Reading Is Good for You

Reader Charlie sent me a link to this article in the September 2007 Scientific American, in which professors from the Harvard School of Public Health argue that the idea of fat not being intrinsically unhealthy is "complete nonsense." Specifically, it criticizes a 2005 paper by Katherine Flegal et al, published in the Journal of the American Medical Association. People in the fat acceptance community know that paper well: it was the big one that said people in the overweight category have the lowest mortality rate, and suggested that previous estimates of the death toll of obesity were off by, oh, a couple or three hundred thousand a year.

For some reason, Scientific American decided it was time, two years after the study was published, to attack Flegal's research. So they gave the boys at Harvard a jingle.

[Harvard professor Walter] Willett thinks this assertion is simply the latest recycling of the notion that Americans have been somehow duped about the risks of obesity. “About every 10 years this idea comes along that says it’s better to be overweight. And we have to stomp it out,” he says. Willett’s research has identified profound advantages to keeping weight down—even below the so-called healthy levels.

That's the kind of quote that might have caused me to despair if I hadn't remembered Willett's name from Rethinking Thin. See, there's a whole section in Kolata's "The Fat Wars" chapter (pp. 201-209 in the hardcover) about how, after Flegal published that paper, the Harvard School of Public Health went apeshit. They even

held a seminar to refute it, making sure that newspaper reporters were not only invited to the meeting but able to watch it and listen on a Webcast if they could not attend in person. It was an exercise in attack science.

A seminar to refute one paper. Why on earth would they do that?

Well, The Harvard School of Public Health was home to one of the most important previous studies of the health effects of fat, a long-term one done on nurses. The nurses' study showed that being fat carried a substantially increased mortality risk. But Flegal's paper pretty much said, "You did it wrong and you got it wrong."

So that's pretty much what they said back to her in that seminar. And are now saying 2 years later in Scientific American.

The difference between the Scientific American article and Rethinking Thin is, Gina Kolata actually talked to Katherine Flegal and the other researchers with whom she published that 2005 paper. So, for instance, where the SA article just reports the Harvard boys' assertion that Flegal's study was fucked because she didn't exclude smokers or the chronically ill, Kolata adds that A) Flegal deliberately didn't, because she felt that previous studies (like, uh, the Harvard nurses' study) had cherry-picked their subjects, and she wanted a group that was actually representative of the general population, and B) once she and her co-researchers had all that raw data,

They looked at the results both with and without current or former smokers and people who had chronic diseases. They posted the extensive analysis on the Internet -- journals like the Journal of the American Medical Association allow for only so much additional data -- and the results always came out the same: There was no mortality risk from being overweight and little from being obese, with the exception of the extremely obese, whose death risk was slightly higher.

Here's how Scientific American presented that:

Flegal has acknowledged that she did not exclude the chronically ill from her study but argued in a follow-up report that she had done further analyses that showed it would not have made a difference. The disagreement turns on subtle statistical arguments. What is clear, however, is that Flegal’s paper is one of a handful that contradict many studies that support the conclusion that being overweight is harmful. Flegal is not necessarily wrong, but the preponderance of evidence clearly points in the other direction.

Do you see what just happened there? First, they make it seem as if Flegal acknowledged a mistake and scrambled to compensate for it, as opposed to acknowledging only that she deliberately included those populations and then analyzed the data both with and without them. Then, they bury the fact that she did it on purpose under the real point of this article: a gazillion obesity researchers can't be wrong!

Seriously, that's the entire argument the article makes. "Flegal is not necessarily wrong," but boy, a whole lot of other studies found something else! Um, yeah, that was kind of the point of Flegal's study. Flegal and David Williamson, an epidemiologist at the CDC, looked at previous studies, determined that those had gotten the statistics wrong, and designed a study that made a good faith effort to get the statistics right. Oddly enough, their conclusions turned out vastly different from those of previous studies. And two years later, this article is here to tell us Flegal, Williamson, et al, are wrong because... their conclusions are vastly different from those of previous studies.

Here's Kolata again:

Flegal says she had a real education in the politics of obesity.

"Everyone thinks they already know the answer," she says. "Anything that doesn't fit, they have to explain it away or ignore it. All these people who just know weight loss is good for you. It's just taken for granted regardless of the evidence." She was not naive about her findings, she said. "I expected people would get perturbed, but I really didn't expect the way they did it. All these erroneous so-called fact sheets. And these misinterpretations and making up things we'd said."

Paints a slightly different picture than the Scientific American article, don't it?

And how many people who read that article are going to know off the top of their head that The Harvard School of Public Health A) did one of the studies Flegal and Williamson deemed egregiously flawed before beginning their research, and B) has apparently had a hate-on for them ever since their paper about that was published?

Not many, I'm guessing. But Gina Kolata took the time to look into Flegal and Williamson's side of the story -- like, you know, an actual journalist might -- so I knew it off the top of my head, and now you know it, too. And it changes everything, doesn't it? Knowing that, you can see quite clearly that the SA article isn't actually saying Flegal's wrong -- in fact, the author takes pains not to say that -- it's just saying a whole lot of people in the obesity research community don't agree with her. And I mean, duh. That's worth writing about two years later?

So the Scientific American article did not remotely convince me that Flegal is wrong and fat is bad for you. It did, however, strongly reinforce one thing I've believed to be true for most of my life: reading is good for you.