Good Calories, Bad Calories
by Gary Taubes
My rating: 4 of 5 stars
In this book, Gary Taubes looks at how nutrition scientists in the 60s and 70s developed the hypthesis that dietary fat was responsible for rising rates of cardiovascular disease, diabetes, and obesity; how this untested hypothesis was used to make sweeping dietary recommendations to the American people; and how subsequent research has demonstrated that this hypothesis was, at best, an oversimplification. Taubes then proposes his own hypothesis: that insulin, as a key regulator of carbohydrate and fat metabolism, is more important than was previously recognized in the development of heart disease and obesity. (Of course, insulin's role in diabetes has long been recognized.) Or, in short, it's not the fat, it's the carbs. Taubes acknowledges that the carbohydrate hypothesis is still largely untested, but there is enough interesting biochemical evidence to make it seem worth further study.
Regardless of what you think of the carbohydrate hypothesis, Taubes's book is well worth reading for one of the best and clearest explanations I've ever seen for why scientific study of nutrition and health is so problematic. He lays out several problems that are worth keeping in mind when looking at any of the evidence in this book, whether it's on the side of the fat hypothesis or the carbohydrate hypothesis:
First, scientists study what they can measure. In the early days of research into the effects of diet on heart disease, total cholesterol was much easier to measure than the different subtypes of cholesterol. And so total cholesterol was what researchers measured, and tried to correlate against incidence of heart disease. We now know that it makes a difference whether the cholesterol is HDL, LDL or triglycerides, but they couldn't measure those in the early studies.
Second, a properly controlled double-blind study is almost impossible to conduct. You can almost never vary just one variable at a time - if you take something out of someone's diet (like saturated fat), you have to replace it with something else, or total calories fall. And in most cases, the double-blind part is impossible - people generally know what they're eating.
Third, if you want to tease out meaningful statistical differences in disease outcomes or life expectancy, you have to get together a huge group of study participants and follow them for a very long time. Because this is both time consuming and expensive, it's very tempting to rely instead on observational studies where you look at populations with different diets and look at relative rates of disease. This means that a surprising amount of what we think we know about diet and diseases comes from studies of the Japanese, Greek fisherman, and Finnish lumberjacks. All of whom probably differ from modern Americans in lots of ways that don't have anything to do with diet.
However, by laying out the flaws of nutrition science so clearly, Taubes somewhat undermines his own argument. Because he then goes on to deploy a barrage of studies and evidence that it's really carbohydrates that are the dietary demon, and the skepticism about nutritional studies that I learned in the first part of the book remains with me through the rest of it. To be fair to Taubes, he acknowledges that his hypothesis about carbohydrates has not been properly tested. And he never once presumes to advise or instruct the reader on how they should eat.
I did find Taubes pretty convincing on the following points:
*That rates of heart disease, diabetes, and obesity in America statistically correlate at least as well or better with consumption of carbohydrates (especially sugar, high fructose corn syrup, and other refined carbohydrates) as with consumption of fat or saturated fat.
*That the importance of insulin in regulating the balance between burning calories in the muscles and storing them in fat cells is well-established enough from biochemical studies to warrant further studies to see how this contributes to the development of obesity in humans.
*That low carb diets might not be as nutritionally deficient as doctors and nutritionists schooled in low-fat-diet orthodoxy might think, and probably merit further study.
Will this book change the way I eat? If I hadn't already cut back the added sugars, high fructose corn syrup, and refined carbohydrates in my diet to a fairly low level, this book would have persuaded me to do so. Beyond that, while Taubes has convinced me that I might be thinner if I further cut back on the legumes, fruit, and whole grains that now contribute the bulk of carbohydrates in my diet, he hasn't convinced me that I'd be healthier, happier, or more satisfied with my diet.
I will be watching the nutrition news, though, to see how further research on this questions plays out. View all my reviews