Listen to this article
As many of you know, I’m currently at the Ancestral Health Symposium at Harvard University. Yesterday I participated on a panel organized and moderated by Jimmy Moore called “Safe Starches: Are They Essential on an Ancestral Diet?” The panelists were myself and Paul Jaminet on the “pro-starch” side, and Dr. Ron Rosedale and Dr. Cate Shanahan on the “anti-starch” side (though Cate’s position is not quite as cut-and-dry as Ron’s).
I’m giving my talk on iron overload today, so I don’t have a lot of time, but I wanted to at least summarize the “anti-starch” side’s arguments and then list some bullet points of my arguments in favor of starch for those of you who aren’t here. I’m not sure if the panels will be made available after the fact (the talks will be).
Ron and Cate believe that glucose is toxic in any concentration, and it’s just a matter of scale. In fact, Ron is fond of saying that “everyone is diabetic”. Since starch breaks down into glucose, then by definition starch is toxic and should be avoided – by everyone. I’m a little less clear on Cate’s position, but she seemed to argue that glucose raises insulin, and insulin causes problems, so everyone should be on a low-carb diet ranging from 20 – 70 grams of carbohydrate a day, starch included.
My arguments in favor of starch
- Let’s define the terms: are we debating whether starch is “safe” in healthy people or people with particular health conditions like diabetes or small-intestine bacterial overgrowth? These are very different conversations. People with hereditary hemochromatosis (a disorder that causes iron overload) should not eat iron-rich foods like liver and mussels; does that mean everyone should avoid these foods? Even if starch/glucose is “toxic” for diabetics, should everyone avoid starch/glucose?
- If the argument is that starch is not safe for healthy people, I would say there’s little to no scientific or anthropological evidence to support that idea, and overwhelming evidence opposing it.
- There are literally billions of people eating high-starch diets worldwide, and you can find many examples of cultures that consume a large percentage of calories from starch where obesity, metabolic problems and modern, inflammatory disease are rare or nonexistent. These include the Kitava in the Pacific Islands, Tukisenta in the Papa New Guinea Highlands and Okinawans in Japan among others. The Kitavan diet is 69% carb, 21% fat, and 10% protein. The Okinawan diet is even more carb-heavy, at 85% carb, 9% protein and 6% fat. The Tukisenta diet is astonishingly high in carbohydrate: 94.6% according to extensive studies in the 60s and 70s. All of these cultures are fit and lean with low and practically non-existent rates of heart disease and other modern chronic disease.
- Amylase is thought to have played a key role in human evolution in allowing humans an alternative to fruit and protein. Compared with primates, humans have many more copies of a gene (AMY1) essential for breaking down calorie-rich starches. The ability to digest starch, along with the discovery of fire and cooking, gave humans a new food source that allowed us to thrive even in marginal environments. Some scientists have even argued that consumption of starch, along with meat, was primarily responsible for the increase in our brain size.
- Dr. Rosedale argues that evolution is optimized for fertility, not longevity, and that starch consumption decreases longevity. The evidence he cites from this come from studies of roundworm, C. Elegans. However, I am not aware of any evidence in humans showing that starch consumption decreases longevity, and some of the longest lived cultures in the world consume large amounts of starch. Okinawans over the age of 65 (who grew up eating a traditional diet) are a prime example. According to a study of the traditional Okinawan diet in 1949, they obtained 85% of calories from starch, mostly from sweet potato. Life expectancy was 86 years for women and 77.6 years for men. Life expectancy at age 65 is the highest in the world, at 24.1 years for females and 18.5 years for males. Finally, the Okinawan population has the highest prevalence of centenarians in the world. This is especially remarkable when you consider that Okinawans did not have access to modern medical care during the 40s & 50s and and higher rates of death due to infections like tuberculosis as a result. If glucose is toxic and promotes short lifespan, how do the Okinawans live so long?
- There is no one-size-fits-all approach. The amount of starch (and carbohydrate in general) will depend upon genetic/epigenetic factors (like amylase production), existing health conditions and the volume and intensity of activity – among others.
- If the argument is that starch isn’t safe for those with impaired glucose tolerance, I concede that may be true in many cases. However, I’d like to point out that there’s some evidence that suggests starch may be safe in this population as well. For example, low-fat diets also cause fat loss (even without deliberate calorie restriction), though to a lesser extent than low-carb diets. And there are documented cases of people losing significant amounts of weight and improving metabolic parameters by eating nothing but potatoes. For example, Chris Voigt lost 21 pounds over the course of two months by eating only potatoes and not deliberately restricting calories. Furthermore, his fasting glucose decreased by 10 mg/dL (104 to 94 mg/dL), his serum triglycerides dropped by nearly 50%, his HDL cholesterol increased slightly, and his calculated LDL cholesterol dropped by a stunning 41% (142 to 84 mg/dL).
There’s more, but I don’t think it’s necessary to go further. If Drs. Rosedale and Shanahan are going to advise us to avoid an entire class of food that has been eaten for a couple of million years by humans, the burden of proof is on them to tell us why that food isn’t safe. Evidence from roundworm experiments and biochemical/mechanistic speculation is not enough in the face of overwhelming evidence that starch and glucose are safe in the absence of certain existing health conditions.