The more damaged you are, the more carbohydrate restriction is likely to benefit you long term.
- Peter @Hyperlipid

I don’t think there are too many people out there familiar with the mechanisms of diabetes and insulin resistance that would disagree with that statement.

But just because a low-carb diet causes fat loss in this population, that doesn’t mean that carbs caused the fat gain or damaged metabolism in the first place.

I wrote about this in a previous article, There is No Single Cause of (or Treatment or) Obesity, but based on some of the comments and discussion I’ve been seeing online recently, I think it bears repeating: in order to properly frame this debate, it’s essential to separate the causes and treatment of obesity. If we don’t do this, we might as well not even have a debate at all because we won’t be talking about the same thing.

We know without a doubt that statins lower cholesterol. But does that mean high cholesterol is caused by a statin deficiency? If you break your arm, your doctor will probably put a cast on to help it heal. Does that mean we should all wear casts on our arms to make sure they don’t break?

Metabolically damaged vs. healthy: apples & oranges

Another important distinction that should be made – but often isn’t – is the difference between how people that are metabolically healthy and metabolically damaged respond to food. Of course excess carbohydrates are more likely to cause problems in someone with leptin and insulin resistance and impaired glucose tolerance. But it doesn’t follow that the same will be true in someone without metabolic problems.

People without gall bladders don’t digest fat very well. Does that mean fat causes indigestion? I have patients with iron overload due to a genetic condition called hemachromatosis. They need to limit their red meat intake because of this. Does this mean we should all avoid red meat to prevent iron overload?

We’re not robot clones. We have different genes, lifestyles, gut flora, immune fitness and exposures to toxins, stress and infections, as well as different emotional and psychological relationships with food. All of these factors play a role in weight regulation.

This explains why two people can react to the same diet in entirely different ways. And it also explains why it’s so ridiculous to extrapolate something that you experience personally to everyone else. (If I read another comment from someone saying that a low-carb diet worked for them, so the insulin-carb theory must be correct, I’m going to lose it. And it takes a lot to make me lose it!)

My unified theory of obesity

For what it’s worth, here’s my “unified theory” on what causes obesity.

Modern lifestyle + genetic predisposition = obesity.

It really is that simple.

Modern lifestyle includes processed, refined and highly rewarding and palatable foods, excess fructose, unprepared grains (especially flour), industrial seed oils, environmental toxins, sedentary behavior, stress, infections and dysregulated gut flora.

But the modern lifestyle doesn’t cause obesity in all people. I’m sure we all know someone who eats a horrible diet, doesn’t exercise, is under tons of stress and lives a shockingly unhealthy lifestyle – but doesn’t gain a single pound.

That’s where genetics come in.

Human evolution didn’t stop in the Paleolithic

A commonly held belief in the Paleo-sphere (I held it myself, until fairly recently) is that our genes haven’t changed much since the Paleolithic era. But recent evidence suggests a much more rapid pace of genetic change in humans than was previously estimated.

The birth of agriculture introduced significant selection pressure, and thus mutation, because humans were not well-adapted to this new way of life. And the evolutionary response to agricultural diet differs because different peoples adopted agriculture at different times and in different places.

Agriculture began in the Middle East 10,000 years ago, but it was never adopted by Aboriginal Australians. Might we expect the descendants of people from these two regions to have different responses when exposed to agricultural diets?

Absolutely. Researchers in Iceland have discovered a gene that regulates blood sugar tolerance. (I discussed the role of genetics in obesity and diabetes in a previous article, Are You At Risk For Diabetes and Obesity?) And we know that Aboriginal Australians have a 4 times greater risk of developing adult-onset diabetes than Australians of European descent.

Lactase persistence is another example. During Paleolithic times, humans stopped producing lactase (the enzyme required to digest lactose, the sugar in milk) shortly after weaning. There was no need for it, since Paleo people didn’t raise cattle or drink milk. Skeletal remains from northern Europeans 8,000 – 9,000 years ago confirm that there was no lactose tolerance at that time.

