have they?

Observational Studies, Association, Causality

195

Many have voiced criticisms of epidemiology as its currently practiced in

nutrition. One way to look at the current problems in nutrition is that we

have a large number of research groups doing epidemiology in violation of

Hill s criteria.

Is It Science?

Science is a human activity. What we dont like about philosophy of science

is that it is about the structure of science, rather than about what scientists

really do, and so there arent even any real definitions. Izja Lederhandler, a

colleague at the NIH, put it well: “What you do in science is, you make a

hypothesis and then you try to shoot yourself down.” A good experiment

puts the experimenter s theory to the test. An experiment whose outcome

only shows consistency is not strong.

One of the most interesting insights on the work of Hill and Doll, as

described in The Emperor of All Maladies, was that during breaks from the

taxing work of analyzing the questionnaires on smoking, Doll himself

would step out for a smoke. Doll believed that cigarettes were unlikely to

be a cause—he favored tar from paved highways as the causative agent—

but as the data came in, “in the middle of the survey, sufficiently alarmed,

he gave up smoking.” 6 In science, you try to shoot yourself down and you

go with the data. The mass pf papers demonizing fat, and the current flood

of papers demonizing sugar, fail most of Hills criteria. A major reason

that they fail is that they set out to show consistency between the data

and the theory, rather than to challenge the theory. They dont try to shoot

themselves down. As a result, they dont consider what other facts could

equally explain the data.

Navigating the Mess

The goal of this chapter, and those that surround it, is to help the consumer—

and perhaps other scientists—read scientific publications. There really are

no set principles of science. The existence of a “gold standard”—that is, the

one best type of experiment that answers all questions—is not recognized

in any science. The best experiment is the one that answers the question at

hand. Observational studies are appropriate if researchers cant intervene

196

Nutrition in Crisis

for practical or ethical reasons. Such experiments imply causality if they

show strong associations and if they have underlying mechanisms in basic

chemistry or biology.

Bradford Hill laid down principles for dealing with observational stud¬

ies, which he recognized as attempts to turn common sense into practice.

Hill’s criteria, once again, are as follows:

1. Strength

2. Consistency

3. Specificity

4. Temporality

5. Biological gradient

6. Plausibility

7. Coherence

8. Experiment

9. Analogy

Hill showed a causal link between cigarettes and cancer because he

found that deaths from lung cancer for cigarette smokers were nine to ten

times that of nonsmokers, a ratio that went up to twenty for heavy smok¬

ers compared to nonsmokers. He was less sure about coronary thrombosis

because the relative risk (RR) was in the range of 2:1. This standard is not

even proposed in modern nutritional experimentation. A nearly continu¬

ous flow of papers in the medical literature showing that something that

you eat will cause some disease you don’t want to have, even if the RR is as

low as 1.3, has marred the field substantially. The majority of epidemiologic

studies, in fact, violate Hill’s criteria.

The most important feature of scientific behavior is, again, the mind-set.

You have to have the courage to try to shoot your own theory down. This is

consistent with the general principle that excluding a hypothesis is always

stronger than showing consistency. The problem in nutrition is that experi¬

menters are trying to prove things, instead of trying to disprove things.

-CHAPTER 14-

Red Meat and

the New Puritans

Dost thou think, because thou art virtuous, there shall be no

more cakes and ale ?

—William Shakespeare,^^

E xperts on nutrition are like experts on sexuality. No matter how

professional they might be, in some way, they are always trying to

justify their own lifestyle. Sure that others should follow their own

standards of behavior, they are always tempted to save us from our own sins,

whether sexual or dietary. While our own New England Puritans were not

actually down on sex (as described in Sara VowelTs Wordy Shipmates, they

actually thought sex was an argument for Gods existence), they did have

the obsessive and literal insistence on what God really wanted. Sin was on

their minds.

The new Puritans of contemporary America want to save us from red

meat. It is unknown whether Michael Pollans In Defense of Food 1 was

reporting the news or making it, but Pollan’s recommendation to limit

meat consumption has become commonplace. Vegetarian Times says that

3.2 percent of the US population is vegetarian, and the Harvard School of

Public Health thinks that they are the righteous few among us.

There are good reasons for avoiding meat—better not to kill

anything, generally—but for most people, the health angle is not one

of the reasons. We are also suspicious when an epidemiologist doth

protest too much. Protein is chemically complex; or more precisely,

there are hundreds of different proteins that might have opposite

effects in the body. There are twenty amino acids that make up these

proteins and they, in turn, have many individual effects and some work

198

Nutrition in Crisis

together with the others. We are asked to believe that after they are

digested, our bodies can tell which amino acids came from animals and

which from vegetables.

It is not even clear that most of us eat a lot of meat. Protein tends to be a

stable part of the diet. There was no change in protein consumption in the

last forty years during the period of the obesity and diabetes epidemic. We

might eat more than recommended by health agencies, but their creden¬

tials are up for renewal. In fact many individuals, particularly the elderly,

don’t get enough meat. Given the complexities, it seems that the burden of

proof should be on those who want to show the dangers of meat. Both the

scientific and popular press, however, give you the idea that meat should be

considered guilty until proven innocent.

Red meat, in particular, is “linked to” just about every disease imag¬

inable, including diabetes, where if anything, it is likely to be beneficial.

The lipophobes’ approach is epidemiological and their numerous papers

have a common theme: find a weak association and claim that if the risk,

no matter how miniscule, is multiplied by the whole population, we can

save thousands of lives by reducing meat consumption. Here, again, I am

asking you to accept the idea that the best and the brightest are not doing

acceptable science. Using the principles from the previous chapters, I will

analyze a couple of specific papers. These reports provide examples of the

particular errors that you can look for when trying to decide if a scientific

paper is valid or not.

One important idea: All of statistics is subjective. Assumptions are

made and numbers are calculated from those assumptions. The numbers

are meaningful only if the assumptions fit the question to be answered.

You have to remember that statistical significance is a mathematical

term, meaning that differences between an experimental group and the

controls in a particular experiment did not arise by chance. It does not

mean, however, that those differences are of sufficient magnitude or are of

sufficient reliability that if the experiment were repeated it would not turn

out differently. It does not mean that the differences have any practical or

clinical significance, either. Til begin with an older red meat study, which

illustrates the limitations of relative risk and odds ratios. This particular

case has a punch line, and Til show you that the whole paper was probably

some kind of flimflam.

Red Meat and the New Puritans

199

Red Meat Scare of 2009

“Daily Red Meat Raises Chances of Dying Early” was the headline in the

March 24,2009, issue of the Washington Post? This scare story was accom¬

panied by a photo of a gloved hand slicing beef with a scalpel-like knife,

probably intended to evoke a CSI autopsy (although it still looked pretty

good to me, if slightly overcooked). I don’t know the reporter, Rob Stein,

but I don’t think we’re talking about anyone on the level of Woodward

and Bernstein. For those too young to remember Watergate, the reporters

from the Post were encouraged to “follow the money” by Deep Throat,

their anonymous whistleblower. The movie Fat Head suggests the variation

—more appropriate here—that we “follow the data.”

In the strange world of nutrition, scandalous behavior is right out in

the open. Researchers don’t like to be whistleblowers, because unless the

issue has some major impact, a breakdown in research principles makes

us all look bad. The missteps in published papers in nutrition, however,

require no insider information to pick apart. The Washington Post story was

based on a study, “Meat Intake and Mortality,” published in the profes¬

sional medical journal Archives of Internal Medicine by Rashmi Sinha and

coauthors. 3 It got a certain amount of press at the time of publication, but

following the usual pattern, it soon disappeared from view into the pile

of “accumulating evidence” that meat would kill you. Nobody looked at it

closely. Certainly not Rob Stein. To be fair, it’s not his fault. It takes time

to analyze such papers and he was likely assigned to a flower show the

next day. Although it is not unreasonable that he took the authors at their

word, it is worth looking at the details now. When analyzing studies on the

risk of a diet or drugs, looking back to Bradford Hill’s criteria, we should

expect that there first be specificity. (Recall how Hill emphasized that it

was lung and nose cancer specifically that were associated with cigarette

smoking.) However, specificity is exactly what was lacking in Sinha et al.’s

paper, whose main outcome measure was all-cause mortality—that is, the

number of deaths from any disease. If a paper is about all-cause mortality,

then the check on specificity is lost and we should be sure that this is for

real. Unsurprisingly, the conclusion will be that the paper fails to make a

good case for the danger of red meat, and in fact, its greatest virtue may be

in showing you how to see through the flaws that populate the literature.

200

Nutrition in Crisis

Common Sense and Experience First

Your best bet in dealing with potential bias is to demand that, no matter

how complicated the statistics are, the results should not violate common

sense—or if they do seem

to be counterintuitive,

that they are convincingly

justified. The further the

conclusion deviates from

common sense or previ¬

ous established results,

the greater the demand

for that justification. The

salient fact about red meat

is that during the thirty years that we describe as the obesity and diabetes

epidemic, overall protein intake was relatively constant, and red meat

consumption actually went down by a significant degree. As previously

shown in chapter 1, almost all of the caloric increase in that period was

due to an increase in carbohydrates. Fat, if anything, went down. However,

during this period, consumption of almost everything else went up: Wheat

and corn, of course, skyrocketed, but so did fruits and vegetables. The two

notable foods whose consumption went down were red meat and eggs. In

addition, we need to remember that much research shows the benefits of

replacing carbohydrate with protein, especially for the elderly. 4 So, here’s

something that you can use as a rule in analyzing papers in the literature.

Simple Statistics: Who's Likely to Die?

The paper by Rashmi Sinha et al. 5 is an observational study of the type

discussed in the previous chapter: They tried to match outcomes for differ¬

ent groups in an existing population with their dietary behavior. In terms

of informing the public, the report in the Washington Post is quite a bit

more accessible than the original paper. It reads:

Researchers analyzed data from 545,653 predominantly white

volunteers, ages 50 to 71, participating in the National Institutes

Principle 1

Biology, common sense, and

experience come before statis¬

tics. Do the results make sense?

Red Meat and the New Puritans

201

of Health-AARP Diet and Health Study. In 1995, the subjects

filled out detailed questionnaires about their diets, including

meat consumption. Over the next 10 years, 47,976 men and

23,276 women died.

So the risk, or probability, of dying if you were in the study population

is easy to calculate from the Washington Post report: overall probability

= (number of people who

died) / (all the people in

the study). That is equal to

(47,976 + 23,276)/545,653 Principle 2

= 0.13 or 13 percent. This Are the results meaningful—

is a good reference point, that is, large numbers?

something that we re sure of

before anybody starts doing

the statistics. The risk is not

great. Only slightly more than one in ten people died in the course of the

experiment. The article goes on to say:

Those who ate the most red meat—about a quarter-pound a

day—were more likely to die of any reason, and from heart

disease and cancer in particular, than those who ate the least—

the equivalent of a couple of slices of ham a day. Among women,

those who ate the most red meat were 36 percent more likely to

die for any reason, 20 percent more likely to die of cancer and

50 percent more likely to die of heart disease. Men who ate the

most meat were 31 percent more likely to die for any reason, 22

percent more likely to die of cancer and 27 percent more likely

to die of heart disease.

That sounds fairly scary, but is it? By this point you might be experienc¬

ing an emotional reaction to relative risk. This is almost always misleading.

So, let’s do what scientists call a back-of-the-envelope calculation to see if

we can get meaningful numbers. You can skip this if you don’t like math,

but its really not bad—just high school algebra. We have the overall risk of

mortality: 13 percent. Now, for both men and women, the increase in risk

202

Nutrition in Crisis

from eating meat is about 33 percent (36 percent for women; 31 percent

for men), so if N people died in the low-meat group, then 1.33iV died

among the high-meat group. Together they make up 13 percent of the

total, so we have: (N+ 1.33iV) = 2.33N= 13 percent. If you solve for A you

get 0.0558, or about 5.6 percent. From there we can determine that the risk

in the low-meat group is 5.6 percent, and the risk in the high-meat group

is 7.4 percent. So the absolute difference in risk is only 1.8 percent. Before

you even look at the original paper, we are now talking about a difference

in risk in the ballpark of a few percent. Your steak is already sounding a

lot better.

What we are trying to do here is to go beyond relative risk and get at

absolute risk. The dire warning that “those who ate the most red meat

were 36 percent more likely to die for any reason” lost its kick when

we did an absolute risk calculation. Why? Obviously, when it is a case

of relative anything, you need to know: Relative to what? In diseases

with low prevalence, or events with low probability, relative risk is almost

always misleading.

What Was Measured?

Another problem is that the paper says “A 124-item food frequency

questionnaire . . . was completed at baseline.” Many are suspicious of

food questionnaires because people might not accurately report what

they ate. While this can be a source of error, the amount of error

depends on how the data is interpreted. All scientific measurements

have error. Here the problem is the word “baseline.” The study was

started in 1995, continued for ten years, and mortality was followed

during this period. This means that some people died as long as ten

years after their reported food intake. The take-home question here:

Are you eating the same thing you were eating ten years ago? More to

the point, with all the kvetching about red meat in the media, is there

a chance somebody in this study reduced red meat consumption? Is

there a chance that they did it halfway through the study and would

have appeared in a different group if data had been collected for them

at that halfway point? To simplify even further: If I get sick, it is due to

the junk I ate in college?

Red Meat and the New Puritans

203

But, let's take them at their word and assume the study is okay. What do

we learn from the outcome? The damned statistics, as Mark Twain might

have called them, go like this: The study population of 322,263 men and

223,390 women was broken up into five groups (quintiles) according to

meat consumption, with the highest group taking in about seven times

as much as the lowest group. The groups were compared according to a

statistic called hazard ratio (HR), which, as explained below, in cases like

this is more or less the same as relative risk. The authors report that the

HR for eating red meat every day compared to eating red meat rarely is

1.44.This is pretty weak. Before discussing further, Til explain the different

terms for expressing risk. Again, the math is simple, but you can skip ahead

if you're daunted.

Understanding OR, RR, and HR

The common measures of relative outcomes in comparing a nutritional

or medical intervention with controls are odds ratio (OR), relative risk

(RR), and hazard ratio (HR). The good news is that for most of the stud¬

ies relevant to what were

doing here, these statisti¬

cal measures are all pretty

much the same (because

the incidence of disease

or other outcome is low).

As described above, “risk”

in medicine is the prob¬

ability of a given outcome,

and relative risk is what it

sounds like—the probability

of an outcome occurring in

one group compared to the probability of it occurring in a second group.

Let’s break things down. Probability is equal to the number of particular

outcomes divided by the total number of possible outcomes. We can look

at some examples from the world of gambling, using the word winning as

a stand-in for any particular outcome, such as getting sick or losing a target

amount of weight:

Principle 3

If the effect is not large, then

the data have to be very reliable.

Conversely, if the data have big

potential error, then the conclu¬

sion must be substantial.

204

Nutrition in Crisis

Probability = Number of ways of winning/Total possible outcomes

Probability of drawing = 4 aces/52 total cards = 1/13 = 0 .0769

an ace from a fair deck

Probability of drawing the _ j ^ of ^53 tMll caris , 1/s2 ,0.0192

ace of spades from a fair deck

In some cases, rather than the probability, the odds might be reported.

Odds are slightly different from probability:

Odds = Number of ways of winning/Number of ways of not winning

Odds of drawing an, 4 aces/48 non _ aces cards , 1/12 . 0 . 0833

ace from a fair deck

Odds of drawing the ace

of spades from a fair deck

= 1 ace of spades /

51 n ° n_aceof =1 /5 1= 0.0196

spades cards

As the likelihood of an event occurring diminishes, the odds and the

probability move closer together and are used similarly in conversation.

This is evident in the examples above: The probability of drawing the ace of

spades from a fair deck is 1 out of 52, or .0192, while the odds are 1 out of

51, or .0196. Because the event is unlikely to occur, the difference between

odds and probability becomes almost negligible.

In an experiment comparing outcomes between two groups, for example

cancer in meat eaters and cancer in vegetarians, you might report relative

risk (RR):

(Probability of an (Probability of

Relative Risk = . . x

outcome m group 1 ; outcome in group 2)

Relative risk of drawing any ace, compared to drawing an ace of spades:

(4/54) / (1/54) = 4

In a medical experiment you might not be able to wait until the

experiment is over to calculate the probability of a given outcome. This

is where hazard comes into play. Hazard is the same as the probability,

but measured over fixed time intervals. For example, you might measure

Red Meat and the New Puritans

205

how many people get sick in the first month in each group in a study.

Hazard is a rate and is the probability of “winning” for a fixed time

period. There is slightly more to hazard, mathematically, but in most

papers it is reported as the hazard ratio (HR) between two interven¬

tions. The bottom line in these cases is that you can think of HR as the

same as RR.

It is important to realize that once you calculate RR, HR, or OR, you

have lost track of how much absolute risk there was to begin with. If a

paper tells you the RR of a disease, you don’t know whether it is a rare

disease or one that everybody has. Bottom line: In reading a scientific

paper, you can take RR, HR, and OR as roughly the same. They all

provide a comparison between how likely two events are to occur. The

big caveat is that you have to make sure you know what the individual

probability of each event is. The problem is described well in the follow¬

ing narrative.

You are in Las Vegas. There are two blackjack tables, and for some reason

they have different probabilities of paying out (different number of decks,

for example). The probability of winning at Table One is 1 in 100 hands,

or 1 percent. At Table Two the probability of winning is 1 in 80 hands, or

about 1.27 percent. The ratio of the probabilities (RR of winning) is 1.27

/1,0 = 1.27. (A ratio of 1 would mean that there is no difference between

the tables.) Suppose all that you know is the RR of 1.27. Right off, some¬

thing is wrong: You are missing a lot of information. One gambling table

is definitely better than the other, but you have no way of knowing whether

the odds at either table are particularly good in the first place.

Suppose, however, that you did get a glimpse at the real info and you

could find out what the absolute odds are at the different tables: 1 percent at

Table One and 1.27 percent at Table Two. Does that help? Well, it depends

on who you are. For the guy who is sitting at the black-jack table when

you go up to sleep in your room at the hotel, and who is still there when

you come down for the breakfast buffet, he is going to be much better off

at Table Two. He will play hundreds of hands and the slightly better odds

ratio of 1.27 is likely to pay off. On the other hand, imagine that you are

somebody who will take the advice of my cousin, the statistician, who says

to just go and play one hand for the fun of it, just to see if the universe loves

you (that’s what gamblers are really trying to find out). You’re going to

206

Nutrition in Crisis

play the hand, win or lose, and then go off and do something else. Does it

matter which table you play at? Obviously it doesn’t. The odds ratio doesn’t

tell you anything useful because your chances of winning are pretty slim

either way.

Differences in Absolute Risk

Returning to the original red meat paper, there are a number of differ¬

ent “models” (reworking of the data) and there are mind-numbing tables

giving you the different hazard ratios. Using the worst case HR between

high and low red meat intakes for men, we get HR = 1.48, or, as they like

to report in the media, 48 percent higher risk of dying from all causes. It

sounds bad, but wait: What is the absolute difference in risk? Well, the

paper says that the whole group of 322,263 men was divided into five

subgroups (quintiles) of 64,453 people each. (Note that you usually have

to dig this information out of tables. Authors don’t always make it easy to

find the data, which is itself cause for suspicion.) The people who don’t eat

much red meat had 6,437 deaths, or 10.0 percent by population. The big

meat eaters must then have had 14.8 percent deaths. That’s an absolute

difference of only 5 percent. The authors then corrected for some things

that might have contributed to the outcome, bringing the absolute differ¬

ence for men down to a mere 3 percent.

Bottom line: There is an absolute difference in risk of about 3 percent.

It is unreasonable to think that this represents a call to change your life.

Remember, too, that this is for big changes—like six or seven times as

much meat. So, what is a meaningful HR? For comparison, the HR for

smoking versus not smoking, with regard to lung disease, was about 20.

For heavy smokers, the HR was about 30. Going back to Bradford Hill’s

criteria: “First upon my list I would put the strength of the association.”

Relative risk does not have meaning by itself. You must know the changes

in absolute risk. Another common way of looking at the data that is more

meaningful is the number needed to treat (NNT), which is the reciprocal

of the absolute risk. For our 3 percent risk in Sinha’s red meat study, you

would have to treat twenty to thirty people to save one life. That’s not great

but it’s something. Or is it?

Red Meat and the New Puritans

207

What About Public Health? Good News or Not?

Many people would say that, sure, for a single person, red meat might not

make a difference, but if the population reduced meat by half, we could save

thousands of lives. At this point, before you and your family take part in a big

experiment to save health statistics in the country, you have to apply Principles

2 and 3. You have to ask how strong the relations are, and to understand how

good the data is, you must look for things that would not be expected to have

a correlation. “There was an increased risk associated with death from injuries

and sudden death with higher consumption of red meat in men but not in

women,” which sounds like we are dealing with a good deal of randomness.

More important, what is the risk in reducing meat intake? The data

don’t really tell you that. Unlike cigarettes, where there is little reason to

believe that anybody’s lungs benefit from cigarette smoke, we know that

there are many benefits to protein, especially if it replaces carbohydrate in

the diet, especially for the elderly, and especially for all kinds of people. So

with odds ratios around 1, you are almost as likely to benefit from adding

red meat as you are reducing it. Technically, it is called a two-tailed distri¬

bution, which is to say that things can change in both directions. The odds

still favor things getting worse but it is a risk in both directions. You are at

the gaming tables. You don’t get your chips back. If you bet on reducing red

meat and it does not reduce your risk, it will increase your risk.

The Fine Print: The Smoking Gun

In writing this chapter, which was based on a blog post, I went back to the

original paper to check the calculations. I had used the numbers for men

because it was a worst case (and it still has the problems that I described),

but in recalculating things, I looked at the numbers for women as well.The

data unfolded like this.

The population was again broken up into five groups, or quintiles. The

lower numbered quintiles are for the lowest consumption of red meat.

