Metabolical: The Lure and the Lies of Processed Food, Nutrition, and Modern Medicine

by Robert H. Lustig

can’t I just exercise past my bad diet? Won’t an extra ten minutes on the elliptical trainer solve everything? Amateur Finnish triathlete Sami Inkinen tried and failed. Sami was one of the original founders of Nokia, sold his share early, and moved to the US to attend Stanford Business School. There he started the real estate website Trulia, which was bought by Zillow for $2.5 billion. In other words, Sami had more money than God—and he exercised five hours per day.

exercise without dietary change can help to affect five of the eight subcellular pathologies (see Chapter 7): mitochondrial dysfunction by generating newer and fresher mitochondria; insulin resistance by reducing skeletal muscle and liver fat; improvement of propensities toward autophagy and reduce inflammatory markers; and maybe even epigenetics, although this effect appears to be mediated through exercise’s suppression of inflammation. However, exercise alone won’t improve glycation, oxidative stress (exercise actually makes this worse), or membrane integrity and fluidity.

Virta Health, a ketogenic diet start-up that proves diet matters more than exercise in reversing type 2 diabetes. The results have been impressive, so much so that the former chief medical officer of the American Diabetes Association, Dr. Robert Ratner, signed on to be their chief executive officer after having previously eschewed the low-carb diet.

bad food isn’t the result of upheavals, as it’s been engineered to be insidious. It’s very bad chronically, as it is the cause of NCDs, social disparities, healthcare collapse, mental health crises, societal devolution, and in time, an even greater risk for social upheaval. Ultimately, even more people die—just slower, and it’s under the wire, as the cause isn’t as clear, so no one does anything about it, and it only gets worse. In addition, bad food puts you at increased risk of getting seriously ill or dying from COVID-19

Nutrition vs. Nutritionism Nutrition is the most important and malleable factor influencing people’s life span (how long we live) and health span (how well we live). Studies on fraternal vs. identical twins show that genetics account for 25 to 30 percent of a person’s longevity. The other 70 to 75 percent proves that while favorable genetics clearly play a role, the environment, including a bad diet, can easily overcome those gifts, hence why the US has seen reduced life expectancy four years in a row.

Since the publication of The Omnivore’s Dilemma (2006), food journalist Michael Pollan has made the point that nutrition is religion—because it requires believing without seeing. After all, nutrition must be all about what’s in the food. With the discovery of the first vitamin (B1, or thiamine) in 1912, scientists became convinced that there were chemicals in food that conferred health, so there must also be chemicals in food that conferred illness. This in turn has led to the concept of nutrients as being the lowest common denominator for any eating paradigm, giving rise to the religion of nutritionism.

Check out the latest fodder on vegan diets—for instance, Forks Over Knives (2011) or The Game Changers (2018), and find any mention of alternatives. I’ve had my own public run-in with Kip Andersen and protein-phobic Dr. Garth Davis, the director and talking head, over their scaremongering movie What the Health (2017) (an egg is the same as five cigarettes?); and all I can say is there wasn’t any science offered in explanation, either in the movie or in the debate. But it’s true on the other side of the street as well. Check out The C Word (2016), a story with an “n of 1” that extols the benefits of a ketogenic diet for cancer, and The Magic Pill (2017), which lauds dietary fat—no mention of any alternatives.

Kwashiorkor is a different disease, resulting from protein deficiency without calorie deficiency. These babies have huge bellies because their livers are filled with fat—they’ve got nonalcoholic fatty liver disease (NAFLD). What caused the fatty liver? Cassava flour—a high-carbohydrate, low-fiber food, resulting in glycation, oxidative stress, mitochondrial dysfunction, insulin resistance, poor membrane integrity, and inflammation (see Chapter 7). In other words, they have “developing world” metabolic syndrome. Well, guess what? We have instead “developed world” kwashiorkor.

First of all, most pundits in the field aren’t bench scientists or clinicians; they tend to be nutritional epidemiologists, and nutritional epidemiology has significant limitations. Epidemiology means correlation, not causation. Like John Snow’s cholera/Broad Street pump exercise (see Chapter 2), nutritional epidemiology studies are discovery, and discovery can be very important in posing the questions that truly need answering. However, it almost never answers the questions by itself; you need to design a proper study to answer them (see below). Just because A is associated with B, does that mean that A causes B? Or could it be reverse causality (B causes A)? Or could it be intermediate causality (C causes A or B)? Could it be irrelevant (C is associated with B and D, and D causes A)? As an example, ice cream consumption correlates with frequency of drownings. Does that mean eating ice cream causes you to drown?

