I’m working on a post about the role of serotonin in the development and maintenance of anorexia, but it is taking me some time as I want to include a sufficient amount of background information. So, in the meantime, I’m going to blog about a short paper that was brought to my attention by Sarah. As you might have guessed by the title, the paper’s author, Casper Schoemaker, wanted to do some fact-checking on Naomi Wolf’s first book, titled The Beauty Myth. A quick glance at the Amazon ratings reveals that a lot of people like it, and many find it shocking and “eye-opening”. Now, I’ll admit: I haven’t read it, and I don’t plan to. (I don’t read books like this.) But I have come across blatantly wrong statistics on eating disorders from people citing her work.
(Books are not peer-reviewed. The main thing you need to get a book published is to convince the publisher that people will buy it. There’s no guarantee that the claims in a particular book are correct. A lot of times they are not. See Kevin Trudeau, though not everyone lies on purpose.)
If you are wondering who cares? I mean, lots of people get things wrong all the time. And that’s true. But, Wolf is a public figure, her statistics are heavily cited in the media, and the media has a huge impact on the public understanding of eating disorders, and even on clinicians’ attitudes toward patients with EDs. She is not helping the feminist cause nor the eating disorder awareness campaigns by fudging facts.
In her first book The Beauty Myth (Wolf, 1990), the American feminist writer Naomi Wolf asserted that the second feminist wave had not been very successful. In 1990, according to Wolf, women were underpaid, and not judged by the quality of their work, but by their looks. Most well educated women suffered from the eating disorders anorexia and bulimia nervosa, and were forced to end their education. The few women that did not diet put all their money in cosmetic surgery. And the purported cause of all this was the “Beauty Myth.”
Thanks to her debut book, Wolf obtained instantaneous fame all over the world. She traveled, gave lectures, took part in discussions, and appeared on television. Her book was a best seller. In the public debate the anorexia statistics in her chapter “Hunger” were frequently talked about. The eating disorder figures were dramatically high: according to Wolf (1990), 20% of American female students suffered from anorexia and 60% from bulimia. Only small minority had no eating disorder! Her mortality figure—150,000 deaths from anorexia each year in the U.S.—caught the most attention in the media…….
In 1994, professor in philosophy Christina Hoff Sommers—not an eating disorder expert—responded in her book Who Stole Feminism? (Sommers, 1994). She tracked down the mortality figure to the source. Wolf had cited a book by Brumberg, who had referred to a newsletter of the American Anorexia and Bulimia Association (AABA). Brumberg misquoted this newsletter, however: The AABA had referred to 150,000 sufferers (not fatalities) from anorexia nervosa.
So anyway, Wolf corrected herself in later edition of the book. But, was that all? Schoemaker wanted to investigate all the other statistics that Wolf cited (in the newest edition of the book) and compared them with findings in peer-reviewed literature.
Schoemaker relied on statistics from epidemiological reviews (Hoek, 2002 and Van Hoeken et al., 2003). To illustrate Wolf’s absurd statistics, Schoemaker calculated the factor by which Wolf was wrong, called WOLF (Wolf’s Overdo and Lie Factor). An example: taking Wolf’s original 150,000 annual deaths due to AN, and dividing it by the real number (around .56% of AN patients die a year, which is about 525), you get 286.
Naomi’s original statistics was almost 300 times higher than the real number.
Below is my adapted version of the table from Schoemaker’s paper. I lifted the most outrageous WOLF numbers as well as the statistics that Naomi Wolf got right. There’s only one statistics Wolf underestimates: the number of anorexia patients that fully recover (WOLF = 0.81).
Of course, one can argue that the data in Hoek’s and Van Hoeken’s review papers are not very accurate. They might not be. I haven’t checked, for example, the criteria used to assess whether someone is recovered or not (these vary a lot, there’s no agreed upon definition of recovery). Nonetheless, two reviews of many epidemiological studies to date, published in peer-reviewed journals, are likely to have statistics as good as we can get. And given just how wrong Naomi Wolf was in some of the statistics she cites, I doubt a few percent fluctuation either way in the “epidemiological #’s” would make much difference.
The mean WOLF for all 23 anorexia statistics was 8.28. This is after the correction Wolf mentioned in her letter to the editor of Time [about the annual mortality rate]… In summary: On average, an anorexia statistic in any edition of the The Beauty Myth should be divided by eight to get near the real statistic.
In the discussion, Schoemaker makes a few salient points, but the main one is that Wolf is not helping anyone by overdoing the numbers or calling it an “epidemic”. It is not an epidemic, but that doesn’t mean we shouldn’t care about it: “anorexia nervosa is a very serious health problem. Not because of the “epidemic” numbers of patients, however, but because of the devastating effects on the few individuals who develop anorexia.”
Exactly. Clinicians and scientists shouldn’t, and hopefully don’t, start caring about diseases and mental disorders when they reach epidemic proportions. There are lots of health problems that are much rarer than anorexia nervosa that get proper research and medical attention. As they should, because the impact on the patient, their family and society as a whole, is huge. Patients shouldn’t be turned away because their problems are so rare, it is just not worth, from a health economics perspective, to spend money researching the causes and cures, and treating them. Things hopefully don’t work that way. We don’t need to (I hope) exaggerate statistics in a misguided attempt to raise awareness.
I love the snarky ending:
I suggest the experts enter the public debate more often. For that purpose, the WOLF seems to be an easy and useful tool to quantify the exaggerations in the media. Its acronym refers to a writer that should never be trusted with anorexia statistics again.
Schoemaker, C. (2004). A critical appraisal of the anorexia statistics in The Beauty Myth: introducing Wolf’s Overdo and Lie Factor (WOLF). Eating Disorders, 12 (2), 97-102 PMID: 16864310