Looking at the reported data on all-cause mortality, there were 5,314 deaths

for low consumption. When you go up to quintile 5—that is, highest red meat

consumption—there are 3,752 deaths. What? The more red meat, the lower

the death rate? In other words, the raw data show the opposite of the conclusion

208

Nutrition in Crisis

of the paper. There were fewer deaths at high red meat. The confounders are

listed in the legend to the figure. For the “basic model,” the data were corrected

for race and total energy intake, and risk went up. Why? We can’t tell if we

cant see what the effect was. Do you have the sense of flimflam?

A useful way to look at this data is from the standpoint of conditional

probability. We ask: What is the probability of dying in this experiment

if you consume a high-meat diet? The answer is simply the number of

people who both died during the experiment and were big meat-eaters

(Q5) divided by the total number of individuals in Q5. Here’s the calcula¬

tion: 3,752/(223,3905) = 0.0839, or about 8 percent. If you are not a big

meat-eater, your risk is (5,314 + 5,081+ 4,734 + 4,395)/(0.8 x 223,390) =

0.109, or about 11 percent.

This paper tested the hypothesis that red meat is associated with all¬

cause mortality. The data showed that it wasn’t. Meat wasn’t a risk unless

you dragged in other factors: education, marital status, family history of

cancer, body mass index, smoking history using smoking status (never,

former, current), time since quitting for former smokers, physical activity,

alcohol intake, vitamin supplement user, fruit consumption, and vegetable

consumption. All of these have to be added in to make the authors’ conclu¬

sion true. Once you do that, though, you have to ask why you would single

out red meat among these other ten inputs as the key variable. Wouldn’t it

be better to find out which of these factors had the biggest effect? What I

offer here is a professional scientist’s view, and I try to make my description

dispassionate and not insulting, but what is this if not deception?

What Is It About Red Meat?

Red meat isn’t a chemical, so what is it about the red meat that upsets people?

The meat? The red? To be fair to the authors, they also studied white meat,

which they found by the same prestidigitation mostly beneficial. But what

about potatoes? Cupcakes? Breakfast cereal? Are these completely neutral?

If we ran all these factors through the same computer, what would we see?

Unspoken in everybody’s mind is saturated fat, that Rasputin of nutritional

risk factors who will come after you despite enough bullets in its body to have

killed several scientific theories. Maybe it wasn’t the red meat per se but the

way it was procured. Maybe it’s ritual slaughter that conferred eternal life (or

Red Meat and the New Puritans

209

lack of it) on the consumer. For a gripping description of the total spiritual

effect of bad behavior, nothing is as harrowing as the Isaac Bashevis Singer

stories equating meat eating with other sins of the flesh. Finally, there is the

elephant in the room: carbohydrate. Basic biochemistry suggests that a roast

beef sandwich might have a different effect than roast beef in a lettuce wrap.

The Right Questions

My hope is that you will come away from this chapter with some basic

rules for dealing with the scientific literature—questions to ask and prin¬

ciples to apply:

1. Do the results make sense? Biology comes before statistics. Experience

comes before statistics.

2. Is there a big association? Are we talking about meaningful, that is, large

numbers?

3 . What was measured? If the effect is not large, then the data have to be

very reliable. If the data have big potential error, then the conclusion

must be substantial.

In nutrition, as in other fields, recommendations are often tinged by

the personal preferences of those dishing them out. In combination with

the dogmatic state of government and private recommendations, it takes

some work to know if you are reading a meaningful study. The point of this

chapter, and of this entire section of the book, is that you have the right to

ask for, and the author has the obligation to provide, clear explanations. The

practical application of Hills criteria is that results should make sense in

terms of magnitude and what you know about biology. You should also be

very suspicious if only relative risk is reported, even if just in the Abstract.

Remember, Alice has 30 percent more money in the bank than Bob, but

we don’t know whether she is rich.

You can usually calculate absolute risk (the number of cases divided by

the number of participants) so you can see if the report is about something

that is very rare. The specific case in this chapter, the effect of red meat, fails

to meet the criterion of meaningful associations and seems to be deployed

to distract from the real culprit, the real elephant in the room: carbohydrate.

210

Nutrition in Crisis

Harvard:

Making Americans Afraid of Red Meat

“There was a sense of Jeji-vu about the paper by Pan, et al.

entitled 'Red meat consumption and mortality; results from 2

prospective cohort studies’ 6 that came out in April of 2012.”

That s what I wrote online in May of 2012. Other bloggers

worked it over pretty well, so I ignored the paper for the most

part. Then I came across a remarkable article from the Harvard

Health Blog. Titled “Study Urges Moderation in Red Meat

Intake,” it was about the Pan et al. study and it described how

the “study linking red meat and mortality lit up the media.” It

claimed that “headline writers had a field day, with entries like

‘Red meat death study/ ‘Will red meat kill you?’ and ‘Singing

the blues about red meat/” This was too much for me.

What was odd about the post from the Harvard blog was

that “the field day for headline writers” was all described from

a distance, as if the study by Pan et al. (and the content of the

Harvard blogpost itself) hadn’t come from Harvard, but was

rather a natural phenomenon, similar to the way every seminar

on obesity begins with a graphic of the state-by-state progres¬

sion of obesity as if it were some kind of meteorological event.

The reference to “headline writers,”1 think, was intended to

conjure images of sleazy tabloid publishers like the ones who

are always pushing the limits of First Amendment rights in

the old Law & Order episodes. The Harvard blogpost itself,

however, was not any less exaggerated. It is not true that the

Harvard study urged moderation. In fact, the article admitted

that the original paper “sounded ominous. Every extra daily

serving of unprocessed red meat (steak, hamburger, pork, etc.)

increased the risk of dying prematurely by 13%. Processed red

meat (hot dogs, sausage, bacon, and the like) upped the risk

by 20%. ”

Red Meat and the New Puritans

211

That is what the paper urged. Not moderation. Prohibition.

“Increased the risk of dying prematurely by 13%. ” Who wants

to buck odds like that? Who wants to die prematurely?

It wasn’t just the media. Critics in the blogosphere were also

working overtime deconstructing the study. Among the faults

cited was the reporting of relative risk. Once again: Relative

risk is relative. It doesn’t tell you what the risk is to begin

with. Relative risk destroys information. The extreme example

remains that you can double your odds of winning the lottery

if you buy two tickets instead of one. So why do people keep

reporting it? One reason, of course, is that it makes your work

look more significant than it is, but if you don’t report the abso¬

lute change in risk, you might be scaring people about risks that

aren’t real. The nutritional establishment is not good at facing

their critics, but in this case, Harvard admitted that they didn’t

wish to push back against their detractors.

Nolo Contendere

Having turned the media loose to scare the American public,

Harvard admitted that the bloggers were correct. The Harvard

Health Blog allocated to having reported “relative risks,

comparing death rates in the group eating the least meat with

those eating the most.” “The absolute risks,” they admitted,

“sometimes help tell the story a bit more clearly. These numbers

are somewhat less scary” Why not try to tell the story as clearly

as possible in the original article then?

Anyway, there was a table on the Harvard Health Blog

that presented the raw data. This is what you need to know.

Unfortunately, the Harvard Health Blog didn’t actually calcu¬

late the absolute risk for you. You’d think that they would want

to make up for Dr. Pan scaring you. After all, an allocution is

supposed to remove doubt about the details of the crimes. Let’s

calculate the absolute risk. It s not hard.

212

Nutrition in Crisis

As previously, risk is a probability: that is, number of cases

divided by total number of participants. Looking at the data for

the men first, the risk of death with three servings per week is

equal to 12.3 cases per 1,000 people, or 1.23 percent. Now going

to fourteen servings a week the risk of death is 13 cases per 1,000,

or 1.3 percent. So, for men, the absolute difference in risk is less

than 0.1 percent. Definitely less scary. In fact, not scary at all.

Still, at least according to the public health professionals,

the risk could add up for millions of people. Or could it? We

have to step back and ask what is predictable about showing a

change of less than one-tenth of 1 percent risk. It means that a

small unpredictable event—a couple of guys getting hit by cars

in one or another of the groups—might throw the whole thing

off (remember we’re talking total deaths). Many other things

have a lower risk than that, but might not have been considered.

Maybe a handful of guys in upscale, vegetarian social circles lied

about their late night trips to Dinosaur Barbecue. The number

has no real-world significance. When can you scale up a small

outcome? If the outcome number is small, and you want to scale

it up, it must be secure and have little room for error.

There is an underlying theme here.There is the possibility that the

mass of epidemiology studies from the Harvard School of Public

Health and other groups are simply not real. Poor understanding

of science, cognitive dissonance, or something else altogether might

be the cause. Whatever it is, the progression of epidemiologic stud¬

ies showing that meat causes diabetes, that sugar causes gout, all

with low-hazard ratios—that is, small association size—might he

meaningless. It’s discouraging and almost impossible to believe. A

big piece of medical research is a house of cards.

Who Paid for This and What Should Be Done

We paid for it. Pan et al were funded in part by six National

Institutes of Health grants and NIH is still paying for it. The

Red Meat and the New Puritans

213

latest production from this group now emphasizes the “change

in meat consumption. 07 Remarkably, my objections to this

paper were published as a letter to the editor. 8 My main point

was that “Red meat consumption decreased as T2DM [type 2

diabetes mellitus] increased during the past 30 years.”

It is hard to believe with all the flaws pointed out by myself

and others, and in the end, admitted by the Harvard Health

Blog, that this work was subject to any meaningful peer review.

A plea of no contest does not imply negligence or intent to

do harm, but something is wrong. There is the clear attempt

to influence the dietary habits of the population. This cannot

sensibly be justified by an absolute risk reduction of less than

one-tenth of 1 percent, especially given that others have made

the case that some part of the population, particularly the

elderly, might not get adequate protein. The need for an over¬

sight committee of impartial scientists is the most reasonable

conclusion from Pan et al.

-CHAPTER 15-

The Seventh Egg

When Studies Defy Common Sense

S tepping back and looking at the recent nutritional literature, I am

struck by the miracle of life. How could humans have evolved in

the face of threats from red meat, eggs, and even shaving? (The

Caerphilly Prospective Study shows you the dangers of shaving ... or is it

the dangers of not shaving? 1 ) With 28 percent greater risk of diabetes here

and 57 percent greater risk of heart disease there, how could our ances¬

tors ever have come of childbearing age? Considering the daily revelations

from the Harvard School of Public Health, showing the Scylla of saturated

fat and the Charybdis of sugar between which our forefathers sailed, it is

a miracle that we're here.

These sensational stories that the popular media writes about, are

they real? They are, after all, based on scientific papers. Although not all

members of the media have the expertise to decipher them, reporters

generally talk to the researchers—and the papers must have gone through

peer review in the first place, right? However, as the previous chapters

suggest, the gatekeepers, as we think of peer reviewers, are less vigilant

than they should be. In fact, many papers that are published in the major

medical journals defy common sense. While it is hard to believe, the medi¬

cal literature really does have a remarkably high degree of error. Medical

publications are burdened with many examples of misinterpretation and

poor understanding of scientific design and analysis. While there is little

outright falsification of data, the researchers are not always doing credible

science. How can the consumer decide?

I am going to illustrate the problem with the example of a paper by L.

Djousse: “Egg Consumption and Risk of Type 2 Diabetes in Men and

Women.” 2 In the study by Djousse et al., participants were asked how

216

Nutrition in Crisis

many eggs they ate and then, ten years later, it was determined whether

they had developed diabetes. If they had, it was assumed to be because

of the number of eggs. Is this for real? Do eggs really cause changes in

your body that accumulate until you develop a disease—a disease that is,

after all, primarily one of carbohydrate intolerance? Type 2 diabetes, recall,

is due to impaired response of the body to the insulin produced by beta

cells of the pancreas, as well as a progressive deterioration of the insulin-

producing cells. Common sense tells us its unlikely that eggs could play a

major role, but it is still important to understand the methodology and see

if there is a something that justifies this obvious departure from common

sense. Again, Hills principles might be useful.

What did the experimenters actually do? First, subjects were specifically

asked “to report how often, on average, they had eaten one egg during the

past year,” and subjects where thus classified into one the following catego¬

ries of egg consumption: 0, <1 per week, 1 per week, 2-4 per week, 5-6 per

week, and 7+ eggs per week. The researchers collected these data every other

year for ten years. With these data in hand, they then followed subjects

“from baseline until the first occurrence of one of the following: (a) diag¬

nosis of type 2 diabetes; (b) death; or (c) censoring date, the date of receipt

of the last follow-up questionnaire,” which for men was up to twenty years.

Take a second and consider the last year of your life. Is it possible that

you might not be able to remember whether you had one versus two eggs

per week on average during the year? Is there any possibility that some of

the subjects who were diagnosed with diabetes ten years after their report

on egg consumption had changed their eating pattern over the course of

that decade-long experiment? Are you eating the same food you ate ten

years ago? Quick, how many eggs per week did you eat last year?

The Golden Rule Again

Right off, there is a problem in people accurately reporting what they ate.

This is a limitation of many—probably most—nutritional studies, and

while it can be a source of error, it is really a question of how you interpret

the data. All scientific measurements have error. You simply have to be sure

that the results that you are trying to find do not depend on any greater

accuracy than the data that you have collected. Eyeballing the paper by

The Seventh Egg

217

Djousse et al., we see that there are no figures, which is already a suspi¬

cious sign. We would have expected, at least, a graph of the number of

eggs consumed versus the number of cases of diabetes. The results, instead,

are stated in the abstract of the paper in the form of this mind-numbing

conclusion (don't actually try to read this; it is merely an illustration of the

tedious writing in the medical literature):

Compared with no egg consumption, multivariable adjusted

hazard ratios (95% Cl) for type 2 diabetes were 1.09 (0.87-1.37),

1.09 (0.88-1.34), 1.18 (0.95-1.45), 1.46 (1.14-1.86), and 1.58

(1.25-2.01) for consumption of <1, 1, 2-4, 5-6, and 7+ eggs^

week, respectively, in men (p for trend <0.0001). Corresponding

multivariable hazard ratios (95% Cl) for women were 1.06

(0.92-1.22), 0.97 (0.83-1.12), 1.19 (1.03-1.38), 1.18 (0.88-

1.58), and 1.77 (1.28-2.43), respectively (p for trend <0.0001).

What does all this mean? Very little. These “statistical shenanigans” are,

in fact, an argument against a correlation. If you look at the paragraph,

almost every number that you see is very close to 1. Without going through

a detailed analysis, you can simply extract from the tables some simple infor¬

mation: There were 1,921 men who developed diabetes. Of these, 197 were

in the high egg consumption group, or about 10 percent. For women, there

were 2,112 cases of diabetes, of whom 46 were high egg consumers, or a little

more than 2 percent. To me this suggests that diabetes is associated with

something other than eggs, and that it is probably unjustified of the authors

to conclude: “These data suggest that high levels of egg consumption (daily)

are associated with an increased risk of type 2 diabetes in men and women.”

What I described are the raw data, and as we saw in chapter 14, we have

to consider confounders. In fact, if we analyzed the data in detail, we would

find that the conclusion is actually poorly supported by the data, but lets

take the authors' conclusion at face value for the sake of argument.

The Seventh Egg

If the authors' conclusion is correct, this means that there was no additional

risk of diabetes from consuming 1 egg per week as compared to eating

218

Nutrition in Crisis

none. Similarly, there was no additional risk in eating 2-4 eggs per week

or 5-6 eggs per week. However if you upped your intake to 7 eggs or more

per week, then thats it: You were at risk for diabetes. It makes me think of

the movie The Seventh Seal , directed by Ingmar Bergman. Very popular in

the fifties and sixties, these movies had a captivating if pretentious style:

They sometimes seemed to be designed for the Woody Allen's parodies

that would follow. One of the famous scenes in The Seventh Seal is the

protagonist's chess game with Death. It requires only a little imagination

to see the egg as the incarnation of the Grim Reaper.

Mindless Statistics

A study of 20,703 men and 36,295 women makes a very weak case for

a connection between egg consumption and type 2 diabetes. Few of the

people who developed diabetes were big egg eaters. The problem is the

mindless use of statistics. If statistics ever go against common sense, it is

the authors' responsibility to explain why—in detail and in the abstract.

Sometimes, you can get a sense of how real the statistics are by looking

for simple things: How many people were in the study and how many got

sick? In other words, are we talking about a rare disease or one that had

low probability? In the case of this particular experiment from Djousse et

al., diabetes is a major health risk, but if you take 1,000 men for ten years,

and only 1 in 10 might develop diabetes, you need to be sure there is a big

difference between the group that followed the behavior you are looking at

(egg consumption) and those who didn't.

-CHAPTER 16-

Intention- to-Treat

What It Is and Why You Should Care

T he medical literature has some strange customs, but nothing beats

intention-to-treat (ITT), the most absurd and amusing statistical

method that has appeared recently According to ITT, the data from

a subject assigned at random to an experimental group must be included in

the reported outcome data for that group, even if the subject does not follow

the protocol, or even if they drop out of the experiment. In other words, it

doesn't matter if you eat what the experimenter told you to eat—your lipid

profile has to be included in the final report. It s easy enough to tell that the

idea is counterintuitive if not completely idiotic—why would you include

people who are not in the experiment in your data? The burden of proof in

using such a method should rest with its proponents, but no such obligation

is felt, and particularly in nutrition studies—such as comparisons of isoca¬

loric weight Toss diets—ITT is frequently used with no justification at all.

Astoundingly, the practice is sometimes actually demanded by reviewers in

the scientific journals. As one might expect, there is a good deal of controversy

on this subject. Physiologists or chemists, or really anybody with scientific

training, who hears about ITT requirements will usually walk away shaking

their head or uttering some obvious reductio ad absurdum —for instance, “You

mean, if nobody takes the pill, you report whether or not they got better

anyway?”That’s exactly what it means. In this chapter Til describe a couple of

interesting cases from the medical literature, and one relatively new instance

Fosters two-year study of low-carbohydrate diets—to demonstrate the abuse

of common sense that is the major characteristic of ITT.

On the naive assumption that some people really didn't understand

what was wrong with ITT—Ive been known to make a few elementary

mistakes in my life—I wrote a paper on the subject. It received negative,

220

Nutrition in Crisis

even hostile, reviews from two public health journals. I even got substan¬

tial grief from reviewers at Nutrition and Metabolism , where it was finally

published. I was the editor at the time, and I had extensive communica¬

tions with one reviewer who was willing to forego anonymity.

The title of my paper was “Intention-to-Treat. What Is the Question?”

My point was that there’s nothing inherently wrong with ITT if you

are explicit about what you are measuring. If you use ITT, you are really

asking: What is the effect of assigning subjects to an experimental proto¬

col? However, is anybody really interested in what the patient was told to

do rather than what they actually did?

The practice of ITT comes from clinical trials, where you cant always

tell whether patients have taken the recommended pills, just as in the real

situation where you never know what people will do once they leave the

doctor’s office. In that case you do an analysis based on your intention; that

is, you have no other choice than to designate those who were assigned to the

intervention as the “experimental group,” even though they might not have

participated in the experiment. That’s what we always did without giving it a

special name. When you do know who took the pill and who didn’t, however,

there are two separate questions: Did they take the pill, and is the pill any

good? That’s the data. You have to know both, and if you want to collapse

them into one number, you have to be sure you make clear what you are talk¬

ing about. You lose information if you collapse efficacy and adherence into

one number. It is common for the abstract of a paper to correctly state that

the results are about subjects “assigned to a diet” but by the time the results

are presented, the independent variable has usually become simply “the diet,”

rather than “assignment to diet,” which most people would assume meant

what people ate rather than what they were told to eat. Caveat lector.

My paper on ITT was perhaps overkill. I made several different

arguments, but a single commonsense argument gets to the heart of the

problem. I’ll describe that first, along with a couple of real-world examples.

Commonsense Argument Against

Intention-to-Treat

Consider an experimental comparison of two diets in which there is a

simple, discrete outcome (e.g., a threshold amount of weight lost or

Intention-to-Treat

221

Table 16.1. Hypothetical Results from the Thought Experiment for Analysis of Diets A and B

Diet A

Diet B

Compliance (of 100 patients]

100

Success (reached target)

ITT success

50/100 = 50%

“Per protocol” (followed diet] success

50/50 = 100%

50/100 = 50%

remission of an identifiable symptom.) Patients are randomly assigned

to two different diets, diet A or diet B, and a target of, say, 5 kilograms

of weight loss is considered success. As shown in table 16.1, half of the

subjects in diet A are “compliers,” able to stay on the diet, while the other

half are not. The half of the patients on diet A who were compliers were

all able to lose the target 5 kilograms, while the noncompliers did not. In

diet B, on the other hand, everybody stayed on the diet, but somehow only

half were able to lose the required amount of weight. An ITT analysis

shows no difference in the two outcomes—half of group A stayed on the

diet and all lost weight, while in study B, everybody complied but only half

had success.

Now, suppose you are the doctor. With such data in hand, should you

advise a patient: “Well, the diets are pretty much the same. It s largely up

to you which you choose,” as required by ITT? Or, alternatively, looking at

the raw data would the recommendation be: “Diet A is much more effec¬

tive than diet B, but people have trouble staying on it. If you can stay on

diet A, it will be much better for you, so I would encourage you to see if you

could find a way to do so.” You are the doctor. Which makes more sense?

Diet A is obviously better, but it s hard for people to stay on it. This is

one of the characteristics of ITT: It always makes the better diet look worse

than it is. In the submitted manuscript, I made several arguments trying to

explain that there are two factors. One of them, whether it actually works,

is directly due to the diet. The other, whether you follow the diet, is under

control of other factors (whether WebMD tells you that one diet or the

other will kill you, whether the evening news makes you lose your appetite,

etc.). I even dragged in a geometric argument because Newton had used

one in the Principia : “A two-dimensional outcome space where the length

of a vector tells how every subject did . . . ITT represents a projection

222

Nutrition in Crisis

of the vector onto one axis, in other words collapses a two-dimensional

vector to a one-dimensional vector, thereby losing part of the information.”

Pretentious? Moi?

Why You Should Care: Surgery or Medicine?