Correlation does not automatically imply a cause-and-effect relationship. But the media, in its effort to sell newspapers or snatch eyeballs, treats almost all epidemiological studies as causation. Therefore, the public doesn’t understand the difference either.

Some investigators and news sources tout meta-analyses, an attempt to conglomerate multiple studies. It’s the gold standard to prove your point. And meta-analyses can do this well, when the individual studies are independent of industry and are also scientifically sound. But many such analyses are GIGO—“Garbage in garbage out”—as they are only as good as the data they are based on. And when the food industry is in charge, the results are suspect.

Another reason why nutrition remains an academic backwater is because we don’t have good biomarkers (e.g., blood tests) that measure what people are actually eating. Most of the data in nutritional studies are obtained through memory recall to food questionnaires. You can see for yourself—try asking someone what they’ve eaten for the last three days. Most people can’t tell you what they’ve eaten in the last three hours. Which doesn’t even factor in that some...

Dr. John Ioannidis, an internist and accomplished statistician from Stanford University, has proposed that we do away with nutritional epidemiology entirely, as the studies are impossible to control, the data is perennially abused, and the results are virtually guaranteed to be wrong. I disagree. There’s no doubt that people read too much into these studies, but they also need to be educated. No single nutrition epidemiology study is ever the final word, because they don’t rise to the level of causation. There are only two types of studies that can approach the rarified air of causation. One, called econometric analysis, looks at the natural history of both consumption and disease prevalence, but also accounts both for confounders and for time (time is essential to causation).

this level of evidence provides what we call causal medical inference; the level of proof we have today for tobacco and lung cancer. The second is called randomized controlled trials (RCTs), where the investigator varies only the one nutrient under study. However, such studies must have a placebo control group to be able to rise to the level of proof.

even without being pregnant, each of us is always eating for two, because we also have to feed our own intestinal microbiome, which receives and metabolizes about 30 percent of our ingested nutrients. If the nutrients didn’t enter our bloodstream, did we really get them?

We now know that we have to feed it to stay healthy. When we don’t feed it right (e.g., depriving it of dietary protein), those bacteria send blood-borne and neural signals that tell our brains to alter our behavior so that they can get the nutrition that they do need. Whether you like it or not, you’re eating for two—you’re in a symbiotic relationship with your gut, and if you hurt your gut, your gut will hurt you back.

The argument I’ll make throughout the rest of this book is that it’s not what’s in the food, it’s what’s been done to the food that matters. Because the real nutritional question is: who and what are you feeding? Are you feeding the human? Or are you feeding the intestinal microbiome? And is your liver working right based on the share that you

In a population where 88 percent have some level of metabolic dysfunction, the entire concept of healthy has been obfuscated.

my personal all-time favorite is the U.S. Institute of Medicine, which in 2004 codified an upper limit for added sugar at 25 percent of total calories. In what universe is 25 percent of calories as added sugar justifiable? This gave the food industry carte blanche to add as much as they possibly could, making us sicker and sicker.

The key to fending off chronic disease is to keep those eight subcellular pathways running right—and each and every one of them can be made to run right with two simple dictates: Protect the liver. You have to protect the liver from fructose, glucose, branched-chain amino acids, omega-6 fatty acids, iron, and other oxidative stresses—all of which end up causing fat accumulation and liver damage, and generate insulin resistance. This can be done by either reducing the dose of dietary liver stressors (e.g., a low-sugar diet) or their flux (e.g., a high-fiber diet, which blocks sugar absorption, thus reducing the rate by which fructose and branched-chain amino acids reach the liver). Feed the gut. If you don’t feed your microbiome, your microbiome will feed on you; it will literally chew up the mucin layer that protects your intestinal epithelial cells, which increases the risk for leaky gut, inflammation, and more insulin resistance. The goal is to deliver more nutrients farther down the intestine (e.g., a high-fiber diet).

Fiber is an essential nutrient—not for only you, but also for your microbiome. The fiber in Real Food is of two kinds: soluble, which is globular, like what holds jelly together (e.g., psyllium, pectin, inulin); and insoluble, like the stringy stuff in celery (e.g., cellulose, chitin, peptidoglycan). You need both, as they do different jobs; and you also need the geometry in order to make fiber work for you.

intact fiber—found in Real Food—has many benefits, and not just short-chain fatty acids (SCFAs). In the processed food industry, the germ of the grain (the nucleic acids, flavonoids, polyphenols) is removed along with the fiber because they can go rancid (see Chapter 19). Protecting the liver means maintaining the fiber and keeping the germ intact as well. Two simple precepts—protect the liver, feed the gut. Real Food (low-sugar, high-fiber) does both. Processed food (high-sugar, low-fiber) does neither. Processed food is the primary suspect in our current health and healthcare debacle, because it doesn’t improve our eight subcellular pathologies, our three nutrient-sensing enzymes, and our two physiologic precepts.