However, skeletal remains from northern Europeans living in the Bronze Age 3,000 years ago show roughly 25% of adults produced lactase. And today, in certain Scandanavian countries, more than 95% of adults are now lactose tolerant. ¹

All of these genetic changes happened within the last 8,000 years, after the advent of agriculture.

The hand you were dealt: life isn’t always fair

What this means is that some of us are likely better adapted to the modern lifestyle, while others are more susceptible to being harmed by it. Those are very likely the ones that become obese when exposed to a western diet.

But as far as I can tell, they didn’t get obese by eating natural, whole-food carbohydrates. I’ve yet to see a population that got fat eating sweet potatoes, fruit and white rice – without any exposure to modern food. If anyone knows of such a population, please let me know.

Does it even matter what causes obesity? I just want to lose weight!

Some might argue that this discussion is irrelevant, since once someone become obese it’s clear their metabolism is damaged. There are two main problems with that argument.

First, not all obese people have deranged metabolisms. Research over the past several years has defined a subset of “metabolically healthy obese” (MHO) people with normal fasting glucose, triglycerides, insulin sensitivity and other markers. I wrote about this in my article Not All Fat People Get Diabetes, and Not All Diabetics Are Fat.

Second, separating the cause and treatment of obesity is necessary to prevent confusion. Something I see all the time in my practice and in the blogosphere is normal or even underweight people following zero- or very-low-carb diets. Why? Because they’ve absorbed the notion that “carbs are bad” from the “carbs-insulin-fat gain” theory, and they avoid them in a misguided attempt to promote health. While this may work for some people, it doesn’t for many others. I know because they end up coming to me with complaints like low energy, hair loss, bad breath, constipation and more.

Food reward vs. the carbohydrate hypothesis: setting the ground rules

I’d like to see a discussion of obesity that acknowledges the difference between cause and effect, considers the varying impact of food on the metabolically healthy and unhealthy, and recognizes the role of genetics in weight regulation.

Unfortunately, these important distinctions seem to be missing from the current debate – which, in my mind, makes it far less compelling.

As a clinician, my primary goal is to help my patients. While I am fascinated by scientific theories, I am far more interested in whether those theories can be applied to practical solutions for the patients that I work with.

I feel grateful for this perspective, because I believe it helps me to keep an open mind and not get too attached to one theory or another. My patients are constantly teaching me that no matter how good a theory or hypothetical treatment is, real live people sometimes do not respond in the ways that textbooks or studies tell us they should.

When I first read Stephan Guyenet’s series on the role of food reward and palatability in fat gain, I was intrigued. As far as I know, it’s the only theory of weight regulation that can embrace and explain all of the other prevailing theories. For example, food reward can explain why both low-fat and low-carb diets can be effective for weight loss on a short-term basis, and provide insight into why they often fail over the long-term.

Another reason it appealed to me is that I have a number of patients that have tried numerous diets (low-carb, low-fat, Paleo, ketogenic, etc.), but still haven’t lost the weight they’d like to lose. I can’t help feeling curious about whether reducing reward might be of benefit to them. (Note: we will not be testing that particular hypothesis in this study. See below for details.)

Stephan has outlined a substantial body of evidence supporting the role of food reward/palatability in fat gain, and some suggesting that reducing it can be helpful for fat loss as well. But what we’re missing is a study evaluating the possibility that reducing food reward (without resorting to liquid diets) can be a practical fat loss strategy in free-living people.

To remedy this, Stephan and I designed a randomized, controlled food reward fat loss trial. But we’re still missing one thing: volunteers! Please read the study design and volunteer criteria below. If you’d like to apply, click the “Volunteer Application Form” link at the end of the post.

Study design

We’re looking for twelve overweight/obese volunteers (men or women of any age) who would like to eat a low-reward diet for one month, in the name of science. The diet will focus on simple, gently cooked food, and minimize flavorings including salt, herbs and spices, added fats and and added sugars. It will also minimize flour products, which have a high energy density and tend to be highly rewarding/palatable. It will strive to leave macronutrient ratios (carbohydrate, fat and protein) unchanged. The control group will be asked not to change diet or lifestyle over the course of the month. Volunteers will be asked to report body weight and waist circumference measurements, and fill out two short questionnaires (adherence, hunger, well-being, etc.). Data will be analyzed and reported publicly, but they will be reported in a 100% anonymous manner.