Does your doctor actually read these academic studies using ITT? One

can only hope not. Consider the analysis by David Newell of the Coronary

Artery Bypass Surgery (CABS) trial. This paper is fascinating for the blan¬

ket, tendentious insistence, without any logical argument, on something

that is obviously fundamentally foolish. Newell wrote: “The CABS research

team was impeccable. They refused to do an as treated’ analysis: We have

refrained from comparing all patients actually operated on with all not

operated on: This does not provide a measure of the value of surgery.” 1

You read it right. The results of surgery do not provide a measure of the

value of surgery.

In the CABS trial, patients were assigned to be treated with either

medicine or surgery. The actual method used and the outcomes are shown

in table 16.2. Looking at the table, you see the effects of assignments

(ITT): 7.8 percent mortality for those assigned to receive medical treat¬

ment (29 deaths out of 373), and 5.3 percent mortality for those assigned

to surgery (21 deaths of 371). On the other hand if you look at the outcomes

of each treatment as actually used, it turns out that medical treatment had

a mortality rate of 9.5 percent (33 deaths of 349), while among those who

Table 16.2. Results from the CABS Trial from Newell

Allocated medicine Allocated surgery

Received

surgery

medicine

surgery

medicine

Survived 2 years

296

354

Died

Total

323

369

Source. David J Newell, "Intention-to-Treat Analysis: Implications for Quantitative

and Qualitative Research,” International Journal of Epidemiology 23, no. 5 [1992]

837-841

Intention-to-Treat

223

actually underwent surgery, the mortality rate was only 4.1 percent (17

deaths of 419). Mortality was less than half in surgery compared to medical

treatment. Making such a simple statement, that surgery was better than

medicine, Newell claimed, “would have wildly exaggerated the apparent

value of surgery.”The “apparent value of surgery”? Common sense suggests

that appearances are not deceiving. If you were one of the 16 people who

were still alive because you chose surgery based on outcome (17 deaths)

rather than because of ITT (33 deaths), you would think that it was the

theoretical report of your death that had been exaggerated.

The thing that is under the control of the patient and the physician, and

that is not a feature of the particular modality, is getting the surgery actually

done. Common sense dictates that a patient is interested in surgery, not the

effect of being told that surgery is good. The patient has a right to expect

that if they comply with the recommendation for surgery, the physician

should try to avoid any mistakes from previous studies that prevented other

patients from actually receiving the operation. In another defense of ITT,

Hollis and Campbell made the somewhat cryptic statement: “Most types

of deviations from protocol would continue to occur in routine practice.” 2

This seems to be saying that the same number of people will always forget

to take their medication and surgeons will continue to have exactly the

same scheduling problems as in the CABS trial. ITT assumes that practi¬

cal considerations are the same everywhere and that any practitioner is

equally capable, or incapable, as the original experimenter when it comes

to getting the patient into the operating room.

One might also ask what happens when two studies give different values

from ITT analysis. In the extreme case, one might suggest that if the same

operation were recommended at a hospital in Newcastle-upon-Tyne as

opposed to a battlefield-in Iraq, the two ITT values would be different. So

which is the appropriate one to attribute to that surgical procedure?

The ITT Controversy

Advocates of ITT see its principles as established and might dismiss a

commonsense approach as naive. They usually say that removing noncom-

pliers introduces bias by destroying the original randomization. In this, they

are confusing the process of randomization with the criteria for inclusion.

224

Nutrition in Crisis

Case Study: Vitamin E Supplementation

A good case for the problems with ITT is a report on the value

of antioxidant supplements. The abstract of the paper in ques¬

tion concluded that “there were no overall effects of ascorbic

acid, vitamin E, or beta carotene on cardiovascular events among

women at high risk for CVD.” This conclusion—that there was

no effect of supplementation—was based on an ITT analysis, but

on the fourth page of the paper one can see the remarkable effect

of not counting subjects who didn’t comply: “Noncompliance

led to a significant 13% reduction in the combined end point of

CVD morbidity and mortality ... with a 22% reduction in MI

. .,, a 27% reduction in stroke ... [, and] a 23% reduction in the

combination of MI, stroke, or CVD death.”

The media universally reported the conclusion from the

abstract, namely, that there was no effect of vitamin E. This

conclusion is correct if you think that you can measure the

effect of vitamin E without taking the pill out of the bottle.

(It s like the old joke about how you make a really dry martini:

You don't remove the cap from the Vermouth bottle before

pouring.) Does this mean that vitamin E is really of value? The

data would certainly be accepted as valuable if the statistics

were applied to a study of the value of, say, replacing barbecued

pork with whole grain cereal. Certainly we can see that there is

“no effect” when you tell a patient to take vitamin E, but that

doesn’t answer the question that most people are interested in:

What are the effects when the patient actually takes the vitamin f

If you excluded people with diabetes from your study and only found out

when the study started that one of the subjects actually had diabetes, like

the tainted juries in courtroom dramas, you would be required to remove

the individual anyway. You would not have included them if their diabetes

Intention-to-Treat

225

came out in the voir dire, so you cant include them now. Of course, if

you don’t know whether a subject has actually complied, you are forced to

include everybody, and that is what we have always done when there is no

choice. That is not the issue here.The question is: What happens when you

know who did and who didn’t comply?

The problem is not easily resolved. Statistics is not axiomatic: There is

nothing analogous to the zeroth law (the idea of thermal equilibrium on

which thermodynamics rests). All statistics rests on interpretation and

intuition. If this is not appreciated—if you do not go back to consideration

of exactly what the question is that you are asking—4t is easy to develop a

dogmatic approach and insist on a particular methodology because it has

become standard.

Assumption versus Consumption

Described in chapter 3 as the “shot heard round the world,” Gary Foster’s

2002 dietary study found that after one year, a low-carbohydrate diet was

substantially better than a low-fat diet for markers of cardiovascular disease.

When it came to weight loss, however, while the low-carbohydrate diet

was better at six months, the two diets were “the same at one year.” This, of

course, was an effect of dwindling adherence in both the intervention and

control groups, (although, in fact, the lipid markers were noticeably better

on the low-carbohydrate group even at one year.)

A follow-up to the landmark one-year study, “Weight and Metabolic

Outcomes after 2 Years on a Low-Carbohydrate Versus Low-Fat Diet,” 3

published in 2010, had a surprisingly limited impact. What went wrong?

The first paper was revolutionary, and in the follow-up, the authors explicitly

addressed the need for including a “comprehensive lifestyle modification

program,” because the original was criticized for simply giving the people

on the Atkins diet a copy of the popular book. The criticism was addressed,

and the conclusion was the same. The low-carbohydrate group had better

outcomes for most CVD risk factors, although again, there was no differ¬

ence in weight loss after two years. As stated in the conclusion: “Neither

dietary fat nor carbohydrate intake influenced weight loss when combined

with a comprehensive lifestyle intervention.”This is, after all, still the party

line and should have been well-received by establishment nutrition. That

226

Nutrition in Crisis

should have been a big win for the only-calories-count crowd. Why was

there so little impact, and why is the paper so rarely cited?

It is likely that the Zeitgeist has shifted since eight years ago. Strict

scientific standards have suffered tremendous blows. The willingness to

disregard the results of big expensive experiments and to ignore common

sense is now much more widespread. Everybody knows—at this point—

that in a big comparison trial, the low-carbohydrate diet will win in some

way, and that it is likely that the authors will try to put a negative spin

on things. As Elizabeth Nabel put it after the embarrassing Womens

Health Initiative, failure: “Nothings changed.” Low-fat is still the name

of the game, a phrase in the USDA guidelines, and a component of the

Healthy Hunger-Free Kids Act school lunch program 4 endorsed by

Michelle Obama. Press releases on such government programs quote a

progression of “suits” with maudlin expressions of optimism about how

well what I call “Fruits ’n Vegetables” are doing in improving obesity. There

are numerous disclaimers about good fats and bad fats, so that you cant

really hold them to anything. However, since Foster s first study, numerous

low-carbohydrate trials have showed it to be superior to any competition.

Foster s 2010 study was, thus, not the challenge that the original was. It

should, however, have been given some attention, at least because it was

a follow-up to the landmark first paper. In my opinion, it deserved more

attention primarily because it represented a new low in misleading science.

The conclusion of the paper: “Neither dietary fat nor carbohydrate intake

influenced weight loss.”

I admit that I had not read Fosters paper carefully before making

the pronouncement that it was not very good. I was even upbraided by

a student for such a rush to judgment. I explained that that is what I do

for a living. I explained that I usually don’t have to spend a lot of time

with a paper to get the general drift. I could easily see a couple of errors

in methodology, which I describe below, but I was probably not totally

convincing, so I went ahead and read the paper, which is quite a bit longer

than usual. The main thing that I was looking for was information on the

nutrients that were actually consumed, since it was their lack of effect that

was the main point of the paper. The problem is that people feel that they

can call anything that they want a low-carbohydrate diet, and of course,

people really believe that being “assigned to a low-carbohydrate diet” is the

Intention-to-Treat

227

same thing as consuming a low-carbohydrate diet. Frequently when you

look carefully at these studies, it turns out the diets were similar, so it is not

surprising that the results are the same.

“Any Reasonably Intelligent High School Student”

In a diet experiment, the food consumed should be right up front, but in this

case I couldn’t find it at all. Foster s 2010 paper is quite long, with a tedious

appendix on the lifestyle intervention, but I read the whole thing carefully.

I really did. The data weren’t there. I was going to write to the authors when

I found out—I think through somebody’s blog—that this paper had been

covered in a story in the Los Angeles Times. As reported by Bob Kaplan:

Of the 307 participants enrolled in the study, not one had their

food intake recorded or analyzed by investigators. The authors

did not monitor, chronicle or report any of the subjects’ diets.

No meals were administered by the authors; no meals were

eaten in front of investigators. There were no self-reports, no

questionnaires. . . . The lead authors, Gary Foster and James

Hill, explained in separate e-mails that self-reported data are

unreliable and therefore they didn’t collect or analyze any.

I confess to feeling a bit shocked. How can you say “neither dietary fat

nor carbohydrate intake influenced weight loss” if you haven’t measured

fat or carbohydrate? If you think that self-reported data are not good,

then you cant make judgments about what was consumed. This would

constitute a remarkable statement, since in fact the whole nutrition field

runs on self-reported data. Are all the papers from the Harvard School of

Public Health and all those epidemiology studies that rely on food records

to be chucked out?

What would have happened if the authors had actually measured the

relevant data—if they had asked what people eat, which, as Kaplan put it,

is “the single most important question . .. that any reasonably intelligent

high school student would ask.” It’s not just bad experimental design. It is

a question of what is on their mind. Do they not realize that it is totally

inappropriate to say that fat or carbohydrate are not important if they

228

Nutrition in Crisis

haven’t measured them? They might be so biased that they don’t see what

is going on. In his first paper, Foster said in public, explicitly, that he set out

to trash the Atkins diet. It’s not a good way to do science. You are supposed

to try to trash your own theory and show that it survives.

Beyond ITT: Fabricating Data

Foster’s 2010 paper shows just how misleading ITT can be. Initially chas¬

tised for jumping to conclusions, I reread the paper carefully and one figure

in particular caught my eye. The title: “Predicted Absolute Mean Change

in Serum Triglycerides.” Predicted? That sounds strange. What about the

real data? The figure indicates changes in triglycerides for the three-, six-,

twelve-, and twenty-four-month time periods.

Reduction in triglycerides is the hallmark of low-carbohydrate diets.

Almost everybody on such a diet lowers their triglycerides (mine fell to half

of the original value, a commonly reported outcome). In figure 16.IT, the

difference in the levels of triglycerides between the two diets (shown by the

double-headed arrow) was quite large at three or six months, the usual result.

However, as the experiment continued, after twelve months, triglyceride

values got closer and actually came together after twenty-four months.

This seemed strange, so I realized I had to find out where the word

predicted came from, and that meant reading the methods section and

going over the statistical analysis on how the data had been handled. In

general, as mentioned before, you usually only read the methods section

in detail if you think that it is a problem, or if a new method has been

introduced, or if you want to repeat the author’s experiment yourself. Large

studies like this usually have a statistician, and they use standard methods

whose details might or might not be understood by a nonstatistician (that’s

me) and most of the authors. They are usually accepted at face value and it

is assumed that authors have adequately explained to the statistician what

the question is that they want to address. As I kept plowing through the

statistical section, I found it increasingly tedious and difficult to read until

I hit this passage:

The previously mentioned longitudinal models preclude the

use of less robust approaches, such as fixed imputation methods

Triglyceride Difference Between Diets 03 Change in Triglyceride Level (mg/dL)

Intention-to-Treat

229

-5

-10

-15

-20

-25

-30

-35

-40

-45

-50

Month

Figure 16.1. A , The double-headed arrow represents the difference between the

change in triglycerides on the two indicated arms of the study. Figure redrawn from

Foster et al. B f The distance shown by the arrow on the y-axis in part A (difference

between the low-carbohydrate and low-fat groups) is shown as a function of the

number of subjects who dropped out of the study.

230

Nutrition in Crisis

(for example, last observation carried forward or the analysis of

participants with complete data [that is, complete case analy¬

ses]). These alternative approaches assume that missing data are

unrelated to previously observed outcomes or baseline covariates,

including treatment (that is, missing completely at random).

Data “missing completely at random”? Whats going on here? In a

nutshell, this is another implementation of ITT. In the study, the authors

used “data” from people who dropped out of the experiment. All they had

to do was “assume that all participants who withdraw would follow first

the maximum and then minimum patient trajectory of weight.” The key

words are withdraw and assume . This is really a step beyond ITT, where

you would include, for example, the weight of people who showed up to be

weighed but had not actually followed one or another diet. Here, nobody

showed up. There are no data. A pattern of behavior is assumed and data

are—lets face it—made up.

The world of nutrition puts big demands on irony and tongue-in-cheek,

but the process in Foster's paper suggests that the results could, in theory,

be fit to a model for a three-year study, or a ten-year study. As people

dropped out you could “impute” the data. In some sense, you could do

without any subjects at all. Nutrition experiments are expensive: Think of

the money that could be saved if you didn’t have to put anybody on a diet

and you could make up all the data.

The Diet or the Lack of Compliance?

It is odd that ITT is controversial. In fact, it’s odd that ITT exists at all.

A reasonable way to deal with concern, however, that would satisfy every¬

body, is simply to publish both the ITT data and the data that include only

the compilers, the so-called “per protocol” group—that is, the subjects that

were actually in the experiment. This is what was done in the vitamin E

study described in an earlier sidebar. It made the authors look bad, sure, but

they did the right thing. Such data are missing from Foster’s paper. Given

the high attrition rate, one could guess that the decline in performance in

both groups was due to including “data” from the large number of people

who failed to complete the study. It turns out we can test this. The number

Intention-to-Treat

231

of people who dropped out is in the paper. To find out whether the decline

in performance is due to including the made-up data from the dropouts,

we can plot the difference in triglycerides between the two groups (the

double-headed arrow in figure 16. L4, for each time point, against the

number of people who discontinued treatment.

Figure 16.15 gives you the answer. You can see a direct correlation

between the number of dropouts and the group differences. The reason

that “decreases in triglyceride levels were greater in the low-carbohydrate

than in the low-fat group at 3 and 6 months but not at 12 or 24 months”

was almost certainly because, with time, more and more people included

in the data weren't on the low-carbohydrate diet, or any diet at all. ITT

always makes the better diet look worse.

This serves as an example of a case where correlation strongly implies

causality: The expectation from previous literature is that there will be a

large difference in the level of triglycerides between the low-carbohydrate

group and the low-fat group. (Decrease in triglycerides is virtually the defi¬

nition of low-carbohydrate diets.) This expected difference was observed at

the beginning of the trial, when the data were correlated with people actu¬

ally taking part in the experiment. With time more people dropped out,

so there were fewer people on either diet and, in turn, there was decreased

difference between the groups—not being on a low-carbohydrate diet

is, unsurprisingly, similar to not being on a low-fat diet. The take-home

message is that ITT and imputing data will reduce the real effect of the

intervention. In a diet experiment, the nature of the diet and compliance

might be related—if the diet is unpalatable, people might not stay on it—

but you have to show this. “Might” is not data. Along these lines, however,

it is likely that the major reinforcer, the major reason people will stay on a

diet, is that it works. In any case, one cannot assume that the two are linked

without specifically testing the idea.

The Bottom Line

Intention-to-treat comes from the realization that in some experiments,

you don't know who followed the protocol and who didn't. In a clinic, you

might write a prescription and not know if it's been filled. In those cases,

you have no choice but to include everybody's performance. However,

232

Nutrition in Crisis

the mechanical application of the idea in situations where you know who

did or didn’t comply is another misguided and dogmatic application of

statistics. ITT usually makes the better diet look worse than it actually is.

Awareness of whether this method has been used is important for evaluat¬

ing a scientific publication.

-CHAPTER 17-

The Fiend

That Lies Like Truth

lpull in resolution and begin

To doubt thequivocation of thefiend

That lies like truth .

—William Shakespeare,M^M

E rrors, inappropriate use of statistics, and misleading presentations

are everywhere in the medical literature, as IVe described in the

previous chapters. Here I will summarize some of the specific

problems weVe covered and try to provide you with a guide to what to

look for in the medical literature, and especially in the popular media.

Look for Visual Representations

You have a right to demand, and the author of a scientific paper has an

obligation to provide, a clear explanation of what the results of their study

really mean. A good test of whether or not the authors are holding up

their end of the arrangement is the number and clarity of the figures.

Visual presentation is almost always stronger than long tabulation of

numbers. This principle is simultaneously so reasonable, and at the same

time so widely violated, that Howard Wainer wrote a whole book, Medical

Illuminations , arguing that scientific papers need more figures instead of

the dense tables that make results hard to understand and force the reader

to take the authors' conclusions at face value. 1

Scientific publication is changing, and increasingly, as more and more

journals become open access and available online, results of scientific stud¬

ies will be universally available. An advantage to an open-access online

234

Nutrition in Crisis

journal is that there is no longer a limit on pages or number of figures. Nor

is there an extra charge to the authors for the use of color in these figures.

It is unknown how frequently authors take advantage of these opportuni¬

ties, but the Golden Rule of Statistics remains the same: Let us see the

data completely and clearly and in as many figures as it takes.

Science, but Not Rocket Science

“Eating breakfast reduces obesity” is not a principle from quantum elec¬

trodynamics. Most of us know whether or not eating breakfast makes us

eat more or less during the day. Most of us also know that eating “good'

anything is not good for losing weight—good might imply nutritious or

tasty but usually means eating a lot. You don't need a physician, one who

may have never studied nutrition, to tell you whether your perception is

right or wrong. A degree in biochemistry is not required to understand

the idea that adding sugar to your diet will increase your blood sugar—and

the burden of proof is on anybody who wants to say that sugar is okay for

people with diabetes. Anything is possible, but we start from what makes

sense. Of course, there is technically sophisticated science and there are

principles that require expertise to understand, but you should not auto¬

matically assume this is the case.

Be Suspicious of Self-Serving Descriptions

If a paper is about a diet that is described as “healthy,” your appropri¬

ate answer should be, “thats for me to decide.” The way the media tosses

around the word healthy is bad enough, but in the context of a scientific

paper it has to be considered an intellectual kiss of death. Nothing that you

read after that can be taken at face value. If we knew what was healthy, we

would not have an obesity epidemic and we would not need another paper

to describe it.

Similarly, guidelines, data, or analyses that the authors describe as

“evidence-based medicine” are likely to be deeply flawed. To use the court

of law as an analogy: There must be a judge to decide admissibility of

evidence. You cant pat yourself on the back and expect to be considered

impartial. The courts have ruled that testimony by experts has to make

The Fiend That Lies Like Truth

235

sense, too. Credentials are not enough. As in the first principle, experts

have to be able to explain data to the jury on the data’s own merit. Be

suspicious if the authors tell you how many other people agree with their

position. Science does not run on “majority rules” or consensus.

Leaping Tall Buildings

As indicated in chapter 12, the new grand principle of doing science is habeas

corpus datorum —in other words, let’s see the body of the data. If the conclu¬

sion is counterintuitive and goes against previous work or common sense,

then the data must be strong and clearly presented. Here’s another way to put

it: If you say you can jump over a chair, I can cut you a lot of slack and assume

you’re telling the truth. If you say you can jump over a building, however, I

need to see you do it to be convinced. My daughter, at age nine, suggested

an additional requirement. In a discussion of superheroes I pointed out that

Superman used to be described as being “able to leap tall buildings in a single

bound.” She pointed out that if you try to leap tall buildings, you only get

a single bound. You can’t say your hundred-million-dollar, eight-year-long

randomized controlled trial was not a fair test. The fat-cholesterol-heart

hypothesis was sold as an absolute fact. None of the big clinical trials should

have failed, but in the end, almost every single one did.

To return to the idea of “Let’s see the body of the data”: Show me

what was done before you start running it through the computer. Statistics

might be important, but in a diet experiment where one has to assume

that even a well-defined population is heterogeneous, you want to under¬

stand how the individuals performed. The compelling work of Nuttall and

Gannon, 2 for example, showing that diabetes can be improved even in the

absence of weight loss, is increased in impact by the presentation of the

individual performance. Not only is there general good response in reduc¬

tion of blood glucose excursions, but all but two of the individual subjects

benefited substantially, and all but one got at least somewhat better.

Understand Observational Studies

The usual warning offered by bloggers and others is that association

does not imply causality, and that observational studies can only provide

236

Nutrition in Crisis

hypotheses for future testing. A more accurate description, as detailed in

chapter 13, is that observational studies do not necessarily imply causality.

Sometimes they do and sometimes they don’t. The association between

cigarette smoke and lung disease has a causal relation because the associa¬

tions are very strong and because the underlying reason for making the

measurement was based on basic physiology, including the understanding

of nicotine as a toxin.

In this sense, observational studies test hypotheses rather than generat¬

ing them. There are an infinite number of phenomena, but when you try

to make a specific comparison between two, you usually have an idea in

mind, conscious or otherwise. When a study tries to find an association

between egg consumption and diabetes, it is testing the hypothesis that

eggs are a factor in the generation of diabetes. It might not be sensible or

based on sound scientific principles, but its a hypothesis. If you do find

a strong association with an unlikely hypothesis (this is the definition of

intuition) then you have a new plausible hypothesis, and it must be further

tested. It is important to question the hypothesis being tested in the first

place, though. If the introduction section to the study says eggs have been

associated with diabetes, you can at least check whether the reference is to

an experimental study rather than a previous recommendation from some

health agency.

