I ran across this little research tidbit several times this morning:

"Therapy hope for eating disorders."

It's a study in the American Journal of Psychiatry, titled "Transdiagnostic Cognitive-Behavioral Therapy for Patients With Eating Disorders: A Two-Site Trial With 60-Week Follow-Up." The lead researcher was expert Chris Fairburn at the University of Oxford in the UK, who compared two different types of cognitive behavioral therapy (CBT) in treating bulimia nervosa or ED-NOS (participants had to have a BMI greater than 17.5). One type of CBT focused almost exclusively on resolving the disordered eating behavior; the other type of CBT focused both on eating and on other issues like perfectionism, low self esteem, depression, etc.

Treatment consisted of 20 weekly hour-long visits, and had nearly a one-year followup period. What I found interesting was that people did just as well in the CBT that focused on resolving the ED behaviors, indicating that many of the other issues resolved (at least somewhat) as eating improved.

The results seem promising: "At the 60-week follow-up assessment, 51.3% of the sample had a level of eating disorder features less than one standard deviation above the community mean. Treatment outcome did not depend on eating disorder diagnosis. Patients with marked mood intolerance, clinical perfectionism, low self-esteem, or interpersonal difficulties appeared to respond better to the more complex treatment, with the reverse pattern evident among the remaining patients," the study concludes.

The disparity between this (the 51.3%) and the other media reports on the study have me a little baffled. The BBC News report cited above says this:

Two-thirds made a "complete and lasting" response, with many of the other third showing substantial improvement.

I can't see the actual paper- I don't have a subscription (if anyone can get ahold of it and wouldn't mind emailing me, that would be fantastic! Hint hint.), but I don't know where those numbers come from.

The "control" was an 8-week waiting period before the study started, however, and not a comparison to non-specific CBT or other forms of psychotherapy. Nor did I read anything on drop-out rates, which tend to be astonishingly high. The last thing would be getting sufferers to present for treatment. Could the sufferers who volunteered for such a study be more motivated for change than those who didn't? In other words, are the people in the study representative of the ED population as a whole? Anyway, in many countries, even if you're motivated for treatment, finding a provider (even one who is marginally qualified) is easier said than done, not to mention wait periods, insurance, travel time, etc.

The other comment that left me baffled was a quote from Susan Ringwood, chief executive of B-EAT, the UK's eating disorder charity (equivalent to the US NEDA), said that "This research shows that people can benefit from psychological therapy even at a very low weight. There has been so little research into eating disorders and anorexia in particular, and this has really added to our knowledge in a challenging field."

Although most people are going to be underweight at a BMI of 17.5, this study didn't look at anorexia, just bulimia and EDNOS, so I'm not sure weight restoration was a part of this. Historically, food is the best therapy for those at very low weights, though this is typically augmented by other treatments to "help the medicine go down," as it were. A current study for CBT of anorexia is in the works, and other research has shown that CBT can help prevent relapse after weight restoration.

There are still many questions to be answered, but I think this study is a good start. I do, however, remain slightly doubtful until I can see more data.

UPDATED: Here's a fantastic article in The Independent that broaches many of the questions I raised in my post (and is a fantastic example of science writing that goes beyond the press release). Thanks, Tiptoe, for the link!