Since this is a randomized trial with an experimental group (n=7) and a control group (n=5), each volunteer will have a 58 percent chance of being selected for the experimental group, and a 42 percent chance of being in the control group. Unfortunately, the nature of a randomized trial means you don’t get to choose which group you end up in, but that’s a critical element of the study design for statistical reasons.

One limitation of this study is that if we do see fat loss in the intervention group, we will not be able to conclude that food reward/palatability is the critical factor, since other variables may change at the same time (e.g., sugar). We don’t have the resources to conduct a study that alters nothing but reward/palatability, which would require an experimental kitchen, a professional staff and local volunteers. However, our study will be able to answer the question “is advice to reduce the reward/palatability of the diet useful for fat loss”? It’s more of a practical question than a mechanistic one.

Volunteer criteria

Here are the criteria you must meet to qualify:

• must be sufficiently motivated to complete a one-month diet trial
• must carry substantial excess body fat (25+ lbs / 11+ kg excess fat). This correlates approximately with a body mass index of 25 or more in a person of average muscularity (BMI calculator: http://www.nhlbisupport.com/bmi/)
• must have an accurate scale and a tape measure
• must eat mostly food cooked at home
• must not rely heavily on processed convenience food and restaurants
• no diagnosed endocrine disorder (diabetes, hypothyroidism, etc.)
• must not currently be weight reduced relative to a prior weight
• must not currently be on a weight reducing diet (low-carbohydrate, low-fat, Paleolithic, Zone, Ornish, etc.)

If you meet all of these criteria, please click the link below to volunteer for the study. If you don’t meet the criteria yourself, but have a friend or family member that might, please forward them this post.

Volunteer Application Form

One of the most hotly debated subjects over the past few years has been the cause of the obesity epidemic, and along with that, the best strategy for weight loss.

Some folks (Atkins, Taubes, Eades, etc.) believe that carbohydrates are to blame. Others (Ornish, Campbell, Esselstyn, Fuhrman, etc.) believe that fat is the problem. More recently, researchers like Seth Roberts and Stephan Guyenet and clinicians like Dr. Sharma have raised awareness of another hypothesis, called the food reward theory, which holds that the consumption of highly palatable foods leads to overeating and weight gain. And Paul Jaminet and others have argued that micronutrient deficiencies, toxins and infections may play a significant role in the obesity epidemic.

Here’s what I think: the most accurate answer to “why do people get fat?” and “what’s the most effective weight loss strategy?” is: “it depends.”

Separating cause from mechanism and effect

One of the biggest mistakes often made in this debate is the confounding of cause, mechanism and effect. A classic example is the assumption that if reducing carbohydrate or fat intake leads to weight loss, then the original weight gain must have been caused by excess carbohydrate or fat consumption.

While it’s tempting to make such an assumption, the logic is faulty. It’s kind of like saying “Advil cures headaches. Therefore, headaches must be caused by Advil deficiency.”

Let’s look at some definitions.

Cause: something that brings about an effect or a result

Mechanism: the fundamental processes involved in or responsible for an action, reaction, or other natural phenomenon

Effect: an outward sign

Obesity is an effect. Insulin resistance, leptin resistance, lipotoxicity, disruption of the mesolimbic dopamine reward pathway and inflammation of the hypothalamus are presumed mechanisms. Excess consumption of carbohydrates, fat, highly palatable food and food toxins (wheat, seed oils, liquid fructose, etc.), exposure to environmental toxins (chemicals), stress, infections, etc. are presumed causes.

Say we do a study on obese people and we observe that they eat a lot of carbohydrates and are insulin and leptin resistance. It’s easy to assume that the chain of causality worked like this: normal weight person eats high-carbohydrate diet, becomes insulin and leptin resistant, and then becomes obese.