Meta-Analysis and the End of Science

Doctors prefer a large study that is bad y to a small study that

is good.

—Anonymous

While intention-to-treat is the most foolish activity plaguing modern

medical literature, the meta-analysis is the most dangerous. In a meta¬

analysis you pool different studies to see if more information is available

from the combination. As such it is inherently suspicious: If two studies

have very different conditions, the results cannot sensibly be pooled. If,

on the other hand, they are very similar, you have to ask why the results

from the first study were not accepted. Finally, if the pooled studies give a

different result than any of the individual studies, the authors are supposed

The Fiend That Lies Like Truth

237

Group Statistics:

Bill Gates Walks into a Bar . ..

Everybody has their favorite example of how averages dont really

tell you what you want to know, or how they are inappropriate

for some situations. The reason most of these are funny is because

they apply averages in cases where single events are important.

One of my favorites, from the title of this sidebar: If Bill Gates

walks into a bar, on average, everybody in the bar is a millionaire.

Technically speaking, averages start with the assumption that

there is a kind oPtrue’Value and that deviations are due to random

error. If we could only control things well—if there were no wind

resistance and each ball was absolutely uniform—all of the balls

would always fall in the same place all the time. Any spread in

values must be random rather than systematic. The familiar

bell-shaped curve that illustrates the mean (average) value and

deviations from the mean is referred to alternatively as the normal

distribution or Gaussian distribution. Without going into details

of statistics, the standard deviation, or SD, is supposed to give you

a feel for how the real points deviate from the mean. From the SD,

you can tell how reliable the data are. A low SD means that the

mean value can be expected to be a reliable indication of what the

actual data look Eke. A high SD, on the other hand, means that

you can expect the data to be more spread out.

There are many examples. The textbooks cite the blitz in

London in World War II when the V-l “flying bombs” were

distributed in a characteristic random pattern. (“Aiming” was

presumed due to fixed launch sites and fixed fuel; deviations

came from wind resistance and other random factors.) You can

see a pretty good fit to the 3-D version of, in this case, the

Poisson distribution.

On the other hand, Allied bombing of the German town

of Aachen destroyed much of the city while leaving intact the

238

Nutrition in Crisis

famous cathedral where Charlemagne was crowned. The joke

was that the cathedral was unharmed because that’s what

we were aiming at It is generally assumed that what is good

enough for bombing is good enough for social and biological

science. But is it? Going from a large collection of individual

points to two mathematical parameters (the mean and standard

deviation), you obscure a good deal of information. A literally

infinite number of different arrangements of points will give

you the same mean and standard deviation.

Even more importantly are you really sure that your data are

uniform—that is, as the statisticians describe, that they come

from the same population? Getting to the point: How are group

statistics used in nutrition and medicine? The underlying prin¬

ciple is usually stated as the idea that "one size fits all.” But is

this a good idea? When we see a spread in weight loss on differ¬

ent diets, for example, is that due to minor individual variations

(one subject always goes to the gym right after breakfast), or is

it due to fundamental differences between people? For example,

might there be two classes of people (Eg., insulin-sensitive and

insulin - resistant)?

Nobody Loses an Average Amount of Weight

Use of averages depends on uniformity in the population. Yet

in diet, we don’t really believe that. Calling attention to all the

people who eat more than you do and gain less weight than you

do is considered an anecdotal observation. (The fact that you

don’t want to believe it adds some credibility.) There have been

experiments, however. David Allisons group 3 studied a number

of identical twins, putting them all on a weight-reduction diet

for a month and measuring how much weight they lost. They

calculated an energy deficit (differences between calories in and

calories out) and compared that to the actual energy deficit

(weight loss minus food intake).

The Fiend That Lies Like Truth

239

Estimated Energy Deficit JMJ]

Figure 17.1. Relationship of estimated and measured energy deficit for

fourteen twin pairs (same numbers]. The dotted line is the regression

of estimate on measured energy deficit and the solid line represents

identity (measured deficit « estimated). Subjects above the solid line

(of identity) are absolutely less efficient than those below the line. Data

from V Rainer et at, '‘A Twin Study of Weight Loss and Metabolic Efficiency”

International Journal of Obesity and Related Metabohc Disorders 25, no 4

(2001): 533-537.

The pairs of twins in figure 17.1 were similar in their energy

deficit, but there were major variations between twins. The

study was carried out in a hospital but, given the possible

random variations of twenty-eight people over the course of a

month, this is pretty good evidence for what is popularly called

“metabolic advantage” (people below the solid line) or maybe

metabolic “disadvantage” (people above).

240

Nutrition in Crisis

to point out what the original study did wrong (and the original authors

are supposed to agree).

Simply adding more subjects is not considered a guarantee of reliability;

however, papers in the medical literature frequently cite their large number

of subjects as one of their strengths. If a meta-analysis is good for anything

(which is questionable), it is for its originally intended role of evaluating

small, underpowered studies with the hope that putting them together

might reveal something you didn't originally see. It is a kind of “Hail Mary”

last-ditch play. It was not intended for appropriately powered studies with

a large number of subjects.

One of the benefits, conscious or unconscious, that keeps meta-analysis

going is that it is perfect for current medical research. With meta-analysis,

no experiment ever fails and no principle is ever disproved. Sugar causes

heart attacks, cholesterol causes heart attacks, red meat causes heart

attacks, and statins prevent heart attacks—it doesn't matter how many

studies show no effect. One winner and you can do a meta-analysis. Just

one more expensive trial and we'll show that saturated fat is bad. Plus you

don’t even have to explain what the previous researcher did wrong, as you

might in a real experiment.

The idea that simply adding more subjects will improve reliability is not

reasonable. Most of us think that if a phenomenon has large variability,

then mixing studies will reduce predictability although it might sharpen

statistical significance. And again, in science it is expected that if your

results contradict previous experiments, you will provide evidence as to the

cause of the differences. What did previous investigators do wrong? Do

those investigators now agree that you've improved things? Probably not,

especially if you don't ask them.

Be Suspicious of Grand Principles

“Randomized controlled trials are the gold standard.” “Metabolic ward

studies are the gold standard.” “Observational studies are only good for

generating hypotheses.” Such grand principles do not play a part in the

physical sciences, where the method that we choose depends on the ques¬

tion that we want to answer. You do not need to carry out a long-term trial

to see if a treatment is appropriate for an acute condition. You do not need

The Fiend That Lies Like Truth

241

Saturated Fatty Acids,

Carbohydrates, and Meta-Analysis

A number of important meta-analyses have examined the effect

of saturated fatty acids on cardiovascular risk. They all show a

very small benefit if the saturated fatty acids are replaced with

carbohydrate. Examination of the data, however, reveals that

almost all of the included studies failed to show any significant

effect of replacing saturated fat, and yet the authors came up

with an answer* How is this possible? It is the consequence of

group-think* If everybody—the editors, the reviewers, and ulti¬

mately the reader—assumes that meta-analysis is an acceptable

method, then peer review will be meaningless.

a randomized controlled study to find out whether penicillin is effective for

treating gram-positive infections. Penicillin is a drug and you might have

to do a randomized controlled trial to determine safety, but at the efficacy

end, if the results are clear-cut, only a small number of tests are necessary.

How many? It depends on how many people recover spontaneously. A

statistician can figure this out for you if you ask the question appropriately.

The idea of the randomized controlled trial as a gold standard has really

never recovered from Smith and Pells landmark paper, “Parachute Use

to Prevent Death and Major Trauma Related to Gravitational Challenge:

Systematic Review of Randomised Controlled Trials.” 4 In the paper, Smith

and Pell concluded that “the effectiveness of parachutes has not been

subjected to rigorous evaluation by using randomised controlled trials.”

Described as both funny and profound, the paper included all the relevant

information, making fun of the excessive statistical detail in the literature:

“We chose the Mantel-Haenszel test to assess heterogeneity, and sensitivity

and subgroup analyses and fixed effects weighted regression techniques to

explore causes of heterogeneity. We selected a funnel plot to assess publica¬

tion bias visually and Egger s and B egg’s tests to test it quantitatively.”

242

Nutrition in Crisis

Smith and Pell pointed out that there were only two solutions to the

problem of a lack of randomized controlled trials: “The first is that we accept

that, under exceptional circumstances, common sense might be applied when

considering the potential risks and benefits of interventions. The second is

that we continue our quest for the holy grail of exclusively evidence-based

interventions and preclude parachute use outside the context of a properly

conducted trial.” (Emphasis added)

In the end, the authors suggested that “those who advocate evidence-

based medicine and criticise use of interventions that lack an evidence

base ... demonstrate their commitment by volunteering for a double blind,

randomised, placebo controlled, crossover trial.”

The bottom line is that science has to make sense and does not depend

on arbitrary rules. Smith and Pell made fun of how often authors attempt

to snow us. There really are statistical tests with those names, but science

is not about fixed rules and accounting. It is about understanding each

experiment and knowing the question to be asked.

The faults that plague the medical literature have made it a minefield for

readers, and more often than not, readers will not be happy with the latest

study showing that processed food or one or another of the usual suspects

will kill you. Armed with a few principles, you can understand what is

explicitly wrong with these studies that so strongly violate common sense.

Failure to show individual data; the use of questionable practices, such as

intention-to-treat and meta-analyses; and, most of all, failure to actually

apply common sense, are the things to look for. It will be hard to accept that

the Harvard School of Public Health, all the best and the brightest, could

be making elementary mistakes, but the proof and precedents are there.

PART 5

The Second

Low-Carbohydrate

Revolution

-CHAPTER 18-

Nutrition in Crisis

M ore work needs to be done” is the standard conclusion of weak

dietary studies or those that got the “wrong answer.” As in any

biological science, what we know is much less than what we

don't know. When it comes to moving forward on the low-carbohydrate

argument, however, no more work needs to be done: Switching from your

current diet to a low-carbohydrate or ketogenic diet is usually the healthi¬

est change you can make. We know this because it has been put to the test.

After being subjected to forty years of persistent criticism—some rigorous,

some preposterous and outlandish, some virulent—carbohydrate restric¬

tion has proven to be the most effective way to lose weight and the first

line of treatment for diabetes.

Yet for forty years, the medical and nutritional establishment have pulled

out all the stops in a compulsive effort to find something medically wrong

with low-carbohydrate diets. They've come up with nothing at all. In head-

to-head comparisons, regardless of the length of the trials, low-carbohydrate

almost always wins. At the same time, we have forty years of multicenter,

multimillion-dollar trials that fail to show any risk from total fat or saturated

fat or dietary cholesterol. We need to face—and capitalize on—the scientific

data that stand clearly before us before any more work needs to be done.

For the individual dieter, we have found the optimal strategy. It's not a

magic bullet, and it's not something that works the same for everybody, but

it's the best we have. The challenge now is to work out a plan to implement

the low-carbohydrate strategy. The crisis in nutrition is that you might be

on your own. Your physician or health provider is not guaranteed to be on

your side and might, in fact, be trying to dissuade you from the best course

of action. If you're lucky, and your provider is reasonable, you might be

able to teach them. A physician, in particular, might have no training at

all in nutrition, and the practice of analyzing original research critically is

246

Nutrition in Crisis

unlikely to be part of their daily activity, so you can point them to relevant

literature. Nutrition in Crisis would be a good start.

Twilight of the Lipophobes

So how close are we to good science and good guidance on nutrition? I

spoke at the LEO Conference, held in Gothenburg, Sweden, which cele¬

brates nonconformity in science. The 2008 meeting honored Uffe Ravnskov,

the arch cholesterol skeptic. 1 Quoting Max Planck, speakers before me

suggested that if you really wanted to introduce a new idea in science, you

had to wait for the old generation to die out. When I spoke, 2 I suggested

that since I was in that generation, we might want to do it a little sooner.

How far do we have to go? As in any highly contentious field, the answer

you get depends on whom you ask. While low-fat still hangs over everything,

there are now numerous disclaimers: “Only saturated fat is a risk” or maybe

it s only a problem if “you replace polyunsaturated fat with saturated fat.”

The large, expensive, randomized controlled trials that successfully showed no

effect of dietary total saturated fat or cholesterol are still assumed to have never

taken place—and although the case for low-carbohydrate and its benefits is

established, the best that can be done according to many is to claim that “it

is not any better” than low-fat diets, despite the fact that low-fat never wins,

and only ever draws. Attacking sugar is considered a first step—presumably

because sugar is an easy target. Whether the second step, if there really is one,

is a more comprehensive meaningful low-carbohydrate recommendation, is

never stated. It is a political strategy that involves fundamental disdain for the

audience, and one that is almost guaranteed to backfire.

Things are progressing, yes. Many researchers and organizations, while

admitting no wrongdoing, are slowly giving in to low-carbohydrate. Yet

there is simultaneously unrestrained backlash, which has taken a very

ugly turn toward personal and professional attacks on low-carbohydrate

researchers and journalists.

Quashing Dissent around the World

Coincidentally, it was in Sweden that the medical establishment revealed

its resistance to criticism and to new ideas. Dr. Annika Dahlqvist lives in

Nutrition in Crisis

247

Njurunda, Sundsvall. She described on her blog 3 how she discovered that

a low-carbohydrate diet would help in her own battle with obesity and

various health problems that included enteritis (irritable bowel syndrome,

or IBS), gastritis, fibromyalgia, chronic fatigue syndrome, insomnia, and

snoring. Recommending low-carbohydrate diets for her patients and

publicly advertising her ideas drew a certain amount of media attention,

leading to a run-in with the authorities. In November 2006 she lost her

job at Njurunda Medical Center. She was ultimately exonerated in January

2008, when the National Board of Health and Welfare found that a low-

carbohydrate diet was “consistent with good clinical practice.”This was the

likely prelude to the announcement in 2013 that the Swedish Agency for

Health Technology Assessment, which is charged with assessing national

health care treatment, endorsed low-carbohydrate diets for weight loss.

While their statements were far from enthusiastic, it was one of a number

of events that indicated the fall of the low-fat paradigm and the no-holds-

barred backlash of nutritional medicine.

The most bizarre case, which was resolved as I was finishing this

book, occurred in Australia. An orthopedic surgeon, Dr. Gary Fettke,

had become the target of the Australian Health Practitioner Regulation

Agency (AHPRA) after he publically discussed the benefits of a low-carb

diet. In a letter dated November 1,2016, the AHPRA told Fettke, “There

is nothing associated with your medical training or education that makes

you an expert or authority in the field of nutrition, diabetes or cancer.” 4

The ensuing persecution was remarkably unprofessional and ugly, and it

wasn t until nearly two years later that the AHPRA exonerated Fettke and

offered a belated and inadequate apology. The training of any physician in

nutrition is, in my view, problematic, but it has been rightly pointed out,

in response to the original letter, that Fettke received the same training as

other doctors.

The letter continued: “Even if, in the future, your views on the benefits of

the [low-carb, high-fat] LCHF lifestyle become the accepted best medical

practice, this does not change the fundamental fact that you are not suit¬

ably trained or educated as a medical practitioner to be providing advice

or recommendations on this topic as a medical practitioner. ”In short, even

if the AHPRA recognized the truth, they would not defend Fettke s right

to say it. As many noted at the time, Dr. Fettke is in fact trained to remove

248

Nutrition in Crisis

the limbs of those whose diabetes has reached a critical point, despite the

AHPRA’s attempts to diminish his credentials. This lack of due process

and common decency speaks to the need for some kind of regulation. The

issue hasn’t receive significant coverage in traditional media, which have

kept it relatively invisible, but the recent turn of events may bring out just

how bad things are. The fallout from this scandal is just being uncovered,

and Marika Sboros’s blog, foodmed.net, is a likely source for the unfolding

story. Her recent blog post titled “Fettke: Cover-Up Grows After Case

Against Him Collapses?” is both gripping and intensely distressing.

Stateside, a North Carolina blogger named Steve Cooksey was enjoined

from recommending a low-carbohydrate diet for diabetes by the North

Carolina Board of Dietetics/Nutrition. After a near-death experience from

his own diabetes, and after experiencing the benefits of carbohydrate restric¬

tion, he felt it might be good to share the information with others, with due

disclaimers about not offering medical advice. That was too much for the

professional nutritionists who somehow have gotten the state legislature to

designate them as sole purveyors of nutritional advice. Enlisting the aid of

the Institute for Justice, Cooksey won a First Amendment lawsuit and the

North Carolina board realized that it could not tell American citizens what

they could and could not say. The board changed its guidelines, but the fury

of a nutritionist spurned is great—and even worse down under.

The Dietitians Association of Australia (DAA) registers dietitians and

essentially controls whether they can do their jobs in hospitals, universi¬

ties, or private practice. In 2015, the DAA expelled Jennifer Elliott, a

dietitian with thirty years’ experience and a highly regarded book for

patients. She had also published a peer-reviewed critique of the diet-heart

hypothesis, which likely contributed to the DAA action. The complaint

from a DAA-registered dietitian was that Elliott’s recommendation

of low-carbohydrate diets for her patients with type 2 diabetes was

not “evidence-based.” The DAA upheld the complaint. In my personal

communication with Claire Hewat, head of the DAA, however, she claimed

that the cause of dismissal had not been the recommendations of low-

carbohydrate diet but something else that could not be revealed because it

was confidential. The power of low-carbohydrate diets to provoke hostile

behavior remains fascinating if incomprehensible. In a blog post on the

subject, I made an analogy to the Protestant Reformation, although unlike

Nutrition in Crisis

249

The Tim Noakes Case

It is troublesome to hear that Tim is being attacked

so strongly for what seems to be a trivial matter ;

when there are plenty of real problems in health

care and our world more broadly .

—Walter Willett 5

Most bizarre for its virulence remains the determination with

which the Health Professionals Council of South Africa

(HPSCA) tried to maintain control by attacking Tim Noakes.

A retired physician and widely admired sports expert, and only a

relatively recent spokesman for carbohydrate restriction, Noakes

answered a comment on Twitter that suggested a newborn

could be weaned to a low-carbohydrate diet, not particularly

different from the recommendations of Australian authorities.

Yes, this is about a single tweet, which wasn't even taken into

account by the individual who J d asked the question in first place.

A nutritionist was insulted by the affair and brought the case

to the HPSCA. The question become whether Noakes s tweet

constituted medical advice—the primary 7 affirmative argument

being that because he was a medical doctor, anything Noakes

said could be considered advice. This is just as ridiculous as the

folk myth that a black belt in karate has to register his hands at

local police stations as lethal weapons.

The HPSCA came down hard on Noakes but, after two

years of trial, he was exonerated. The HPCSAs own committee

found that no harm derived from the tweet, and that his advice

was “evidence-based” and within normal standards. HPSCA

was able to establish a warning for other medical profession¬

als, since as Noakes explained, “It s been very, very demanding

on us and on our lives and financially it's been huge.” Noakes

asked, further, “Did they ever consider the consequences for my

250

Nutrition in Crisis

wife and myself and our family?” Indicating that it did not, the

HPSCA immediately appealed the decision of its own commit¬

tee. While the decision of the new body was supposed to be

produced with all deliberate speed, innate pettiness presumably

caused them to drag it out for months. Noakes was finally exon¬

erated. Besides being a person of statute in sports medicine,

he is a charming, friendly person—I met him at a landmark

conference at Ohio State University just as the final edits on

this book were prepared.

Luther Jennifer Elliott was not particularly rebellious and was not trying

to reform anything except her patients’ dietary habits. The DAA func¬

tions, after all, as a professional dietitians’ organization and should, as in

Macbeth , against the murder shut the door, not bear the knife itself. There

was some press coverage but little public awareness or professional outcry

over the lack of natural justice and common decency.

Dissent will not be tolerated. When Nina Teicholz, author of The Big

Fat Surprise, published a critical analysis in the BMJ describing excesses

of the expert report by the HHS-USDA Dietary Guidelines Advisory

Committee (DGAC)—notably that it did not even discuss low-carbo¬

hydrate diets—Michael Jacobson, the obsessive and humorless leader of

the Center for Science in the Public Interest (CSPI, a largely vegetarian

watchdog group), initiated a petition to have the paper retracted because

it was “so riddled with errors.” The alleged errors turned out to be less

about Teicholz’s scientific conclusions than about who used what database

as evidence. From Jacobson’s petition: “Teicholz states that ‘in the NEL

systematic review on saturated fats from 2010 ... fewer than 12 small trials

are cited, and none supports the hypothesis that saturated fats cause heart

disease_’ Correction: It is incorrect to state that none of the trials cited

in the 2010 NEL review supports the hypothesis.”

The paper was not retracted, presumably because no scientific issue

was raised and Teicholz’s critique was judged “within the realm of

Nutrition in Crisis

251

scientific debate.” A revealing bit of information was exposed along

the way, however: Frank Hu of Harvard School of Public Health,

lead author of the DGAC review of saturated fats, personally sent out

multiple emails to colleagues, putting together a posse of more than

150 people, including many graduate students, who ultimately signed

the CSPI petition. The pjetition remains problematic in that it did not

represent an assembly of people who had expressed concern, but rather

an assortment of people who were contacted by Dr. Hu with an unsolic¬

ited request to sign the petition.

The Big Picture

The extent to which repressive behavior actually reduces research on or

acceptance of dietary carbohydrate restriction is unknown. However, the

hostile, personal attacks—generally uncharacteristic of scientific behav¬

ior—represent desperation. If they could actually deal with the issue on a

scientific basis, they wouldn't have to behave like this. It is likely, however,

that the aggression is a response to seeing people, including their patients,

find out the answer for themselves.

The studies from Volek's lab are particularly compelling, but in fact, almost

every comparison in the scientific literature shows that low-carbohydrate

diets are more effective for weight loss and most other metabolic distur¬

bances. The rationale is that dietary carbohydrate is the major stimulus for

secretion of the anabolic hormone insulin. Chapter 9 explained how it is

possible to lose more weight, calorie for calorie, on a low-carbohydrate

diet: The bottom line is that insulin slows the breakdown (lipolysis) of fat,

and if you consume another meal before the system has had a chance to

deal with stored fat, it will accumulate. Also, on a low-carbohydrate diet,

your appetite goes down—satiety is increased, so most people find it easier

to adhere to a carbohydrate-restricted diet than to any other.

