Healthy Skeptic Podcast and Reader Questions

Chris Kresser, Danny Roddy and I just finished recording the podcast that will be released on May 24th.  It went really well, and we think you'll find it informative and maybe even practical!

Unfortunately, we only got around to answering three of the questions I had selected:

1. How does one lose fat?
2. What do I (Stephan) eat?
3. Why do many people gain fat with age, especially postmenopausal women?

I feel guilty about that, so I'm going to answer three more right now.


What do you think about cutting carbohydrate for fat loss?  Why does my weight increase after adding carbohydrate back to a low-carbohydrate diet?

As carbohydrate is a major factor in food reward, reducing it will allow fat loss in some people.  Long-term studies of low-carbohydrate diets without deliberate calorie restriction indicate that the average person maintains a 5-10 pound weight loss at one year, mostly as a result of fat loss.  Studies show similar trends for low-fat diets unrestricted in calories, although the fat loss is a bit less.

I think that an optimal diet for lean healthy people is probably not restricted in macronutrients, and if anything a diet biased toward carbohydrate is better for overall long-term health than one biased toward fat.  However, obesity and diabetes are important factors in health, and if a person finds that reducing carbohydrate attenuates them, then it's worth considering. 

In susceptible people, body fat increases after adding carbohydrate back to a low-carbohydrate diet because you're increasing the overall reward value of the diet, not because you're spiking insulin more.  Carbohydrate is not inherently fattening, it just depends on the context.  Replacing more fat with carbohydrate in the context of a low-fat diet will not increase the diet's reward value and will if anything accelerate fat loss.  Conversely, while adding fat to a low-fat diet will increase its reward value and favor fat gain, replacing more carbohydrate with fat in a low-carbohydrate diet will if anything favor fat loss.  But there will always be individual variation as the reward value of specific foods differs between individuals.

What is the role of insulin in body weight control?

Insulin has a role in body fat control, but it's the opposite of what many people believe it to be.  In fact, insulin acts in a manner similar to leptin, although less potently:

• When injected into the brain, it reduces food intake and body weight in rodents and primates (1, 2)
• Knocking out the insulin receptor in the brain leads to increased fat mass (3)
• Foods that spike insulin the most in humans lead to the greatest satiety and lowest food intake at subsequent meals (4)
• Insulin is co-secreted with amylin, a hormone that reduces food intake and body weight, probably by increasing leptin sensitivity (5)

Originally, insulin was the leading candidate for the body fat homeostasis hormone.  Why?  Because it increases roughly in proportion to fat mass, it reaches the brain roughly in proportion to circulating concentration, and it reduces food intake and fat mass when injected into the brain (6).  We now know that leptin is the main homeostasis hormone, but insulin does a similar job, just less potently.  Insulin's main job is to coordinate glucose and fatty acid metabolism, and leptin's main job is to regulate long-term energy balance, but each hormone does a little bit of the other's job.

Insulin does act directly on fat cells to promote fat storage, but under normal circumstances, that's not a significant factor in 24 hour fat mass balance if you consider its effect in a whole, live organism (as opposed to cells in a dish).  Type 1 diabetes is an exception, because insulin is completely lacking and fat can't get into fat cells at all, so people become very lean.  Fat cells require some insulin to function normally.  But T1DM isn't exactly a prescription for health!

Is fat mass controlled by a setpoint or a settling point?

There is no agreed-upon definition of what exactly a setpoint and settling point are, but here are two definitions that I find reasonable:

• A setpoint is regulated by an active, centralized system that measures the variable in question and defends against changes.
• A settling point is regulated by the passive confluence of many factors that influence the variable in question, and can also defend against changes.

There are many, many factors that influence body fat mass, so this has lead some to favor the settling point hypothesis.  However, the leptin system is different from all other known factors that influence fat mass (except insulin, see above) in that it's a negative feedback system that measures fat mass and responds to changes in fat mass by attempting to restore it to its "favored" level.  Alterations in other systems can influence fat mass to some extent, as the leptin system reaches a new equilibrium, but I doubt anything can cause frank obesity without the participation of the leptin system, for a number of reasons.  Since we know that the dominant system that regulates fat mass is an active, centralized negative feedback loop, I favor the term "setpoint" because it acknowledges that fact.

One of the objections people have to the term "setpoint" is it sounds immutable and unresponsive to other factors.  I believe the leptin system is modulated by food reward, and probably other factors to a lesser extent.  So you could call it a setpoint that's responsive to other factors, especially food reward, or you could call it a settling point.  I don't feel very dogmatic about it.  Call it what you want.