But again, this is faulty logic. There’s no proof that A (high carbohydrate intake) was what led to B (insulin and leptin resistance) was what led to C (obesity).

In fact, we could disprove that theory simply by observing another individual or group that eats a very high carbohydrate diet, but does not develop insulin or leptin resistance and obesity. Guess what? Such individuals and groups most certainly exist. There goes that theory.

Likewise, we could also disprove this theory by observing people that are insulin and leptin resistant, but don’t become obese. Such people do exist, and I’ve written about them in my series on diabesity here.

A more rigorous approach

How have we developed our theories on obesity and weight regulation? It seems to me they come from a blend of personal experience, belief and facts. And I think it’s time to become more rigorous about keeping them separate. Here’s an example of what I mean:

Personal experience: I lose weight on an low-carb diet, therefore low-carb diets must be best for weight loss.

Belief: carbohydrates are responsible for the obesity epidemic, via their effects on insulin.

Fact: many cultures around the world eat high-carbohydrate diets and are exceptionally lean.

Those who’ve lost a lot of weight on a low-carb diet have a tendency to become convinced that their wife, friends, family, plumber and everyone else will also lose weight following the same diet.

From this personal experience, a belief is formed. And once we believe in something, we have a remarkable ability to filter out any evidence that might contradict that belief.

This is especially true if our reputation or financial livelihood is tied to said belief. As Upton Sinclair famously said:

It’s difficult to get a man to understand something when his salary is dependent upon him not understanding it.

When a belief like “carbs cause obesity” is shared between enough individuals, it becomes a meme. Once that happens, it is accepted by most as fact – regardless of whether it has any scientific basis. Hence we had the idea for decades that eating fat makes you fat, and now the more recent idea that eating carbs makes you fat.

There’s no single cause (or treatment) of obesity

Perhaps one of the reasons it’s so easy to confuse cause, mechanism and effect and personal experience, belief and fact is that obesity is an incredibly complex disease. Just how complex is it?

Click on the Obesity Systems Influence Diagram below to find out.

Click image for larger version

Wow. That should give you a rough idea of how many variables are potentially involved in weight regulation. Now you know why it has been such a challenge to come up with a single, unified theory of obesity.

That said, of all of the hypotheses advanced to explain the mechanisms behind obesity, I think the food reward theory is the most inclusive.

However, as even its proponents would agree, it doesn’t tell the whole story because there are people and groups that eat large amounts of highly palatable foods that do not become obese.

My opinion is that the modern lifestyle (i.e. food and environmental toxins, stress, poor gut health, infections, micronutrient deficiencies, sleep deprivation, etc.) interfere with hypothalamic hormonal regulation, dopamine signaling, leptin and insulin sensitivity at the cellular level, glucose metabolism and a range of other mechanisms that lead to obesity.

This is consistent with the observation that obesity is extremely rare or nonexistent in traditional cultures that do not consume modern foods and do not live a “modern” lifestyle.

But even this theory is incomplete, because there are people fully exposed to the modern lifestyle that do not become overweight or obese. This suggests that genetics, and perhaps other undiscovered factors, also play a role.

We’re not robots

Humans are not robots. We’re living, breathing, dynamic organisms influenced by varying genetics and environmental conditions.

Anthropological evidence combined with modern research has helped us to reveal the basic template of a species-appropriate diet. However, it has also shown us that humans can thrive on a wide variety of macronutrient ratios and foods within that basic template.

This is not a belief. It’s a fact, supported by the evidence as a whole. Ignoring the evidence doesn’t make it go away. Believing passionately in something doesn’t make it true. Experiencing something personally doesn’t make it fact for everybody else.

19th century philosopher Charles Peirce said:

The state of belief is a calm and satisfactory state which we do not wish to avoid, or to change to a belief in anything else.

And Tolstoy said:

I know that most men, including those at ease with problems of the greatest complexity, can seldom accept even the simplest and most obvious truth if it be such as would oblige them to admit the falsity of conclusions which they have delighted in explaining to colleagues, which they have proudly taught to others, and which they have woven, thread by thread, into the fabric of their lives.