The real win for low-carbohydrate diets is personal: It changes your

interaction with food. For people with a weight problem, every meal is a

battle. Every time you sit down at the table you're trying to make sure you

don't cross some line of calories, fat, or whatever other metric you're keep¬

ing an eye on. If you make a substantial cut in the amount of carbohydrates

that you eat, even if you are not overly precise, even if you spend a couple of

252

Nutrition in Crisis

days at the conference buffet, you are unlikely to gain any weight. You lose

a substantial part of your obsession with and anxiety about food.

It is reassuring, too, to know that we don’t have a biological need for

carbohydrate and that the opposite of lowering fat intake is not eating all

the fat in sight. It might well be true, as critics of low-carbohydrate diet

say, that the recommended low-carbohydrate strategy includes assurances

that “you can eat all the fat that you want,” but the emphasis is on want.

Fat is filling. How much do you want? It is important to stress that the

low-fat idea was not originally instituted to deal with obesity but rather

to prevent heart disease (which it didn’t) and it always had a moralistic

overtone. It was somebody’s idea of what we should have been eating

during the millennia in which haute cuisine evolved and people perfected

sausages and other food in ethnic cuisines. The low-fat idea was the work

of Puritans. It gave rise to the obesity, epidemic, and the fact that many

people did not get fat is only a testament to the adaptability of humans in

dealing with all kinds of food.

The real impact of low-carbohydrate and ketogenic diets is the tie to

biologic mechanism and the generality of control of metabolism by insulin

and associated hormones. From a scientific standpoint, the main ideas are

established, and all that remains is political and social acceptance. As this

acceptance sets in, our understanding of metabolic control mechanisms

might unlock new and exciting possibilities.

-CHAPTER 19-

Cancer

A New Frontier for Low-Carbohydrate

I t was in July 2012 that I realized we'd won. It was now clear that we

had a consistent set of scientific ideas supporting the central role of

insulin signaling in basic biochemistry. We could see a continuum

with the effectiveness of dietary carbohydrate restriction for obesity, diabe¬

tes, and general health. The practical considerations—how much to eat of

this, how much to eat of that—hadn't yet been fully resolved, but we had

the kernel of a scientific principle. In fact, it was not so much that we had

the answer as that we had the right question. We were, at last, looking at

diet, metabolism, and disease as a single entity. We were asking whether

pathologic changes could be precisely traced back to disease, and that

meant focusing on the downstream effects of high-carbohydrate intake.

In science, the question is frequently more important than the answer. If

you know what to look for, you are more likely to find it. Of course it

wasn't originally about winning, or even fighting. When my colleagues

and I got into this about ten years earlier, coming from a background of

basic biochemistry, we hadn't anticipated that it would be such a battle,

that there would be so much resistance to what we thought was normal

scientific practice. Surprisingly, it was cancer studies that made clear how it

all fit together, and seemed to signal the success of all our work.

We generally recognize cancer as a genetic disease—not necessarily

inherited, but a disease associated with genetic mutations. A great deal of

research has gone into identifying the mutations and where the products

of the mutant genes fit in. While this has given rise to much valuable

information, it has been impossible to identify a single mechanism for

cancer. Numerous mutations have been identified but finding a common

thread has rarely been possible—and the information we do have has led

254

Nutrition in Crisis

to only a small number strategies, pharmaceutical or otherwise, for treat¬

ment or prevention.

The focus of cancer research, however, has recently begun to turn

back to earlier studies on the relation of energy metabolism to the

cancerous state. With this shift in focus, researchers have paid greater

attention to the Warburg effect, the observation that many tumors rely on

glycolysis—the anaerobic processing of glucose to pyruvic acid described

in chapter 5. The re-appreciation of energy metabolism has had wide-

ranging consequence. This is not to say that the genetic mutation theory

has somehow been replaced by a metabolic theory. In fact, it is usually

impossible to separate genetic and metabolic effects. Metabolism runs

on enzymes, which are proteins, and proteins are primarily controlled by

their synthesis, the gene products. In other words, although you could

consider DNA to be a blueprint, that blueprint is not simply copied and

sent out to the factory. Rather, we keep going back to the blueprint, and

the workers vote on whether the body will work on one or another part of

the blueprint, and choose what to do with each piece created. This slightly

strained analogy means that the total state of the cell, its energy profile

and cell constituents, will determine which proteins stipulated by the

genome will be expressed and the extent to which they will be active. The

focus on metabolism was the thread that brought us back to nutrition and

the control of energy production and utilization. There suddenly seemed

to be better appreciation of the need for a nutritional approach. I felt that

we had reached some kind of turning point.

The locale was a conference in Washington, DC, called “Metabolism,

Diet and Disease.” It might have been better billed as “Metabolism, Drugs

and Cancers,” at least based on the first day, but there was a surprising

thread in the various talks—surprising to me, at least—about the very

strong association between obesity and risk of cancer. Less surprising

to me, because of the numerous studies I’d read, were the presentations

about the effect of total dietary calorie restriction, the apparent basis for

the connection between obesity and cancer: Reducing calories has been

shown to reliably increase longevity and to control cancer in animals. An

additional element of the conversation was the hormone insulin, which

had popped up as a major player in various experiments on cancer. Then

there was the identification of downstream signaling elements—the

Cancer

255

compounds and proteins that transmit the information about cell stimu¬

lation to the interior of the cell. These were important results, and again,

they frequently pointed to the components of insulin pathways and even

to an association between cancer and diabetes. This last point was not

entirely new to me: Outstanding experiments had indicated the critical

role of insulin, and I always wondered why the connection to dietary

carbohydrate hadn't been made. In any case, the second day of the confer¬

ence included presentations on dietary carbohydrate restriction, and

although he was not listed on the organizing committee, Gary Taubes had

helped set up the event and deserves credit for bringing together cancer

people and low-carbohydrate people.

My colleague Dr. Eugene Fine presented a poster at the Washington

conference. Many conferences have poster sessions: Presenters pin posters,

typically four by six feet, to easels, and attendees at the conference can then

discuss the subject matter with the presenter. Posters don't always have a

big impact, so we were grateful to Gary Taubes for making Eugene’s poster

known to the main speakers.The paper on which the poster was based, now

published in the journal Nutrition , a describes a small study conducted with

ten seriously ill cancer patients. The study had the modest goal of show¬

ing that a ketogenic diet was a safe and feasible regimen, and in fact, the

patients did well. Six of the ten stabilized or went into partial remission.

By itself, this study would be considered only a small step forward, but in

fact, it was key in tying together the fields of carbohydrate restriction and

cell signaling in the context of normal and cancer cells. The experiment

was difficult to do:The patients had to have refused or failed chemotherapy

and exhausted all the traditional regimens. Yet it was incredibly significant.

Given what we know about insulin and low-carbohydrate diets, the experi¬

ment should have been done twenty years ago. These two lines of thought

simply had to be brought together.

In hindsight, it seems that workers in carbohydrate restriction should

have paid more attention to downstream cell signaling. We thought that

the role of insulin in system biochemistry made it clear that carbohydrate

restriction was built on a solid foundation. The resistance to the idea seemed

incomprehensible and parochial, but this conference made it obvious that

more was needed to form a consistent biological story. It became clear that

there was a conceptual barrier to acceptance of carbohydrate restriction

256

Nutrition in Crisis

beyond the traditional resistance to anything associated with the Atkins

diet. There were additional factors. There was a mind-set that prevented

adequate synthesis of all the information.

Targeting Cancer through Insulin Inhibition

When approaching cancer, much of the valuable work in cell biology tended

to downplay the biochemistry at the upstream-stimulus level—namely,

what you eat and its immediate biologic consequences. The major goal had

been to characterize the individual components in the inner working of

the cell and to search for those components that were specifically malfunc¬

tioning in pathological states. These agents—primarily proteins and the

associated genes from which they were expressed—could then be targeted

with drugs, either directly or indirectly at the genomic level. With the

important observation that calorie restriction could ameliorate or prevent

cancer, a link with obesity was established, and in some way excess calories

became interchangeable with weight gain. The obesity-cancer link became

a serial link and it was assumed—as it is frequently assumed in nutrition

that preventing obesity was part of preventing associated pathologies. As

a result of this framing, successes of carbohydrate restriction in improving

cardiovascular risk factors, for example, have frequently been dismissed

as due to the attendant weight loss. This, despite the evidence that the

improvements in risk factors or other outcomes persist even in the absence

of weight loss, or even when benefits were demonstrated in eucaloric trials.

The rationale of the research, then, was that calorie restriction—the

recognized approach to obesity—would point to those intracellular

components that could be targeted for drug development. In many cases

this was a conceptual error. In nutrition it is likely that the doctrine of “a

calorie is a calorie” is the single greatest impediment to understanding.

Otherwise sophisticated and informative experiments were compromised

by the identification of obesity with excess calories, and the failure to look

beyond this effect—by the failure to ask how the separate nutrients, carbo¬

hydrate, fat, and protein, individually affect cellular metabolism, and how

these effects interact (there is, after all, no calorie receptor). Similarly, as I

mentioned above, I and other workers in carbohydrate restriction failed to

see how important it was to look at downstream cellular signaling.

Cancer

257

Fines study treated ten seriously ill cancer patients with low-

carbohydrate ketogenic diets and showed that it was a safe and feasible

regimen for such patients.The rationale followed from the fact that rapidly

growing tumors have a requirement for glucose, but simply reducing

carbohydrates to give the host an advantage was unlikely to be effective

because blood glucose is regulated to stay fairly constant and the cancers

are good at getting whatever glucose is there. Tumors overexpress GLUTl

protein, the non-insulin-dependent glucose receptors. The odds were

therefore strongly stacked against us when it came to effectively depriv¬

ing the cancer of glucose. The question was whether ketone bodies, the

specific consequence of very low-carbohydrate intake, might themselves

exert control over the cancer cells.

Ketosis, the state associated with very low-energy or very low-

carbohydrate intake, held some promise. Fine hypothesized that if we

think of cancer in terms of genetics, we could think of cancer cells as

having evolved throughout the life of the individual, an individual whose

systemic environment, in a modern setting, would be unlikely to experi¬

ence any significant level of ketosis—and that would therefore be unlikely

to provide any selective pressure for the adaptation to use of ketone bodies

as a fuel source. The host, on the other hand, was well adapted to this

substrate: It is unlikely that our ancestors regularly had three square meals

a day. Some fraction of cancer lines, then, might not deal well with a

ketotic environment.

In the experiment it turned out that those patients who became stable

or showed partial remission had the highest level of ketone bodies. Figure

19 .\A shows individual time points with different symbols for each

patient. As expected, there was a correlation between ketone bodies and

insulin levels.

Although this was a small sample, figure 19.15 shows that while some

of the patients had progressive disease, some became stable or showed

partial remission. The level of ketone bodies was the best predictor of

this outcome. Figure 19.1 C shows, on the other hand, that improvements

did not depend on calories or weight loss. Patients who demonstrated

stable disease or partial remission had a threefold higher average ketotic

response compared to those with continued progressive disease. However,

both groups showed similar calorie deficits, or, as shown, degree of weight

p-OH-butyrate (diet/baseline] TO p-OH-butyrate (diet/baseline)

258

Nutrition in Crisis

40'

35 T

0.2

0.4

0.6 0.8 1.0

[insulin] (diet/baseline)

1.2

1.4

Stable Disease/ Progressive

Partial Remission Disease

g 60

(/)

l/>

a> 40

CT3

£ 10

Stable Disease/ Progressive

Partial Remission Disease

Figure 19.1. A f Ketonemia versus insulinemia: The lowest insulinemia correlated with

the highest ketonemia levels. Different symbols represent values for each patient.

B, Ketosis versus disease progression: Patients who demonstrated stable disease or

partial remission had much greater ketotic response compared to those with progres¬

sive disease. C, Calorie deficit versus outcome: The stable disease/partial remission

and progressive disease groups showed no difference in calorie reduction.

Cancer

259

loss—suggesting that the well-established benefits in caloric restriction

reflect an underlying mechanism beyond the energy itself.

Nailing the Link

A small study in advanced cancer patients might be considered only

a minor step forward, but in fact it ties into a vast area of research on

the downstream signaling in both cancer and normal cells—that is, the

changes in the cell following stimulation by an external food or hormones.

Of particular interest is the well-known effect of calorie restriction. It was

widely understood that dietary calorie restriction would have a therapeutic

effect on animals with cancer, and in addition, that reducing calories was

the only way to prolonglife in animals. Studies following this idea showed

that it was an insulin pathway that was involved. Eugene Fines study

nailed the link for us. It is an encouraging situation, because if this is the

worst of times in nutrition, it is also the golden age of biology. Bringing all

of the science to bear on the problem has great promise indeed.

-CHAPTER 20-

The Future of Nutrition

T he low-carbohydrate revolution of 2002 was precipitated by

the popular exposes on the low-fat-diet-heart hypothesis and

its consistent failure to find support in large clinical trials. The

well-armed forces mustered by the diet-heart hypothesis fared poorly in

the early confrontations with the actual experimental evidence. It was a

case of failure to accept failure, even as authorities tried to generalize the

proposed harm of dietary fat to obesity and diabetes. Scientific devel¬

opments, however, have continued to reinforce the idea that control of

metabolism by insulin and other hormones is the key factor in weight loss,

diabetes, metabolic syndrome, and—now—perhaps even cancer. Classic,

well-controlled experiments have nailed the idea: Carbohydrate restriction

is the best therapy for all of the features of metabolic syndrome and the

default treatment for diabetes, consistent with its nature as a disease of

carbohydrate intolerance.

At the same time, underwhelming tests of the diet-heart hypothesis

continue to be conducted, despite the base of evidence to the contrary.

Support comes from meta-analyses and epidemiological statistical juggling,

whose minimal positive outcomes—again, an odds ratio of 1.5 seems even

worse if described as 60:40—are just as damning as the well-documented

failures. Bias against publishing studies of low-carbohydrate diets seems as

strong as ever. A low-carbohydrate paper submitted to the New England

Journal of Medicine, the BMJ, or other major journals will be treated with

superficial politeness if you re lucky, or—more likely—palpable disdain.

For scientists and consumers alike, the situation has become more

confusing, more ambiguous. A major factor appears to be a near disin¬

tegration of standards. Epidemiologic studies trying to show the risks in

saturated fat (good fats-bad fats), in red meat, in sugar (good carbs-bad

carbs) continue to proliferate. In the absence of critical peer review, they

262

Nutrition in Crisis

are accepted for publication, and their conclusions are presented at face

value in the media, and notably in the online medical publications from

which physicians receive much of their information. The same journals

and popular medical sites publish papers describing how bad the obesity

epidemic is, and directly, or by implication, blame the patient. Ironically

these journals also publish articles lamenting the very breakdown in stan¬

dards in the medical literature to which they are the primary contributors. 1

The effects of carbohydrate per se are generally ignored. If they are

mentioned at all, it is always to point out the dangers of “refined” carbohy¬

drate. Wild, exaggerated dangers are attributed to sugar and high-fructose

corn syrup, which are misguidedly separated from carbohydrates at large,

and the lipophobes seem to have maintained control of the market:

Low-fat versions of nearly every product, even “half and half,” are pushed

in the supermarkets. Yet, at the same time, there is a sense that it is all over.

The accumulating failures of low-fat and the popular books and articles

exposing those failures are finally taking their toll. The publication in 2014

of Nina Teicholz’s The Big Fat Surprise might well be the second low-

carbohydrate revolution’s Common Sense, but whatever ultimately deals

the death blow, it is clear that the nutritional establishment has cracked.

Squabbling among the proponents of low-fat is a sure sign.

What’s Next?

It is, perhaps, the end of the beginning.

—Winston Churchill

It is clear that low-fat is dead. Burying it seems like a good idea. Continual

failures of large trials, in combination with the success of alternative

approaches, especially control of metabolism through the glucose-insulin

axis, makes a new method of thinking inevitable.

Surprisingly, the key scientific focus of the second low-carbohydrate

revolution might be cancer. It has long been recognized that total caloric

restriction, at least in animals, is of benefit in slowing progression of

disease and extending life. This has led to the questionable paradigm

of identifying calories with obesity and identifying obesity, in turn, as a

stimulus for pathological effects. One problem is that calorie restriction,

The Future of Nutrition

263

as implemented, generally involves intentional or de facto reduction in

carbohydrate. Moreover, calorie restriction is too broad a term to provide

much insight on mechanism, whereas we have substantial understanding

of the biochemical effects of individual macronutrients, and especially the

role of insulin.

Research in carbohydrate restriction has probably underestimated the

importance and value of detailed downstream stimulus—response coupling

in cells, while cell biology has paid insufficient attention to the nature of

upstream signaling—the effect of diet. It is not calories but carbohydrate

that stimulates insulin release, and obesity is largely a response, not a stim¬

ulus. Ketone bodies, too, are vital to our evolving understanding. Originally

considered primarily a marker for fat breakdown, ketone bodies are now

understood to be a more global cell signal. The possibility of cancer treat¬

ment based on the theory that tumors might be more poorly adapted to

use ketone bodies as an energy source because of their evolution during the

life span of the individual is one specific approach tested in Eugene Fine’s

small pilot study. 2

More generally, the use of carbohydrate restriction as a cancer therapy

might be the important battleground in the search for effective dietary

treatment and prevention. The new scientific paradigm might also be

better received in the area of oncology due to the failures of other methods

to contain the disease. With regard to obesity, diabetes, and metabolic

diseases, it may be necessary to clear out the backlog of biased, unscientific,

and statistically flawed studies that have so far impeded progress. New

standards will have to be implemented to improve the future literature.

Scientific truth is its own justification, but in this case, relief of human

suffering is what is truly important to society at large. Progress may be slow

but the crisis is real and the stakes are high.

ACKNOWLEDGMENTS

T wo people, Monika Hendry and Paula Nedved, have made this

book possible. Their continual encouragement, their probing

questions, and their excellent proofreading abilities kept the book

alive through periods during which there were many doubts and discour-

agements. I am also grateful to Professor Wendy Pogozelski for her help

with chapter 2 and for her enthusiasm about the final product.

Insofar as this book is scientifically accurate it is due to my early influ¬

ences, particularly Thomas C. Detwiler, and later interactions with my

colleagues, Doctors Eugene J. Fine, Frederick Sacks JefFS. Volek, Eric C.

Westman, Jay Wortman, Steve Phinney, Ann Childers, and many others

who have for so long served as loyal opposition to the medical and nutri¬

tional monarchy.

Thanks also to members of the Nutrition and Metabolism Society and

Facebook friends, as well as the greater social circle, Gary Taubes, Nina

Teicholz, Carole Friend, Judy Barnes Baker, and especially the pseud¬

onymous Amanda B. Wreckondwith, who have provided continuous

encouragement and endured a certain degree of curmudgeonly behavior.

I am grateful to the State University of New York Downstate Medical

Center, which has given me the freedom to work and think outside the box.

NOTES

Introduction

1. W. Pogozelski, N. Arpaia, and S. Priore, “The Metabolic Effects of Low-Carbohydrate

Diets and Incorporation into a Biochemistry Course,” Biochemistry and Molecular

Biology Education 33 (2005): 91-100; R. D. Feinman and M. Makowske, “Metabolic

Syndrome and Low-Carbohydrate Ketogenic Diets in the Medical School

Biochemistry Curriculum/’ Metabolic Syndrome and Related Disorders 1 (2003):

189-98; M. Makowske and R. D. Feinman, “Nutrition Education: A Questionnaire

for Assessment and Teaching,” Nutrition Journal 4, no. 1 (2005): 2.

2. F. B. Hu et al., “Dietary Fat Intake and the Risk of Coronary Heart Disease in Women,”

New England Journal of Medicine 337, no. 21 (1997): 1491-99.

3. M. Pollan, In Defense of Food: An Eaters Manifesto (New York: Penguin Press, 2008).

4. M. U. Jakobsen et al., Major Types of Dietary Fat and Risk of Coronary Heart Disease:

A Pooled Analysis of 11 Cohort Studies,” American Journal of Clinical Nutrition 89,

no. 5 (2009): 1425—32; P. W. Siri-Tarino et al., “Meta Analysis of Prospective Cohort

Studies Evaluating the Association of Saturated Fat with Cardiovascular Disease,”

American Journal of Clinical Nutrition 91, no. 3 (2010): 535-46; P. W. Siri-Tarino et

al., Saturated Fat, Carbohydrate, and Cardiovascular Disease,” American Journal of

Clinical Nutrition 91, no. 3 (2010): 502-9.

5. American Diabetes Association, “Nutrition Recommendations and Interventions for

Diabetes 2008 J Diabetes Care 3 1 , Supplement 1 (2008): S61-78.

6. A. Rabinovich, The Yom Kippur War: The Epic Encounter That Transformed the Middle East

(New York: Schocken Books, 2004), 56.

7. American Diabetes Association, “Nutrition Recommendations and Interventions for

Diabetes 2013 J Diabetes Care 36, supplement 1 (2013): S12-32.

8. N.Teicholz, The Big Fat Surprise: Why Butter, Meat & Cheese Belong in a Health Diet{ New

York: Simon Sc Schuster, 2014).

9. G. M. Reaven, “Role of Insulin Resistance in Human Disease,” Diabetes 37 (1988):

1595-607.

10. E. J. Fine, C. J. Segal-Isaacson, R. D. Feinman et al., “Targeting Insulin Inhibition as a

Metabolic Therapy in Advanced Cancer: A Pilot Safety and Feasibility Dietary Trial in

10 Patients,” Nutrition 23, no. 10 (2012): 1028-35.

11. W. B. Kannel and T. Gordon, “Diet and Regulation of Serum Cholesterol, Section 25,”

in The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease

(Washington, DC: National Heart, Lung, and Blood Institute, 1970).

268

Nutrition in Crisis

Chapter 1: Handling the Crisis

1. F. Hahn, M. R. Fades, and M. D. Eades, The Slow Burn Fitness Revolution: The Slow

Motion Exercise That Will Change Your Body in 30 Minutes a Week (New York: Broadway

Books, 2003).