Politics is often spread through myths, which themselves are easily turned into propaganda, thus perpetuating the politics—a vicious cycle. These three are replete within nutrition, perhaps more so than any other medical discipline, because there are so many stakeholders with their own beliefs and agendas.

Let’s start with the myths surrounding terminology. Here are just three examples: The word “weight”

It’s not the fat you can see, it’s the fat you can’t see that matters.

The word “fat”—does it mean body fat or dietary fat? Or, as you will soon learn, fatty acid? Or, “do these pants make my butt look fat?” (Pro-tip: never answer this question.) Two-thirds of the US populace continue to believe and perpetuate the myth that “fat makes you fat.” While it’s true that dietary fat could become body fat, it does so only in response to insulin. And so weight isn’t driven by dietary fat, which doesn’t raise insulin, but rather by dietary refined carbohydrate and sugar, which do.

The word “sugar”— does it mean blood sugar (glucose) or dietary sugar (glucose-fructose)? The food industry says “you need sugar to live”—but while you do need a blood glucose level to live, you don’t have to consume that glucose. In fact, your liver can make glucose from the glycerol (see Fig. 7–3d) released from the breakdown of triglyceride...

you don’t need fructose (the molecule that makes food sweet) to live at all. In fact, there’s no biochemical reaction in any animal cell on the planet that requires dietary fructose. Which means you m...

From a physics standpoint, a calorie is a calorie. But so what? This has nothing to do with what happens to those calories in the human body, because weight gain is only about how those calories are stored.

Understanding these various phenomena shows that in fact “a calorie is not a calorie,” and there’s an actual difference between eating a handful of almonds and a donut, even if their calorie count is the same.

Sugar sweetened beverages (SSBs) are causative for at least three diseases of metabolic syndrome—type 2 diabetes, heart disease, and fatty liver disease—plus tooth decay. So, what about noncaloric diet sweeteners, for those with a “sweet tooth”? Stevia, sucralose, aspartame, acesulfame-K, allulose, xylitol, erythritol, and others would seem the obvious choices—no calories, so no heart disease, right? No fructose, so no liver fat or diabetes, right? Not so fast.

diet sweetener consumption is also correlated with metabolic syndrome. Studies of switching out sugar for diet sweeteners don’t show beneficial effects on weight loss. Rather, the data show that sugar is a direct cause of metabolic syndrome—though thus far we only have correlation with diet sweeteners.

In the last fifteen years, American sugar consumption has dropped from 120 to 94 pounds per year, yet obesity and metabolic syndrome persist unabated. Could diet sweeteners be playing a role? The only surefire way to find out is for Americans to de-sweeten their food across the board—drinks, too. And don’t start thinking juice is the answer

As discussed in Chapter 11, the fiber in food is perhaps the most important nutrient for health, because it singlehandedly protects the liver and feeds the gut in six different ways:

a primary goal of improving metabolic health is to get the insulin down. One way is to eat foods that don’t make your blood glucose rise too fast. This is where amylose—the “good Amy”—comes in.

Is it the glucose spikes or the insulin spikes that do the damage?

The important parameter is glycemic load (GL). GL is different from GI—how much food do you have to eat to get the 50 grams of carbohydrate? GL takes into account the beneficial effect of fiber. A good example is carrots, which are high-GI (lots of carbohydrate) but low-GL (even more fiber). More fiber means a larger portion, because there’s less digestible carbohydrate. You can turn any high-GI food into a low-GL food by eating it with its original fiber. Real Food is by definition low-GL.

Fructose! Fructose is the most egregious cause of liver insulin resistance and metabolic syndrome because of how the liver uniquely metabolizes it. Fructose isn’t glucose—when eaten, it doesn’t raise the blood glucose level (it’s not measured in the glucose assay). In fact, by definition, it’s low-GI, because it has no glucose. Still, this hasn’t stopped the food industry from trying to capitalize on the low-GI craze by adding fructose to foods.

The epic battle between British physiologist John Yudkin and Minnesota epidemiologist Ancel Keys for control of the American Diet, detailed in my book Fat Chance (2012) and Nina Teicholz’s The Big Fat Surprise (2017), is now sixty years old. Yudkin wrote Pure, White and Deadly (1972) targeting dietary sugar; Keys wrote The Seven Countries Study (1980) targeting dietary saturated fat. Both scientists had correlation but not causation; both had static data (single points in time) rather than longitudinal data (patterns over time). Both used ecologic (population) data, which is much flimsier than individual data. In other words, both had weak cases.