Recognizing this basic human trait, philosopher of science Karl Popper advised every researcher to earnestly try to discredit their own hypotheses.

That is no easy task, and it asks a lot of us. Yet intellectual rigor, emotional maturity and personal integrity are characterized by the capacity to question our own beliefs, no matter how deeply cherished they are or how much is at stake.

I sometimes wonder why we’re all so sure of ourselves. It helps me to remember that at every point in history scientists (and the general public) were convinced they had the right answers. At one time the world was flat, the earth was the center of the solar system and disease was caused by foul humors and could be cured by bloodletting.

Nowadays we look back on those fallacies with a smirk. But are we so arrogant to assume that our great-grandchildren won’t do the same?

The truth is, there’s far more we don’t know than we do know.

A study was just published in the New England Journal of Medicine on July 17th comparing the effectiveness and safety of three different weight loss diets. 322 moderately obese subjects were assigned to one of three diets: low-fat, restricted-calorie; Mediterranean, restricted-calorie; or low-carbohydrate, non-restricted calorie.

The rate of adherence to the study diet was 95% at year one and 85% at year two. Among the 272 participants who completed the intervention, the mean weight losses were 3.3 kg for the low-fat group, 4.6 kg for the Mediterranean-diet group, and 5.5 kg for the low-carbohydrate group.

Perhaps more significantly, the relative reduction in the ratio of total cholesterol to HDL was 20% in the low carbohydrate group while only 12% in the low-fat group. Among the 35 subjects with diabetes, changes in fasting plasma glucose and insulin levels were more favorable among those assigned to the Mediterranean diet than among those assigned to the low-fat diet.

Unfortunately, the bias against saturated fat and animal products that is still so prevalent in the mainstream (in spite of the lack of evidence to support it) prevailed in this study. The research team advised those following the low-carb diet to “choose vegetarian sources of fat and protein” and moderate their consumption of saturated fats and meat.

This suggests that the low-carb dieters may have consumed a substantial portion of their calories as fat in the form of omega-6 polyunsaturated fatty acids. Excess intake of omega-6 fatty acids contributes to a host of problems including heart disease, diabetes, and cancer; but even more relevant to this study and its results is the fact that omega-6 fatty acids can cause increased water retention. And as everyone knows, increased water retention equals increased weight.

This certainly causes me to wonder how much more dramatic the results of this study might have been if the low-carb subjects were encouraged to significantly restrict their consumption of omega-6 fats (which cause water retention, and thus weight gain) and replace them with saturated fats (which do not cause water retention). What is remarkable is that in spite of the consumption of omega-6 fats, the low-carb group still lost more weight than both the low-fat and Mediterranean groups. That’s a strong endorsement for the benefits of a low-carb diet for weight loss.

The low-carb and Mediterranean (to a lesser degree) diet also had other benefits beyond promoting weight loss and improving cholesterol measures. The level of high-sensitivity C-reactive protein decreased significantly only in the Mediterranean and ow-carb group, with the low-carb group again showing the greatest decrease (29% vs. 21%). C-reactive protein is a measure of inflammation that has been positively correlated with heart disease in recent studies. Once again, one must wonder if the reduction would have been even greater in the low-carb group had the subjects been told to restrict their intake of omega-6 fats, which are known to promote inflammation.

Another interesting finding is that although caloric intake was only restricted in the low-fat and Mediterranean diet groups, the low-carb group also ended up eating fewer calories during the diet. Many people who follow a low-carb, high protein/high fat diet find that they spontaneously eat less because additional protein, and in particular fat, leads to greater levels of satiety (satisfaction).

One limitation of the study is that it relied on self-reported dietary intake (this is true of almost every dietary study except those performed in tightly controlled conditions, such as an inpatient facility). However, the study was somewhat unique in that it was conducted in a workplace at a research center with an on-site medical clinic. It also had several other strengths. The drop-out rate was exceptionally low for a study of this kind; all participants started simultaneously; the duration was relatively long (2 years); the study group was relatively large; and the monthly measurements of weight remitted a better understanding of the weight-loss trajectory than other studies.