2. E. C. Westman, S. D. Phinney, and J. Volek, The New Atkins for a New You: The Ultimate

Diet for Shedding Weight and Feeling Great Forever (New York: Simon & Schuster, 2010).

3. M. R. Eades and M. D. Eades, Protein Power (New York: Bantam Books, 1996).

4. J.S. Volek and S.D.Phinney, The Art and Science of Low-Carbohydrate Living (Charleston,

SC: Beyond Obesity, 2011).

5. R. D. Feinman, M. C. Vernon, and E. C. Westman, “Low Carbohydrate Diets in Family

Practice: What Can We Learn from an Internet-Based Support Group,” Nutrition

Journal 5 (2006): 26.

6. S. Somers, Eat Greats Lose Weight (New York: Crown, 1997).

7. R. C. Atkins, Dr Atkins New Diet Revolution (New York: Avon Books, 2002).

8. R. C. Atkins, New Diet Revolution .

Chapter 2: Whaddaya Know?

1. American Diabetes Association, “Nutrition Recommendations and Interventions for

Diabetes 2008.”

2. American Diabetes Association, “Nutrition Recommendations and Interventions for

Diabetes 2013.”

3. American Diabetes Association, “Standards of Medical Care in Diabetes 2012,

Diabetes Care 35, supplement 1 (2012): Sll-63.

4. American Diabetes Association, “Standards of Medical Care in Diabetes 2010,

Diabetes Care 33, supplement 1 (2010): SI 1—61.

5. A. Keys, “Diet and Blood Cholesterol in Population Survey: Lessons from Analysis of

the Data from a Major Survey in Israel,” American Journal of Clinical Nutrition 48, no. 5

(1988): 1161-65.

6. U. Raynskov, The Cholesterol Myths: Exposing the Fallacy That Cholesterol and Saturated

Fat Cause Heart Disease (Washington, DC: NewTrends Publishing, 2000); G. Taubes,

Good Calories, Bad Calories (New York: Alfred A. Knopf, 2007); A. Colpo, The Great

Cholesterol Con (Morrisville, NC: Lulu Press, 2006).

7. A. Keys, “Coronary Heart Disease in Seven Countries ,'”Annals of Internal Medicine 73, no.

2 (1970): 356.

8. N. Teicholz, The Big Fat Surprise ,; U. Raynskov, The Cholesterol Myths , G. Taubes, Good

Calories, Bad Calories; G. Taubes, “The Soft Science ofDietary Tat,” Science 291 (2001):

2536-45; U. Ravnskov et al., “LDL-C Does Not Cause Cardiovascular Disease:

A Comprehensive Review of the Current Literature,” Expert Review of Clinical

Pharmacology 11, no. 10: 959—970.

9. K. Sarri and A. Kafatos,“The Seven Countries Study in Crete: Olive Oil, Mediterranean

Diet or Fasting?,” Public Health Nutrition 8, no. 6 (2005): 666.

10. N. Teicholz, The Big Fat Surprise .

11. C. E. Forsythe et al., “Limited Effect ofDietary Saturated Fat on Plasma Saturated Fat

the Cor

jOW Carbohydrate Diet,” Lipids 45, no-10 (2010): 947-62: C. E.

Notes

269

12 "i'^'“"‘ h 4?1 ""■ 5,2fJ i 30?-" 1 ' 10 ""’ 8 '“° J M '" b ' J “

13 fTh T” 4 ' ’ Un ' mJ °“"’ l °f c ‘"‘<°l°£y'H,M

Chapter 3- The F' T DiS6aSe in Wome «-”

s<W,i 992) *’“* J, " OT “(N™y„ rk .. Sin , on&

4 rr WDietR£v ° lM ion. J ° Vanovi ch,1978). *

' ^ — — «, Momingitar

s' m 'Z oll T JnDefemeofFood -

10. G. D. Foster et al. “Low-C k l"? ^ 2 ° 82 - 90 -

11- P- L eren, “The Oslo Diet-Heart StudyW]^ 161 v" ^^ny-”

935-42; P. Leren et a]., “The Oslo Study- CHDft "jfj? 0 "’” 42 > no. 5 (1970;-

13 ^ ^ C0r0 ^ Disease,” CV* 28

jo™: change 7 *« ^e

ZZcJI 5 ’ n °' ! (2006); 39 ' 49; * * Howard ^ Me dkal

£sk of Cardtovascuiar Disease: The Women’s Healrt T - Dieta V Pattem and

657. ^ Cati0n TtU ’J^of iheAm ^MZZ Rand ° mized trolled

D' ^° S t^ n0pal ^^ r ^^^^^ 1 ®' a ^’ni«^^Hedtli 1I I^ Kabetes Meflitus'

” Ler i>asic NufnV;^«. at

y «*« ivieaicint

Chapter 4: Basic Nutrition- 7 U,

1. M.Eades m P R j 771 -pj ‘ Mac ronutnents

- anZ c,„„ &Gr ^, m

270

Nutrition in Crisis

4. N.Teicholz, The Big Fat Surprise. rvF!ltan d Risk of Coronary Heart Disease:

5. M-UJakobsen et X of Clinical Nutrition 89,

A Pooled Analysis of 1 ° ^ ^ “Meta-analysis of Prospective Cohort

no. 5 (2009): 1425-32; P. ■ in c atura t e d Fat with Cardiovascular Disease/

Studies Evaluating the Association o , 535-46; P. W. Siri-Tarino et

Glinka!Nutrition 91, no. 3 (2010): 502-9. ^wCarbohydrateDietandaCalorie-

6, ta Healthy

Restricted Low Fa, Die, on 1617-23.

Woma; Journal of CUnmlEnrktnm ogy a a Carbohydrate-fiee

7. S. Borghjid and R. D. F=i»=W^ Mt6n mJ A .scent*

Die,,' ^ Din A „ ociated with Sevete Gluco.e

“Marked Hyperleptinemia After irlig ,™ no 4 (1998): 461-67;

Intolerance in Mice,’ Eurof.JKJonrn.l ® ^ A ' Modcl fo, Studying

M.S,Wi»adl.»dB,Ahten,-U.Htgh-F.,Dt d ft 2 D i, b e,.»-

Mechanisms and Treatment of Imparted Glucose

Trr'

8 . S.Klein and R.R. Won, '-‘U y - ( 1992 ): E631-36.

Fasting,” American.Journal of Physio ogy > • C A: Lange Medical

9. H. A. Harper, Review of Physiological Chemistry, 8th ed. (Los Altos,

Publications, 1961). h d (Los Altos, CA: Lange

10. D. W. Martin et al., Harpers Rev t ev> ofBjochermstry, 20th

Medical Publications, 1985).

2 f'g'Y oung^'Claude ’Bernard and the Discovery of Glycogen: A Century of Retrospect,"

668-75.

Chapter 6 : Sugar, Not Glucose-Sweetened,

'■ L„g“n«»“ Vfeetal Adiposity »d Lipid, and

Metabolism in the Liver

" Co„tw“ »1 Lttpmeu, of Metabolic Syndrome,- —

3. RDrebiman and E.J. Fine,"Fructose in Perspective,” Alwfriften andMotabolism

‘ rpThubes^and C- K. Couzens,“Big Sugar’s Sweet Little Lies," Mother Jonos, 2012,

Notes

271

Chapter 7: Saturated Fat: On Your Plate or in Your Blood?

1. C E. Forsythe et al “Limited Effect of Dietary Saturated Fat on Plasma Saturated Fat

Ee ^ ont “ t of a Low-Carbohydrate Diet "Lipids 45, no. 10 (2010): 947-62

2. C. E Forsythe et al., “Comparison of Low Fat and Low Carbohydrate Diets on

i Fatt ^ Aad Composition and Markers of Inflammation,” Lipids 41

no. ! (20 8). 65-77; J. S. Volek et al„ “Dietary Carbohydrate Restricts Educes a

mque Metabohc State Positively Affecting Artherogenic Dyslipidemia, Fatty Acid

S y ndrome ’” jPro ^ in Lipid Research 47, no. 5 (2008):

the Met h V S 6k f ’ ^ ° hydrate Restriction Has a More Favorable Impact on

the Metabolic Syndrome Than a Low Fat Diet Lipids 44, no. 4 (2009): 297-309

3. C. E. Forsythe et al., “Limited Effect of Dietary Saturated Fat ”

4. S. K. Raate et al., “Total Fat Intake Modifies Plasma Fatty Acid Composition in Humans ”

Journal of Nutrition 131, no. 2 (2001): 231-34. ’

Chapter 8: Hunger: What It Is, What to Do About It

1. B- Behaviorism (New York: Knopf, distributed by Random House 1974)

Fffi F ~. and E 'J' Fine > “Non-equilibrium Thermodynamics and Energy

Efficiency in Weight Loss Diets,” Theoretical Biology and Medical Modelling 4 (2007): 27

“ A CaW 18 3 “ orie ’• to

TO* trBX%7w£ r ’ ° rd " " d “““ L “ ° f ^ Enrow (Nm

2 C/^^/aT a1 '' ' Di ^ Comp ° s iti°n an d Energy Balance in Humans, Journal of

Clinical Nutrition 67, no. 3 (1998): 551S-55S. J

Chapter 10: Diabetes

1. E_C. Westman and M. C. Vernon, “Has Carbohydrate-Restriction Been Forgotten as

rea ment or Diabetes Mellitus? A Perspective on the ACCORD Study Design ”

Nutrition and Metabolism 5 (2008): 10. 6 ’

2 ' Memr"T p W ' S ' Y r Cy ; J '-' “ d “■ H "” Phre?! ' »f Dinbere,

“ W) 77-8 e 3 (1914 -‘ ,22 f p “^«- in «,

3. D. J. Jenki™ et d "Effect of, Low-GIyconic Index or , High-Cereal Fiber Die, on

4 f L C TT man 6 t *!i’ EffeCt ° f a L ° W “ Carbohydrate > Ketogenic Diet Versus

a Low-Glycemic Index Diet on Glycemic Control in Type 2 Diabetes Mellitus ”

Nutrition and Metabolism 5 (2008): 36.

5. W. S. Yancy et al., “A Low-Carbohydrate, Ketogenic Diet to Treat Type 2 Diabetes ”

Nutrition and Metabolism 2 (2005): 34. F J

1' )l' St, 7 t f n, l a ^'’ 1 fffa ) D S5 0:Risk F ai;toreforI n eidenceandP,ogressionofR£dnoparhy

m Type II Diabetes over 6 Years from Diagnosis,” Diabetologia 44, no. 2 (2001): 156-63

272

Nutrition in Crisis

Chapter 11: Metabolic Syndrome: The Big Pitch

1 . G. M. Reaven, “Role of Insulin Resistance in Human Disease,” Diabetes 37 (1988):

1595-607; G. Reaven, “Syndrome X: 10 Years After,” Drugs 58, no. 1 (1999): 19-20.

2. G. M. Reaven,” Role of Insulin Resistance”; G. M. Reaven, “Tie Metabolic Syndrome:

Requiescat in Pace,” Clinical Chemistry 51, no. 6 (2005): 931—38.

3 . J. S. Volek and R. D. Feinman, “Carbohydrate Restriction Improves the Features of

Metabolic Syndrome. Metabolic Syndrome May Be Defined by the Response to

Carbohydrate Restriction,” Nutrition and Metabolism 2 (2005): 31.

Chapter 12: The Medical Literature: A Guide to Flawed Studies

1 . M. C. Gannon and F. Q^Nuttall, “Control of Blood Glucose in Type 2 Diabetes without

Weight Loss by Modification of Diet Composition "Nutrition and Metabolism 3 (2006): 16.

2. R. D. Feinman and E. J. Fines, “Perspective on Fructose.”

3 . G. A. Bray et al., “Effects of Dietary Protein Contention, Weight Gain, Energy

Expenditure, and Body Composition During Overeating: A Randomized Controlled

Trial," Journal of the American Medical Association 307, no. 1 (2012): 47-55.

4. G. R. Norman and D. L. Steiner, PDQ Statistics, 3rd ed. (Hamilton, Ontario: B. C.

Decker, 2003).

5. D, Colquhoun, Lectures on Biostatistics (London: Oxford University Press, 1971), http://

www.dcscience.net/Lectures_on_biostatistics-ocr4.pdf.

6 . H. Wainer, Medical Illuminations (Oxford: Oxford University Press, 2014).

7 . L. J. Appel et al., “Effects of Protein, Monounsaturated Fat, and Carbohydrate Intake

on Blood Pressure and Serum Lipids: Results of the OmniHeart Randomized Trial,

Journal of the American Medical Association 294, no. 19 (2005): 2455-64.

Chapter 13: Observational Studies, Association, Causality

1. S. Weinberg, Dreams of a Final Theory , 1st ed. (New York: Pantheon Books, 1992).

2. S. Mukherjee, The Emperor of All Maladies: A Biography of Cancer (Waterville, ME:

Thorndike Press, 2010).

3. J. LeFanu, The Rise and Fall of Modern Medicine (New York: Carroll & Graf, 1999).

4. J. LeFanu, The Rise and Fall of Modern Medicine .

5. L. E. Cahill et al., “Prospective Study of Breakfast Eating and Incident Coronary Heart

Disease in a Cohort of Male US Health Professionals,” Circulations 128, no. 4 (2013): 337-43.

6 . S. Mukherjee, The Emperor of All Maladies , 246.

Chapter 14: Red Meat and the New Puritans

1. M. Pollan, In Defense of Food.

2. R. Stein,“Daily RedMeat Raises Chances ofDying Early,March24,2009.

3. R. Sinha et al., “Meat Intake and Mortality: A Prospective Study of Over Haifa Million

People,” Archives of Internal Medicine 169, no. 6 (2009): 562-71.

4. D. K. Layman et al., “A Reduced Ratio of Dietary Carbohydrate to Protein Improves

Body Composition and Blood Lipid Profiles during Weight Loss in Adult Women,”

Journal of Nutrition 133, no. 2 (2003): 411-17; D. K. Layman et al., “Dietary Protein

and Exercise Have Additive Effects on Body Composition during Weight Loss in

Notes

273

Adult Women, 1 "Journal of Nutrition 135, no. 8 (2005): 1903-10; M. P. Thorpe et al., “A

in Protein, Dairy, and Calcium Attenuates Bone Loss over Twelve Months

of Weight Loss and Maintenance Relative to a Conventional High-Carbohydrate

Diet in Adults,’ "Journal of Nutrition 138, no, 6 (2008): 1096-100; P.J. Douglas et al,

“Protein, Weight Management, and Satiety,” The American Journal of Clinical Nutrition

87, no. 5 (2008): 1558S-61S; P. J. Douglas et al., “Role of Dietary Protein in the

Sarcopenia of Aging,” The American Journal of Clinical Nutrition 87, no 5 (2008)'

11562S-66S.

5. R. Sinha et al., “Meat Intake and Mortality.”

6. A. Pan et al., “Red Meat Consumption and Risk ofType 2 Diabetes: 3 Cohorts of US

Adults and an Updated Meta-analysis,” The American Journal of Clinical Nutrition 94

no. 4 (2011): 1088-96.

7. A. Pan et al., “Changes in Red Meat Consumption and Subsequent Risk ofType 2

Diabetes Mellitus: Three Cohorts of US Men and Women,” Journal of the American

Medical Association Internal Medicine (2013): 1-8.

8. R. D. Feinman, “Red Meat and Type 2 Diabetes Mellitus J Journal of the American

Medical Association Internal Medicine 174, no. 4 (2014): 646.

Chapter 15: The Seventh Egg: When Studies Defy Common Sense

1. S. Ebrahm et al., “Shaving, Coronary Heart Disease, and Stroke: The Caerphilly Study,”

American Epidemiology 157, no. 3 (2003): 234-38.

2. L. Djousse et al., Egg Consumption and Risk ofType 2 Diabetes in Men and Women,”

Diabetes Care 32, no. 2 (2009): 295-300.

Chapter 16: Intention-to-Treat: What It Is and Why You Should Care

1. D. Newell, “Intention-to-Treat Analysis: Implications for Quantitative and Qualitative

Research,” InternationalJournal ofEpidemiology 21, no. 5 (1992): 837-41.

2. S. Hollis and F. Campbell, “What Is Meant by Intention to Treat Analysis? Survey of

Published Randomized Controlled Trials,” AM/319, no. 7211 (1999): 670-74.

3. G. D. Foster et al., “Weight and Metabolic Outcomes after 2 Years on a Low-

Carbohydrate versus Low-Fat Diet: A Randomized Trial,” Annals of Internal

Medicine 153, no. 3(2010): 147-57.

4. Comparison of Current NSLP Elementary Meals vs. Proposed Elementary Meals,”

Hunger-Free Kids Act of 2010, The White House, 2010, https://obamawhitehouse

.archives.gov/sites/default/files/cnr_chart.pdf.

Chapter 17: The Fiend That Lies Like Truth

1. H. Wainer, Medical Illuminations (Oxford: Oxford University Press, 2014).

2. M. C. Gannon and F. Q^Nuttall, “Control of Blood Glucose in Type 2 Diabetes.”

3. V. Hainer et al., “ATwin Study ofWeight Loss and Metabolic Efficiency,’ "International

Journal of Obesity and Related Metabolic Disorders 25, no. 4 (2001): 533-37.

4. G. C. Smith and J. P. Pell, “Parachute Use to Prevent Death and Major Trauma Related to

Gravitational Challenge: Systematic Review of Randomized Controlled Trails,” BMJ

327, no. 7429 (2003): 1459-61.

274

Nutrition in Crisis

Chapter 18: Nutrition in Crisis

1. U. Ravnskov, The Cholesterol Myth: Exposing the Fallacy That Cholesterol and Saturated Fat

Cause Heart Disease (Washington, DC: New Trends Publishing, 2000); U. Ravnskov,

“Cholesterol: Friend or Foe?”

2. R. D. Feinman and J. S. Volek, “Carbohydrate Restriction as the Default Treatment for

Type 2 Diabetes and Metabolic Syndrome,” Scandinavian Cardiovascular Journal 42,

no. 4 (2008): 256-63.

3. A. Dahlqvists, LCHF-blogg (blog), 2014.

4. H.Monery, “Dr. Gary Fettke’s‘Officially Cautioned,’” The Examiner,November 13,2016.

5. W. Willett, personal communication to the author.

Chapter 19: Cancer: A New Frontier for Low-Carbohydrate

1. E. J. Fine, R. D. Feinman et al., “Targeting Insulin Inhibition as a Metabolic Therapy

in Advanced Cancer: A Pilot Safety and Feasibility Dietary Trial in 10 Patients,

Nutrition 28, no. 10 (2012): 1028—35.

Chapter 20: The Future of Nutrition

1. J. P. Ioannidis, “Meta-Research: The Art of Getting It Wrong,” Research Synthesis

Methods 1 (2010): 169-84; J. P. Ioannidis, A. Tatsioni, and F. B. Karassa, “Who Is

Afraid of Reviewers’Comments? Or, Why Anything Can Be Published and Anything

Can Be Cited J European Journal of Clinical Investigation 40, no. 4 (2010): 285-87;

R. Horton, “Offline: What Is Medicine’s 5 Stigma?,” The Lancet 385 (2015): 1380;

M. Angell, Science on Trial (New York: W. W. Norton & Co., 1996); R. Feinman and

S. Keough, “Ethics in Medical Research and the Low-Fat Diet-Heart Hypothesis,”

Ethics in Biology, Engineering and Medicine 5, no. 2 (2014): 149—59.

