Anonymous, January 22, 2008 7:28 AM
Bubba Links
Baseball Links
Not much going on baseball land today, it being Sunday, and a big day for the NFL. I could blog about today's playoff games, I suppose, but I'm just too depressed after the Cowboys' early exit. 
Other bloggers blog about pop culture on slow days: music, movies, TV shows, books. But my taste in music is boringly bourgeois, I rarely go to the movies, and I watch mostly news and sports on TV. That leaves books. I do like to read, and these days, my taste runs heavily towards science. (Why yes, I am nerd. ;-) I'm reading a truly fascinating science book right now - one that I think will prove to be tremendously important. But it's not much to do with baseball. You've been warned. I'm drifting off-topic today.
The book is Good Calories, Bad Calories
Taubes first broached the idea that low-fat diets might actually be bad for you in a 2002 NY Times article called What if It's All Been a Big Fat Lie? Now he's come out with an entire book on the topic, and it's a masterpiece.
Good Calories, Bad Calories argues, persuasively, that we've been seriously misled over the past 30 or 40 years. It was with the best of intentions - but wrong. Doctors were so eager to help people that they got ahead of the science.
There's more and more evidence that, contrary to what we've been told, it's not dietary fat that causes obesity, cancer, and heart disease. Rather, the problem is carbohydrates - especially refined carbohydrates like sugar, white flour, rice, and pasta.
This is an astonishing claim, but Taubes has the evidence to back it up, and he lays it out in detail. So much of what I thought was proven science on nutrition turns out to be built on shaky ground indeed. Fiber doesn't prevent colon cancer. Fat isn't bad for you - not even saturated fat. High cholesterol, even "bad" cholesterol, isn't the cause of heart disease. The "Mediterranean Diet" may be healthful because it is low in refined carbohydrates, not because it's high in vegetables and olive oil.
How did they get it so wrong? A lot of it has to do with the problems that come with science that involves people's health. There are limits on what kinds of experiments you can do on humans. And animals may not be good stand-ins. (For example, one of the early influential cholesterol studies used rabbits. But rabbits are herbivores. They don't normally eat cholesterol.) Many of the early researchers were doctors, not scientists; they did not know how to properly design studies. And they were so eager to save lives, that they felt they couldn't wait for proof.
But perhaps the greatest factor was mistaking correlation for causation. The classic example of this logic error: claiming that people going through the revolving doors of Macy's department store power the escalator inside. The proof? When you look inside the store when the doors are locked, the escalator is no longer moving!
Similarly, many of the studies of diet and health found a correlation, but did not prove causation. In particular, fat and meat consumption often go up with sugar consumption, so it's hard to say which one is the cause of health problems. Also, for international studies, many of the populations that ate a lot of carbs and not much meat and fat were chronically undernourished. (It's known now that calorie restriction can prolong life. But it's not easy to half-starve yourself your whole life - and maybe, if you eat the right kinds of calories, it's not necessary.)
There was also a lot of "confirmation bias" - the human tendency to interpret data in the light of one's preconceptions, and ignore anything that doesn't fit. An example was a study that clearly showed that high cholesterol was not linked with heart disease, but low cholesterol was strongly linked with cancer. That so contradicted expectations that the researchers simply refused to believe it.
If Taubes is right, then what we've been told for the past 30 years or so is the opposite of what we should be doing. I don't know if he's right or not, but intuitively, I suspect he's onto something. There are so many things that make sense in the light of his work. The so-called "French paradox": the French have fewer heart attacks than Americans, despite eating a lot more butter and cream, and smoking more. That Inuit eating their native diet of red meat and blubber don't get heart disease (and no, it's not fish oil). That Americans have gotten fatter, despite cutting back on fat and red meat, and eating more carbs. The fact that farmers who want to fatten their hogs feed them skim milk; they don't get fat if you feed them whole milk.
Back in the '50s and earlier, it was common knowledge that sweets and starches were what caused weight gain. The "diet plate" offered at restaurants was a hamburger patty, with a scoop of cottage cheese instead of a bun, and no fries. Maybe grandma was right after all...
Labels: books, off-topic, reviews, science of sports