2. E.J. Fine, C.J. Segal-Isaacson, R. D. Feinman et al., “Targeting Insulin Inhibition as a

Metabolic Therapy in Advanced Cancer: A Pilot Safety and Feasibility Dietary Trial in

10 Patients ,”Nutrition 23, no. 10 (2012): 1028—35.

INDEX

Note: Page numbers followed by “f” refer to figures. Page numbers followed by “t” refer

to tables.

abdominal fat, 119

absolute risk, 191,202,205,206,209,

211-12

absolute zero, 138,141

Academy of Nutrition and Dietetics, 68

acetoacetate, 101,104

acetyl-coenzyme A, 76,97-100,101,

103-4,120

acetyldehyde, 99

Adams, John, 15, 61

Adams, Sam, 61

addiction to sugar, 109

adenosine diphosphate, 97,100

adenosine triphosphate, 97,98,100,146

adipocytes, 148

Adler, Alfred, 31

advanced glycation end-products (AGEs),

94, 96

aging, 94

alcohol, 37, 73-74, 99,108-9

Allen, Woody, 127,189,218

Allison, David, 238

American Diabetes Association, 11,13,

44-45,57,151,161,176

American Heart Association, 8,42,45,52,

57,58,62

on fructose, 107

on saturated fat, 77

on total dietary fat, 58,63

American Medical Association, 62,177

amino acids, 88, 89-90, 93,147,197-98

ammonia, 90

amylopectin, 75

amylose, 75

animal models, 82-83, 91-92,118,154

anorexigenic hormones, 126,127

antioxidant supplements, 224

Appel, L.J., 180

appetite, 126

Art and Science of Low-Carbohydrate

Livings 27

ascorbic acid, 224

association and causality, 185-96,206,

235-36

in carbohydrates and obesity, 40-42

criteria on, 190-94,196

in eggs and diabetes, 215-18,236

in meats and disease, 198

in smoking and lung disease, 185,190,

192,195,196,199,236

atherogenic dyslipidemia, 10,14,118,170

atherosclerosis, 88

athletes, carbohydrate loading in, 75

Atkins, Robert, 32, 61-62,63,156

Atkins diet, 25,27,29, 61-62

breakfast in, 127

cheese in, 32

Foster study on, 64-66, 80-81,225-31

generic status of, 13,29

as high-calorie starvation diet, 85-86,

101-2

ketone bodies in, 29,102

professional resistance to, 61, 62, 63, 65,

69,156,168,228,256

276

Nutrition in Crisis

Atkins diet ( continued )

sugar in, 107

weight loss in, 13,23,65, 66, 67,153,

225

Atkins Nutritionals, 67-68

atomic theory, 3

Australia, 247—50

average values in statistics, 237-39

avocado oil, 55

bacon, 1,168,210

bacteria, glycolytic, 98-99

beef tallow, 55,60

bell-shaped curves, 237

Bergman, Ingmar, 218

Bernard, Claude, 91-94,153-54

Bernstein, Carl, 199

Bernstein, Richard, 163

beta carotene, 224

beta cells of pancreas, 42,159,216

Big Fat Surprise (Teicholz), 13,55, 63, 78,

250-51,262

biochemistry, 3-4, 9,10,15

black box approach to metabolism, 94—96,

Bloomberg, Michael, 105

BMJ> 250

body mass, 40,155-56

Boxer, Barbara, 1

brain, 43,76,101,102

Bray, George, 177

bread, 38-39,47,75,106

breakfast, 1,2,124,127-28,191

Brehm, Bonnie, 80, 81, 84

Brillat-Savarin, J. A., 61

butter, 38—39,47,77,118

Cahill, George, 85,101-2

“calorie is a calorie” idea, 23,32,38,84-85,

126,133,138,177

in cancer, 256

and Hesss law, 144-46

calories, 21,23, 37—38,39,40, 60

in carbohydrates, 37,42,59, 84-85, 86,

187,200

consumption trends, 40,41f, 73

in fat, 5,37-38,57,133

in low-carbohydrate diets, 25, 32,33t,

66 , 81, 84,135

in low-fat diets, 25,30, 66, 81, 84,119,

135

reduction in intake of, 21,168,262-63

and thermodynamics, 84—85,131-48

calories in, calories out (CICO), 142

calorimeters, 37-38

Campbell, F., 223

cancer, 172,191,253-59

insulin in, 15,254—55,256-59

ketone bodies in, 257,258f, 263

low-carbohydrate diets in, 15,253-59,

263

of lung, 185,189-90,192,196

and meat consumption, 201,204

and smoking, 185,190,192,195,196,

199

Warburg effect in, 100,254

candy, 30,105,165,186-87

Cannon, Geoffrey, 54

canola oil, 55—56

caproic acid, 76

carbohydrates, dietary, 73-76

accuracy of dietary records on, 82, 84,

135-36,227

ADA on, 13,44-45,151,176

calories in, 37,42,59,84-85,86,187,200

and cardiovascular disease, 6—8,58-59,

60,70,158-59

catalytic effect of, 118

consumption trends, 40,59,73,116

conversion to fat, 88

in DASH diet, 180-81

in diabetes, 151—66. See also diabetes

mellitus

as fattening, 84—85, 86-89

Index

277

feedback response to, 14

and glucose levels, 2,21-22,42-43,59,

151,155,164-65

glycemic index of, 46-48,59, 83,

156-58

in high-carbohydrate diets. See high-

carbohydrate diets

insulin response to, 10,23,79, 80, 82,

164,251,263

intolerance of, 21-22,42-43,152,159,

216,261

and lipoprotein levels, 70

in low-carbohydrate diets. See

low-carbohydrate diets

in low-fat diets, 118

metabolism of, 86, 88,110-13

nonsatiating effect of, 123

in obesity, 40-42,59, 83,107,113,153

recommended daily intake, 33t, 42,

44-45,59

refined, 83,262

replacing fat with, 7, 8,58-59,60

requirement for, 43,59,79,161

and saturated fat in blood, 57, 88-89,

121-22,170

sugar as. See sugars

and thermic effect of feeding, 141

and triglyceride levels, 56-57,65,108,

111,120,159,170,228-31

carbon, 143-44

carbon dioxide, 95, 99

carbon monoxide, 143-44,145

carbophores, 61

cardiovascular disease, 10,191,193

antioxidants in, 224

and carbohydrates, 6-8,58-59, 60, 70,

158-59

cholesterol in, 4,5,15-16,48,56-58,

63,70,159

in diabetes, 159

diet-heart hypothesis on. See diet-heart

hypothesis

and fats, 4-9,13,15-16,30,48, 51-54,

62-64,69-70, 83,113,118,158-59,

193,252,261

Foster study on, 64-66,225-31

Framingham Study on, 15-16, 62-63,

69,193

lipid markers of, 10,14,48-49

and meat consumption, 201

and metabolic syndrome, 14,167,170

NHS study on, 6-8,58-59

observational studies on, 188

Seven Countries Study on, 53-55

WHI study on, 64, 69,193

carnivore diet, 79

cauliflower, 29

causality and association. See association

and causality

Center for Science in the Public Interest,

55,250,251

central nervous system, 101,102

cheese, 32,54,106

chemiosmotic theory, 100

chemistry

atomic theory in, 3

of carbohydrates, 73-76

energy in, 2, 96,97,98

of lipids, 76-78

chocolate, 106,123,161,165

cholesterol, 13,48,56-59

and cardiovascular disease, 4,5,15-16,

48,56-58,63,70,159

in low-carbohydrate diets, 65,70,120,

159

in low-fat diets, 56,57-58, 60,69-70,

120

Churchill, Winston, 262

cigarette smoking. See smoking

citric acid cycle, 90,99-100

coconut oil, 52,118

coenzymes, 97

acetyl-CoA, 76, 97-100,101,103-4,

120

278

Nutrition in Crisis

Colbert, Stephen, 55

Colquohon, David, 177-78

Common Sense (Paine), 62,64,262

compliance

and intention-to-treat, 219-32

in low-carbohydrate diets, 24,32,

135—36,225,226-27,231,251

self-reports on, 227

in vitamin E supplementation, 224,230

Consumer Reports, 52

control loss in obesity, 25

Cooksey, Steve, 68,248

corn, 200

corn syrup, high-fructose, 47,105,107,

113,262

coronary artery bypass surgery, 222-23

coronary thrombosis, 190,196

Couzens, Cristin Kearns, 115

cravings, 30-31,32-33

Crete, 53-55

Crick, Francis, 181

Dahlqvist, Annika, 246-47

dairy fat, 121

Dalton, John, 61

DASH diet, 180-81

degenerative disease, 94

delayed-onset muscle soreness, 99

de novo lipogenesis, 57, 78, 88,120—21

depression, 110 _

desserts and sweets, 26,30-31,165

candy, 30,105,165,186-87

chocolate, 106,123,161,165

cravings for, 30-31

diabetes mellitus, 2,10,12—13,151—66

ADA on diet in, 11,13,44—45,151,

161,176

carbohydrate intolerance in, 21-22,

42-43,152,159,216,261

compensatory feedback in, 14

complications of, 11,153,159,164

default diet in, 4,15,21-22,245,261

drug therapy in, 11,22,28,44,45,151,

158

“eat to the meter” in, 22

effectiveness of low-carbohydrate diet

in, 28,42,59,106,113

and eggs, 215-18,236

epidemic of, 40—42,59,66,198,200

fructose in, 113

glucose blood levels in, 2,11,22,42-43,

44,45,103-4

glucose—insulin axis in, 79,153

high-fat diet in, 42

increase in incidence of, 116

individualization of diet in, 13,20-21,

176

insulin production in, 21,42,103,152,

154,159

insulin resistance in, 14,21,42-43,103,

152,154,159,216

insulin therapy in, 152-53

ketoacidosis in, 103-4

latent autoimmune, 160-66

long-term effects of low-carbohydrate

diet in, 44—45

and meat consumption, 213

and metabolic syndrome, 14

metabolism in, 4,43

and obesity, 155—56,187-88

principle of low-carbohydrate diet in, 33

professional resistance to

low-carbohydrate diet in, 10-11,13,

15,22,44,248

type 1. See type 1 diabetes

type 2. See type 2 diabetes

weight loss in, 22,153,155,156,176,235

diet-heart hypothesis, 5-6,62-63

in diabetes, 159

lack of proof for, 9,13,58,60,63,70,

83,235,261

lipid levels in, 48,58

low-fat diet in, 5—6, 9,13,48,261

peer review of, 248

Index

279

digestion, 46,74-75, 89-90,91,125

disaccharides, 73

Djousse, L., 215-18

DNA, 181,254

dog studies, 91-92,154

Doll, Richard, 185,190,195

Don’t Know Much About History , 61

dose-response curve, 192-93

drug therapy

in coronary artery disease, 222-23

in diabetes, 11,22,28,44,45,151,158

in tuberculosis, 189-90

dyslipidemia, atherogenic, 10,14,118,170

Eades, Mary Dan, 32

Eades,Mike, 32,187

eating behavior, 123-30

‘eat to the meter,”22,32

eggs, 121,123,186,200

and diabetes, 215-18,236

Einstein, Albert, 9, 61,178,187

elderly, diet of, 198,200,207

electron transport chain, 99-100

Elliott, Jennifer, 248,250

Emperor of All Maladies, 185,188-89,194,

195

energy

and calories, 21,37-38,39

in carbohydrates, 28,40,45, 84

from chemical reactions, 2, 96,97,98

excess intake in diabetes, 44, 45

in fasting, 86,101

in fat, 37-38, 85,101

from fatty acid oxidation, 87, 88

from glucose, 87,94

as goal of metabolism, 76,96-97

from ketone bodies, 94,100

and metabolic advantage, 132

from protein, 90

restriction in low-fat diet, 80, 81

and thermodynamics, 85,131-48

Enig, Mary, 78

enthalpy, 143,144

entropy, 138-41,143

epidemiologic studies, 194-95,212

accuracy of self-reports in, 227

association and causality in, 187-89,

190,192

on fat in diet, 261-62

on meat in diet, 197,198

equilibrium constant, 96

equilibrium thermodynamics, 146-48

ester bonds, 49

ethanol, 37, 73-74,99,108-9

euricic acid, 56

evidence-based medicine, 194,234-35,

242,248,249

evolution of diet, 77, 111

exercise, 22,26,75,129

extensive variables, 39,47,60

Facebook, 175-76

fasting, 85-86, 87-88, 94

glucose levels in, 85, 86, 87, 88,152,

159

intermittent, 20

protein in, 88, 94

fat, body, 49, 79,88,119,131

accumulation of, 148

conversion of protein to, 90

and eating behavior, 126-27

in fasting, 85,86, 87, 88

and glucose, 98,101,120,154

oxidation of, 80, 84, 85, 86t

role in metabolism, 86

fat, dietary, 45-59,73,261-62

accuracy of dietary records on, 82, 84,

135-36,227

animal models of, 82

association and causality in, 188

calories in, 5,37-38,57,133

and cardiovascular disease, 4-9,13,

15-16,30,48,51-54, 62-64,69-70,

83,113,118,158-59,193,252,261

280

Nutrition in Crisis

fat, dietary ( continued )

composition of, 55-56

consumption trends, 40,41f, 73

in diabetes, 40-42,43

diet-heart hypothesis on, 5-6,9,13,48,

62-63,83,252,261

energy density of, 37-38

epidemiologic studies on, 261-62

and fat in blood, 48,117,119-20

Foster study on, 64—66,68, 80-81,82,

225-31

Framingham Study on, 15-16,29,

62-63,69,193

good and bad, 76-78,226,261

guidelines on, 42

in high-fat diets. See high-fat diets

in low-carbohydrate diet, 25,29—30,33,

117,118,119-20,170,252

in low-fat diets. See low-fat diets

NHS on, 6-8,58-59

in obesity, 5, 38,40-42, 83

replacing carbohydrates with, 108,116

saturated. See saturated fats

storage of, 79,80, 84

and thermic effect of feeding, 141

unsaturated. See unsaturated fats

WHI study on, 64, 69

Fat Head documentary, 62,199

fattening. See weight gain

fatty acids, 48,49-56, 76-78

in diabetic ketoacidosis, 103-4

in fasting, 85, 86t, 87,88

free, 49, 85, 86t, 103,104

ketone bodies derived from, 101

medium-chain, 52

nonesterified, 49,51,103

oxidation of, 87, 88,100,103,121

plasma levels of, 121—22

role in metabolism, 86

saturated. See saturated fats

structure of, 49,50f, 51,52, 53

synthesis of, 57,88,120-21

trans- and cis- configurations of, 51f,

52-53,77,78

unsaturated. See unsaturated fats

Fay, Allen, 128-29

feedback mechanisms, 13-14

Feinman, Jeffrey, 84

fermentation, 98—99

Fettke, Gary, 247—48

fiber, dietary, 156-57,179

Fine, Eugene J., 15,102,110,188,255,

257,259,263

Finland, 53

fish, 121

flour, enriched, 107

food pyramid, 1

foods

glycemic index of, 46-48,59, 83,

156-58

glycemic load of, 47,59

oxidation of, 95, 98

smell and taste of, 128

Forsythe, Cassandra, 117,121,122

Foster, Gary, 64—66, 68, 80—81, 82,225-31

Framingham Study, 15-16,29, 62-63,69,

193

free fatty acids, 49,85, 86t, 103,104

French Revolution, 91

Freud, S.,20

fructokinase, 112

fructophobia, 30,107-9

fructose, 44,47-48, 57,73,105-16

chemistry of, 74

and fructophobia, 30,107—9

in high-fructose corn syrup, 47,105,

107,113,262

medical literature on, 177

fruits, 28,32,64,187,200,208,226

future of nutrition, 261-63

Gannon, M. C., 155,156,176,235

Gates, Bill, 237

Gaussian distribution, 237

Index

281

gender differences, 33t, 40,41f, 171,201-2,

207

genetic factors, 90,181

in cancer, 253,254,256,257

in cardiovascular disease, 9,58, 70

Gibbs free energy, 142-43

glucagon, 85, 87,103

glucokinase, 113

glucometers, 22

gluconeogenesis, 14,43, 87, 88,133,146

Bernard studies of, 91-94,154

in diabetes, 103,155

fructose in, 111

liver in, 14, 76, 87, 92,93,103,153-54,

155

in low-carbohydrate diet, 76

protein in, 76, 87, 88, 89, 90, 92, 93, 98,

101,154

glucose, 73,76,97-98,120,254

and acetyl-Co A, 76,98,120

amount in diet, 113-15

Bernard studies on, 91-94,153-54

brain requirements for, 43, 76,101,102

in cancer, 254,257

chemistry of, 74, 74f

compared to fructose, 110-13

conversion of fructose to, 105,108, 111,

112f

conversion to fructose, 110

daily requirements for, 102

and fatty acid synthesis, 120

glycogen storage of, 14,43, 75-76, 87

in high-fructose corn syrup, 105,113

oxidation of, 98,100,143,146

in sucrose, 105

synthesis of. See gluconeogenesis

glucose blood levels, 76,93—94

carbohydrates affecting, 2,21-22,

42-43,59,151,155,164-65

constant, as goal of metabolism, 87,96

in diabetic ketoacidosis, 103-4

drugs lowering, 11,22,44,45

in fasting, 85, 86, 87, 88,152,159

feedback system in, 14

and gluconeogenesis, 154

and glycemic index of foods, 46—48,

156-57

in hyperglycemia, 21,43,76,103,154,

164-65

in hypoglycemia, 22, 76,164-65

and insulin, 3,11, 61, 79-80,103-4,

151,154-55,164

in low-carbohydrate diets, 85-86,164—65

in metabolic syndrome, 14,167,168t,

170,171

glucose-insulin axis, 40, 79-80,153,155,

159,262

in metabolic syndrome, 167,169

glucose tolerance test, oral, 152,159

glucose transporter type 4 (GLUT4), 103,

154,257

glycation, 76, 94

glycemic index, 46-48,59, 83,156-58

glycemic load, 47,59

glycerol, 48,49,50f, 87

glycogen, 14,23,43,73,75-76

Bernard studies on, 91-94,153

in fasting, 87, 88

in low-carbohydrate diets, 23, 75-76

in muscle, 75, 87

structure of, 89

synthesis of, 93,108, 111

glycolysis, 98-99,100,254

Good Calories, Bad Calories , 62, 63, 78

Gordon, Tavia, 63

grains, 22,30, 61, 64

grapefruit diet, 118

Handler, Philip, 62

Harpers Illustrated Biochemistry, 86

.Harvard Health Blog, 210-13

Harvard School of Public Health, 27,45,

197,215,242,251

epidemiology studies, 212,227

282

Nutrition in Crisis

hazard ratio, 203-6,217

Head-and-Shoulders effect, 172

healthful diets, 180-81,234

Healthy Hunger-Free Kids Act, 226

hemoglobin Ale, 152,157,159,161

Hesss law, 144-46

Hewat, Claire, 248

high-carbohydrate diets, 33t, 79,89,106,

108,116

ADA on, 151

animal models of, 82-83

with high-fat diet, 30

insulin in, 79,80, 82

in metabolic syndrome, 168,169

triglyceride levels in, 108, 111

high-density lipoprotein, 48,56-58,60,

65,70

in atherogenic dyslipidemia, 14,118,170

in high-carbohydrate diet, 111

in low-carbohydrate diet, 159,170

in metabolic syndrome, 135,168t, 170

high-fat diets, 79, 82,117

in diabetes, 42

with high-carbohydrate diet, 30

saturated fats in, 121-22

high-fructose corn syrup, 47,105,107,113,

262

Hill, Bradford, 185,188-94,195,196,199,

206

Hill, James, 227

Hite, Adele, 86

Hollis, S.,223

homeostasis, 132,135

hormones

and eating behavior, 126,127

insulin as. See insulin

Hu, Frank, 6,251

Huckabee, Mike, 168

Hufhngton Post, 13

Hume, David, 189

hunger, 21,26,123-30

in low-fat diets, 24,30

hydrogen, 95, 96

hydrogenated oils, 52, 77—78

p-hydroxybutyrate, 86t, 101,104,25 8f

hyperglycemia, 21,43, 76,103,154,

164-65

hypertension in metabolic syndrome, 10,

14,167,168t, 170

hypertriglyceridemia, 57

hypoglycemia, 22, 76,164—65

hypothalamus, 126

hypothesis in observational studies, 186,

187,236,240

hypothetico-deductive systems, 185

ice cream, 31, 47-48

imagery techniques, 129—30

inborn errors of metabolism, 75, 90

In Defense of Food (Pollan), 197

Institute for Justice, 248

insulin, 3—4,5,11,14,102-4

anabolic effects of, 61,79,102,127,251

in cancer, 15,254-55,256—59

dosage administered, 151,153

in fasting, 85, 86,87

and fat storage, 80,84

in feedback system, 14

and fructose, 113

and fruit intake, 32

functions of, 79, 80,102

and glucose blood levels, 3,11, 61,

79-80,103-4,151,154-55,164

in glucose—insulin axis, 40,79-80,153,

155,159,167,169,262

in high-carbohydrate diet, 79, 80, 82

and ketoacidosis, 103-4

in latent autoimmune diabetes, 160,161

and lipase activity, 147f, 148

in lipogenesis, 120

in low-carbohydrate diets, 5,23, 85-86,

121,151,251,252

in metabolic syndrome, 10,14,167,

168,169

Index

283

production in type 1 diabetes, 21,42,

103,152,154,159

resistance in type 2 diabetes, 14,21,

42-43,103,152,154,159,216

response to dietary carbohydrates, 10,

23, 79, 80,82,164,251,263

and sucrose-containing foods, 11,44,151

insulin pumps, 164

insulin resistance

in metabolic syndrome, 167,168,169

in type 2 diabetes, 14,21,42-43,103,

152,154,159,216

intensive variables, 39,47, 60

intention-to-treat, 66-67,182,219-32

irritable bowel syndrome, 172

Jacobson, Michael, 55,250

Japanese food, 107

Jenkins, David, 156-57

junk food, 110,128

Kaplan, Bob, 227

ketoacidosis, diabetic, 103-4

ketogenic diets, 23,27,28-29,34

in athletes, 75

in cancer, 15,255,257

coconut oil in, 52

in diabetes, 158

popularity of, 4

very low-carbohydrate, 33t, 118-20,

135-36,171f

ketone bodies, 43,52, 90,100-102,263

in blood (ketosis), 28-29,34,101,103,

257

in cancer, 257,258f, 263

in diabetic ketoacidosis, 103-4

as energy source, 94,100

in fasting, 85, 86, 94

in ketogenic diets. See ketogenic diets

synthesis and utilization of, 94,

100-101,103

in urine (ketonuria), 28

ketonemia, 104

ketonuria, 28

ketosis, 28-29, 34,101,103,257

Keys, Ancel, 53-54

Klein, S., 85

Krai, John, 126

Krebs, Hans, 99

Krebs cycle, 99-100

lactic acid, 98, 99

Lanaspa, Miguel, 110

lard, 51,55,77

latent autoimmune diabetes, 160-66

Lavoisier, Antoine, 91

Lederhandler, Izja, 195

LeFanu, James, 189

Levy, Barbara, 1

lifestyle recommendations

in cardiovascular disease, 225,227

in type 2 diabetes, 152

lipase, 147f, 148

lipids, 4,45-59,76-78

and atherogenic dyslipidemia, 10,14,

118,170

chemistry of, 76-78

cholesterol. See cholesterol

diet-heart hypothesis on, 48,58

fatty acids. See fatty acids

metabolism of, 83—84

triglycerides. See triglyceride levels

lipogenesis, 57(78, 88,120-21

lipolysis, 87,103,104,148,251

lipophobia, 8, 53, 63, 64, 67,107,246,

262

meat in, 198

unintended consequences of, 109

lipoproteins, 48,56-59,70

in atherogenic dyslipidemia, 14,57, 60,

118,170

high-density. See high-density

lipoprotein

low-density. See low-density lipoprotein

284

Nutrition in Crisis

liver, 75,87,112f, 113

acetyl-coenzyme A in, 103,104

in cholesterol synthesis, 5

in glucose synthesis, 14, 76, 87, 92, 93,

103,153-54,155

glycogen in, 14,75, 87,92,153

Lorenzo's Oil\ 56

Los Angeles Times , 227

Low-Carber Forums, 29

low-carbohydrate diets, 3-4, 8,23-25,

151-72

accuracy of dietary records in, 82, 84,

135-36,227

ADA on, 13,44-45

AHA on, 57

AMA on, 62

in athletes, 75

Atkins diet as. See Atkins diet

calories in, 25,32, 33t, 66, 81, 84,135

in cancer, 15,253-59,263

in carbohydrate intolerance, 21-22,261

and cardiovascular disease, 70,158—59

cheese in, 32

cholesterol levels in, 65,70,120,159

compared to low-fat diets, 65-70,

80-82,118-20,134f, 135-36,169-70,

171f, 225-31,246

compared to low-glycemic index diet,

156-58

compliance with, 24,32,135-36,225,

226-27,231,251

cravings in, 30-31, 32-33

desserts and sweets in, 26,30-31

in diabetes, 151—66. See also diabetes

mellitus

fat in, 25,29-30,33,117,118,119-20,

170,252

first revolution in, 61-70

Foster study on, 64-66,68, 80-81,

225-31

fruits in, 32

future of, 261-63

glucose blood levels in, 85—86,164—65

glycemic index in, 46-48

glycogen in, 23, 75-76

as high-calorie starvation diet, 85-86,

101-2

insulin in, 5,23, 85-86,121,151,251,

252

ketogenic. See ketogenic diets

medical education on, 3-4, 69,169,

247-48

medical literature on, 177

metabolic advantage in, 131,132-33

in metabolic syndrome, 15,21,28, 70,

79,106,113,167-72,261

metabolism in, 3-4,23,28,75-76,131,

132-33,252

nuts and nut butters in, 32

in obesity, 113

overeating in, 32-33

in overweight, 79,106

principles of, 27-33

protein in, 89-90,101

resistance to, 10-11,13,15,22,44,

61-63,65,68-69,156,168,176,228,

245-51,256,261

satiety in, 23,24—25,32-33,66,127,251

second revolution in, 12-13,243-63

sugar in, 30—31,105—6,113

techniques for staying on, 26

triglyceride levels in, 57, 65,120,159,

170,228-31

types of, 33-34

vegetables in, 28-29

in weight loss. See weight loss programs

Westman study of, 157-58

low-density lipoprotein, 48,56-58, 65,70

in atherogenic dyslipidemia, 57,60,118

particle size of, 57-58,167

pattern B, 58,60,118

low-fat diets, 4—5, 8,24,29—30, 62

accuracy of dietary records in, 82, 84,

135-36,227

Index

285

calories in, 25,30,66, 81, 84,119,135

and cholesterol levels, 56,57-58, 60,

69-70,120

compared to low-carbohydrate diet,

65-70, 80-82,118-20,134f, 135-36,

169-70,171f, 225-31,246

diet-heart hypothesis on, 5-6,13,48,

261

failures of, 262

hunger in, 24,30

in metabolic syndrome, 168,169

nutritionist support of, 68-69

in obesity, 5, 38

political policies on, 109

recommendations on, 76-77

research findings on, 15-16,29-30,54,

63-70,80-82,135-36

saturated fat and fatty acids in, 29-30,

117,119-20,121

triglyceride levels in, 228-31

low-glycemic index diet, 156-58

lung cancer, 185,189-90,192,196

Lustig, Robert, 105,107,108

malonyl-coenzyme A, 120-21

margarine, 77,109

Maxwell, James Clerk, 140

Maxwells demon, 138-39,140

Mayes, Peter, 88

McDonald s, 55

McGovern, George, 62

meat consumption, 186,197-213

in carnivore diet, 79

epidemiologic studies on, 59

in low-carbohydrate diet, 28,33

and satiety, 123

warnings on, 27, 90

medical education, 3-5,245,247-48

on diabetes and nutrition, 162

on fat in diet, 79

on low-carbohydrate diets, 3-4,69,169,

247-48

quiz on nutritional concepts in, 35-60

on scientific papers, 180

Medical Illuminations ,, 233

medical literature, 175-242. See also

research studies

Mediterranean diets, 27, 53-55

menopause, 64

Men's Healthy 123

meta-analysis, 81,192,236-40,261

metabolic advantage, 131,132-33,141,239

metabolic syndrome, 10,14-15,167-72

insulin in, 10,14,167,168,169

low-carbohydrate diet in, 15,21,28, 70,

79,106,113,157-58,167-72,261

Volek study of, 117-22,135,169-70,

171f

metabolism, 3-5,10,15, 91-104

acetyl-CoA in, 76,97-100,101

in cancer, 254-59

of carbohydrates, 86, 88,110-13

in diabetes, 4,43

in fasting, 85-86, 87-88,94

fuels in, 76, 97-98,120

gluconeogenesis in. See gluconeogenesis

goals of, 76, 87, 88,96-97

homeostasis in, 132,135

inborn errors of, 75, 90

insulin in, 79,80,102-4

lipogenesis in, 57, 78, 88,120-21

in liver, 112f, 113

in low-carbohydrate diets, 3-4,23,28,

75-76,131,132-33,252

of sugar, 110-13

mice studies, 82-83

milk, 1

Milk, Harvey, 109

minerals, 73

Mitchell, Peter, 100

moderate-carbohydrate diet, 33t

monosaccharides, 73

monounsaturated fat and fatty acids, 6-7,

49,50f, 51-52,55,121

286

Nutrition in Crisis

mortality rate

in coronary artery disease, 222-23

in intention-to-treat analysis, 224

in meat consumption, 197-213

Mother Jones, 115

Mukherjee, Siddhartha, 185

muscles, 96, 99

and glycogen, 75, 87

and ketone bodies, 101

and protein, 89,90

starvation affecting, 20,101

Mussolini, Benito, 31

Nabel, Elizabeth, 64,226

National Academy of Sciences, 62

National Health and Nutrition

Examination Survey, 40,42

National Institutes of Health, 48,63,64,

200 - 201,212

Nature , 110

Naughton,Tom, 62

New Atkins for a New You, 27

Newell, David, 222-23

New EnglandJournal of Medicine, 27

Newton, Isaac, 221

New York Times Magazine, 62

nicotinamide adenine dinucleotide, 97-98,

Njurunda Medical Center, 247

Noakes, Tim, 249-50

nonesterified fatty acids, 49,51,103

Nordmann, Alain, 81

Norman, G. R., 178

nose cancer, 192,199

number needed to treat (NNT), 206

Nurses’Health Study, 6-8,58-59

nut butters, 32

Nutrition, 255

nutrition, basic rules of, 19-22,176

nutritional epidemiology, 187,188-89

Nutrition and Metabolism, 110,169—70,220

Nutrition in Crisis, 246

nuts, 32,121

Nuttall, E Qy 155,156,176,235

oatmeal, 2,30

Obama, Michelle, 226

obesity, 10,25,116

and cancer, 254,256

carbohydrates in, 40-42,59,83,107,

113,153

and diabetes, 155-56,187-88

and diet sodas, 192

epidemic of, 39-42, 59,66,73,107,

198,200

and fat in diet, 5, 38,40-42, 83

and metabolic syndrome, 14,168t

sugar in, 105,186-87

vagotomy in, 126

observational studies, 185-96,235-36

criteria on, 190-94,196

hypothesis in, 186,187,236,240

on meat consumption, 200

odds ratio, 203-6,207

oils, 48,49,52,55,77-78

oleic acid, 51, 52,55,60,76,121

olive oil, 51,52,53,55,60,121

orange juice, 106

orexigenic hormones, 126

Oslo Diet Heart Study, 29,69

overeating, 32-33

overweight, 79,106,124,153

and metabolic syndrome, 10,14,135,

167,170

oxidation, 95-100

and calories, 38,60

of carbon, 143-44

of fat, 80,84, 85, 86t

of fatty acids, 87, 88,100,103,121

of glucose, 98,100,143,146

of protein, 145—46

oxidation-reduction reactions, 95-96,97-98

Paine, Thomas, 62

Index

287

paleo diets, 27-28

palmitic acid, 57, 76, 88,120,121

Pan, A, 210,211,212,213

pancreas, 21,42,159,216

parachute use, 241-42

Pasteur, Louis, 94

PDQ Statistics , 178

peer review, 27,175,213,248,261

PellJ.P, 241-42

Phinney, Steve, 75

phosphorylation, oxidative, 98

physics, 38,96

Planck, Max, 246

Pogozelski, Wendy, 160-66

Poisson distribution, 237

politics, 10,13,15,16,109

Pollan, Michael, 8,63,197

polyol reaction, 110

polysaccharides, 46, 73, 74

polyunsaturated fat and fatty acids, 6-7,49,

51,121

Pop-Tarts, 107

portal vein, 93

postabsorptive state, 87

postmenopausal women, 64

postprandial state, 87

potatoes, 28,47

Prigogine, Ilya, 138

Principia, 221

probability, 139,140

prospective experiments, 188

protein, 73, 89-90,197-98

calories in, 37

consumption trends, 73,198,200

in fasting, 88, 94,101

in glucose synthesis, 76, 87, 88, 89, 90,

92,93,98,101,154

in low-carbohydrate diet, 25, 33,101,

118,135

in low-fat diet, 118,135

oxidation of, 145-46

satiating effects of, 123

and thermic effect of feeding, 141

Protein Power, 21 , 32,187

proteinuria, 90

Psychopathology of Everyday Life, 20

Public Health Nutrition, 54

Puritans, 197

pyruvic acid, 98-99,254

quantum mechanics, 131

quiz on nutritional concepts, 35-60

Raatz, S. K., 122

Rabinovitch, Abraham, 11-12

randomized controlled trials, 158,189,193,

240-42,246

rapeseed, 55-56

Ravnskov, Uffe, 246

Reaven, Gerald, 14,167

red meat, 27,59,186,197-213

redox reactions, 95-96, 97-98

reinforcement of behavior, 128-30

relative risk, 190-91,196,198,201-6,209,

211

research studies, 175—242

animal models in, 82-83, 91-92,118,

154

association and causality in. See

association and causality

on cancer, 255,256-59,263

common sense in, 200,234,235,242

comparison of low-fat and

low-carbohydrate diets in, 65-70,

80-82,135-36,169-70,225-31

consistency of results in, 192

on diet-heart hypothesis, 62-63,248,

261

on egg consumption, 215-18

epidemiologic. See epidemiologic

studies

of Foster, 64-66, 68,80-81, 82,225-31

Framingham Study, 15-16,29,62-63,

69,193

288

Nutrition in Crisis

research studies ( continued )

on healthful diets, 180-81

inclusion criteria in, 223-25

intention-to-treat in, 66-67,182,

219-32

interpretation of, 12,180

levels of evidence in, 182

on low-fat diets, 15—16,29-30,54,

63-70,80-82,135-36

on meat consumption, 197-213

meta-analysis of, 81,192,236—40,261

misrepresentation in, 181

number of subjects in, 236,240

Nurses’Health Study, 6—8,58-59

observational. See observational studies

peer review of, 27,175,213,248,261

prospective experiments in, 188

randomized controlled trials in, 158,

189,193,240-42,246

self-reported diet records in, 82, 84,

135-36,202,216,227

self-serving descriptions in, 234—35

Seven Countries Study, 53-55

specificity in, 192,199

statistics in. See statistics

visual representations on, 233-34

of Volek, 75, 89,117-22,135-36,170

of Womens Health Initiative, 29,64,

69,193,226

Reverse Mussolini Fallacy, 31

Review of Physiological Chemistry, 86

rice, 27,59,177,179

Rise and Fall of Modern Medicine, 78,189

risk, 190-91

absolute, 191,202,205,206,209,

211-12

from eggs, 215-18

from meat, 198,200—213

and probability, 208

relative, 190-91,196,198,201-6,209,

211

Robinson, Jackie, 3,11

rodent studies, 82-83,118

sandwich, invention of, 189

satiety, 123-30

in low-carbohydrate diets, 23,24—25,

32-33,66,127,251

saturated fats, 48,60,77-78,117-22

and cardiovascular disease, 6—8,29—30,

51,53-54,58-59,64,77,118

compared to unsaturated fat, 49

consumption trends, 40, 41f

dietary carbohydrates affecting blood

levels of, 57, 88-89,121-22,170

in low-carbohydrate diets, 117,119-20,

170

in low-fat diets, 29-30,117,119—20,

121

research findings on, 59,78, 89,117-22,

170,246,250-51,261

structure and form of, 50f, 51,52,53,78

synthesis of, 57,78, 88,120-21

Sboros, Marika, 248

school lunch programs, 226

science, 195,196

association and causality in, 187—88,

189

philosophy of, 185,189,195

self-deception about evidence, 11-12

self-reported diet history, 227

on egg consumption, 216

on fat and carbohydrate intake, 82,84,

135-36,227

on meat consumption, 202

Semmelweis, Ignaz, 11

Seven Countries Study, 53—55

j Seventh Seal, 218

Shakespeare, William, 27,197,233

Sharon, Ariel, 12

significance, statistical, 198,240

Singer, Isaac Bashevis, 209

Sinha, Rashmi, 199,200,206

Siri-Tarino, R W., 8

Index

289

60 Minutes, 107

Skinner, B. F. } 125

smell of food, 128

Smith, G. C., 241-42

smoking, 2,115

and cancer, 185,190,192,195,196,199

and coronary thrombosis, 190,196

hazard ratio in, 206

snacking, 124

sodas, 1,30,31,186,192

Somers, Suzanne, 32

Sommerfeld, Arnold, 131,136

sorbitol, 110

South Africa, 249-50

sperm cells, 110

standard deviation, 237-38

starch, 30,44,46,73,108,112,113

in bread, 75

chemistry of, 74f, 75

in diabetes, 2,106,153

overconsumption of, 107

in weight-loss diets, 1

starvation, 20, 85-86, 88,101-2

statistics, 177-79,194

absolute risk in, 191,202,205,206,209,

211-12

averages in, 237-39

and common sense, 200

on egg consumption, 215-18

hazard ratio in, 203-6,217

intensity measures in, 39

in intention-to-treat, 219-32

interpretation of, 12

on meat consumption, 200-213

number needed to treat in, 206

in nutritional epidemiology, 194

odds ratio in, 203-6,207

probability in, 208

relative risk in, 190-91,196,198,

201-6,209,211

significance in, 198,240

standard deviation in, 237-38

standard error of the mean in, 80

two-tailed distribution in, 207

visual presentation of, 233-34

stearic acid, 55,120,121 .

stearoyl-CoA desaturase-1,121

Stein, Rob, 199

Stewart, Potter, 185

Streiner, D. L., 178

streptomycin in tuberculosis, 189-90

sucrose, 47,73, 74f, 105,110

in diabetic meal plan, 11,44,151

sugar alcohols, 67-68,73

sugar industry, 115

sugars, 1,73-75,91-94,105-16

dangers attributed to, 262

in desserts and sweets, 30-31

in diabetes, 44,153

as easy target, 246

fructose. See fructose

in fruits, 32

glucose. See glucose

in low-carbohydrate diets, 30—31,

105-6,113

and obesity, 186-87

refined, 83

research findings on, 177,195

in sodas, 1,30

sucrose, 11,44,47, 73,105,110,151

surgery, coronary artery, 222—23

Sweden, 246-47

sweeteners, 31,113-15

sweets. See desserts and sweets

table sugar, 105

taste, 128

Taubes, Gary, 62,63, 64,115,255

Teicholz, Nina, 55,63,250-51,262

temperature regulation, 126

thermic effect of feeding, 133,141

thermodynamics, 84—85,127,131-48,187

thermogenesis, 141

thrombosis, coronary, 190,196

290

Nutrition in Crisis

tobacco industry, 115

trans-fats, 5 If, 52,55,56,63,77,78,109

triacylglycerols, 49,50f, 118,120,122,148

tricarboxylic acid cycle, 99-100,108

triglyceride levels, 49,56-58

in high-carbohydrate diet, 108, 111

in low-carbohydrate diet, 57,65,120,

159,170,228-31

in low-fat diet, 228—31

in metabolic syndrome, 135,168t, 170

triose phosphates, 111, 112

tuberculosis, 189-90

Twain, Mark, 203

Twinkles, 107,109

twins, weight loss in, 238-39,239f

two-tailed distribution, 207

type 1 diabetes, 21

insulin in, 21,42,103,152,154,159

ketoacidosis in, 103-4

latent autoimmune, 160-66

type 2 diabetes, 21,248

complications in, 159

early symptoms in, 153

and eggs, 215-18

glucose blood levels in, 42-43

insulin in, 14,21,42-43,103,152-53,

154,159,216

lifestyle recommendations in, 152

low-glycemic diet in, 156-57

and meat consumption, 213

United Kingdom Prevention of Diabetes

Study, 159

unsaturated fats, 48,51-52,55,60,77-78,

121-22

and cardiovascular disease, 6-7,58-59

monounsaturated, 6-7,49,51—52,55,

121

polyunsaturated, 6—7,49,51,121

structure and form of, 52,53,77

urea, 90

USDA, 1,20,42,69,107,226,250,251

vagotomy, 125-26

vagus nerve, 125-26

vegetable oils, 6, 52,55, 77

vegetables, 28-29,64,187,198,200,208,

226

vegetarian diets, 90,197—98,204

Vegetarian Times , 197

very low-carbohydrate ketogenic diets, 33t,

118-20,135-36,171f

vitamins, 2, 73

and intention-to-treat, 224,230

Volek, Jeff, 24,42, 75, 89,117-22,135-36,

169-70,251

Vowell, Sara, 197

Wainer, Howard, 233

Warburg effect, 100,254

Washington Post, 199,200,201

water, 95

Watergate scandal, 199

Watson, James, 181

weight gain

in animals, 61

carbohydrates in, 84—89,107

and eating behavior, 127

energy density of foods in, 38-39

sugar in, 105

weight loss programs, 1,3,20-21

Atkins diet in, 13,23,65,66,67,153,225

average weight loss in, 238

behavioral reinforcement in, 128-30

breakfast in, 127-28

calorie reduction in, 21, 38

cheese in, 32

compliance with, 24,32,251

cravings in, 32-33

in diabetes, 22,153,155,156,176,235

effectiveness of low-carbohydrate diets

in, 4,13,67,81,245,251

exercise in, 22,129

fasting in, 20, 88

Forsythe study of, 121

Index

291

Foster study of, 65-66, 80-81,225,226,

227

gender differences in, 40,171

graded response in, 28,34

intention-to-treat analysis of, 220-21

ketone bodies in, 28,34

low-fat diet compared to

low-carbohydrate diet in, 65-66,

80-81,118-19,135,225

metabolic advantage in, 131,239

in metabolic syndrome, 14

personal benefits of low-carbohydrate

diet in, 251-52

professional resistance to

low-carbohydrate diet in, 62

satiety in, 23,24-25,32-33,66,127,251

scientific literature on, 61-62

sugar in, 105

Swedish endorsement of

low-carbohydrate diet in, 247

thermodynamics in, 136

twin studies of, 238-39,239f

Volek study of, 24,118-19,135,251

Weinberg, Steven, 185

Western diets, 40

Western Electric Study, 69

Westman, Eric, 46,151,153,156,157-58

wheat, 200

White, Dan, 109

Willett, Walter, 6,32,249

Wills, Gary, 61

Wilmot,John, 124

Wolfe, R. R., 85

women

carbohydrate and fat intake in, 33t, 41f,

glucose-insulin axis in, 40

meat consumption in, 201-2,207

weight loss in, 171

Womens Health Initiative, 29, 64, 69,193,

226

Woodward, Bob, 199

Wordy Shipmates, 197

Yancy, William, 158

yogurt, 99

YouTube, 107

yo-yo dieting, 24

ABOUT THE AUTHOR

RICHARD DAVID FEINMAN is a professor

of cell biology at the State University of

New York Downstate Medical Center in

Brooklyn, New York, where he has been

a pioneer in incorporating nutrition into

the biochemistry curriculum. A graduate

of the University of Rochester and the

University of Oregon (PhD), Dr. Feinman

has published numerous scientific and

popular papers. Dr. Feinman is the founder

and former coeditor-in-chief (2004-2009) of the journal Nutrition and

Metabolism . His current research interest is in the application of ketogenic

diets to cancer.

The Nutrition and Metabolism Society

“[A] Fascinating book ... by one of the original

low-carb researchers whose grounding in the

field goes back decades.”

—Nina Teicholz , author of The Big Fat Surprise

“A must-read for anyone with a serious interest

in health and nutrition.”

-Michael R. Eades.MD,

author of Protein Power

“Every scientific discipline needs a Dr. Feinman. He

is just as skeptical about what he believes as what

he disbelieves. His writing bears eloquent testi¬

mony to why he is such a scientific treasure.”

-Timothy Noakes , MD, emeritus professor,

University of Cape Town; founder,

The Noakes Foundation

“This cant-put-it-down title is the closest thing

to a complete, popular analysis of the biochem¬

istry of human nutrition that you will find and a

superb lesson in how scientific studies have been

manipulated to prove fiction.”

-Richard K. Bernstein , MD, author of

Dr. Bernsteins Diabetes Solution

“This book tells an informative and entertaining

story about what went wrong in the nutrition

establishment and advocates for a rational solu¬

tion to the problem.”

—Dominic D’Agostino, PhD, leading

scientist on ketogenic metabolic therapies

A lmost every day it seems a

new study is published

that shows you are at risk

for diabetes, cardiovascular disease,

or death due to something youVe

just eaten for lunch. Many of us no

longer know what to eat or who to

believe. In Nutrition in Crisis ., distin¬

guished biochemist Dr. Richard

Feinman cuts through the noise,

explaining the intricacies of nutrition

and human metabolism in accessible

terms and providing a commonsense

approach for making intelligent

nutritional choices.

Nutrition in Crisis offers an

unsparing critique of the nutritional

establishment, which continues to

demonize fat and ignore the benefits

of low-carbohydrate and ketogenic

diets—all despite decades of evidence

to the contrary. Entertaining, infor¬

mative, and irreverent, Dr. Feinman

paints a broad picture of the nutri¬

tion world that shows the beauty

of the underlying biochemistry, the

embarrassing failures of the medical

establishment, and what’s wrong

with the constant reports that the

foods we’ve been eating for centu¬

ries represent a threat rather than a

source of pleasure.

Chelsea Green Publishing

85 North Main Street, Suite 120

White River Junction, VT 05001

(802)295-6300

www.chelseagreen.com

Cover design by Melissa Jacobson

$24.95 USD

worth it"

pRmej on recvo-Ed paper

PRINTED IN CANADA

ISBN TPfl-l-bOBSfl-an-S