A wolf in sheep’s clothing

Every researcher in the field of eating disorders tries their best to reduce the burden of suffering for patients.  They try to contribute to the meaningful scientific discussion.  Having said that, however, I am going to proceed to critique an article by Natalie Godart, Sylvie Berthoz, Florence Curt and colleagues at the Institut Mutualiste Montsouris in Paris, France; the French National Institute for Health and Medical Research; King’s College, London Institute of Psychiatry; the Department of Biostatistics, Necker Hospital, Paris; and the Institute of Mental Health Research, University of Ottawa.  I must thank Chris Berka, former board chair of FEAST (US), for making me aware of this article, although it resulted in a huge headache given how much was wrong with it.

I call this article a “wolf in sheep’s clothing” because, while it purports to report more favorable outcome for “treatment as usual (TAU)” when “family therapy (TAU+FT)” is added, what it really does is imply that faulty family dynamics are what sustain or cause anorexia nervosa, and –worse—it states that “our study design made it possible to rule out the hypothesis that the key ingredient for family therapy effectiveness in AN is that it places ‘greater emphasis on getting patients to eat well and maintain a healthy weight’”.

Leaving aside that “treatment as usual (TAU)” as they defined it (“sessions for the patient alone as well as sessions with a psychiatrist for the patient and her parents”) is so inadequate, if this is “treatment as usual” in France, they had better strongly consider joining the 21st century. So what could possibly be improved by adding “family therapy sessions targeting interfamilial dynamics, but not eating disorder symptoms” ?  The answer seems to be “not much”, as the reported outcomes were so bad that I would quit my job if that were all that we were able to achieve for children at Kartini Clinic.

Good and intermediate outcome groups (Morgan and Russell Outcome Categories) were lumped together and defined as those patients who achieved a “healthy weight” and sometimes resumed menstruation.  Now let’s look at what they considered a healthy weight.  They state (emphasis mine):

Regarding weight status assessment, in view of the patients’ age, we considered the Ideal Body Weight (IBW) (which is classically defined as the average body weight of the general population over 15 years of age) to be a less relevant index than BMI percentiles. Hence, to take the ages of our patients into account, we referred to the INSERM (French National Institute for Health and Medical Research) weight curves for the French female population [47], in which a BMI<10th percentile indexes AN [48]. We defined the outcome categories as follows [16], [49]:

1) Good outcome : weight >10th BMI percentile and regular menstruation

2) Intermediate outcome: >10th BMI percentile but amenorrhea (i.e. the absence of menstruation for at least the past three months)

3) Poor outcome: weight <10th BMI percentile and/or presence of bulimic symptoms.

As they melded the “good” and “intermediate” outcome groups into one acceptable group for analysis and contrasted that melded group to the “poor outcome” group, to belong to the successful group a patient could be amenorrheic (no menses) and weigh any number above—even slightly above—the 10th percentile for age. AND, if I am reading their tables correctly, by eighteen months (!!) only slightly more than 17% had achieved this modest outcome in the TAU group and about 40% in the TAU + family therapy (TAU+FT) group.  

Now, that’s some damned poor outcome after a very long time.  A closer look  (see table 3) shows that only about 27% of the TAU group achieved a BMI greater than the 10%tile for age and 53% in the TAU+FT. Worse, about 65% in the TAU group were still amenorrheic and a third were still amenorrheic in the TAU+FT group.  I would not publish this data if I were them, unless I wanted to blow the whistle on a national disgrace.

From my Kartinian point of view I struggled with two main issues of the authors’ approach to pediatric eating disorders:

  1. Their definition of a healthy weight is so flawed as to be malpractice.  How can one define a healthy weight for a given child/teen as “that weight which is above the 10th percentile for their age?”  Have they had no clinical experience with eating disordered children?  How about normal children? Surely it matters what the starting point was?  A child who grew happily along the 50th percentile prior to their illness could not be said to have achieved a “healthy weight” if they have been returned to a weight which represents the 15th pecentile for age.  Who cares about the average BMI centile for age? Individual healthy weights will fluctuate wildly with Tanner stage (SMR), genetics and menstrual history.  This lack of understanding of the crucial role of weight restoration and individual healthy weights is no doubt why they got such abysmal results.
  1. Although I am aware that even people who agree with me up to this point may now depart from my analysis, this article is a perfect example of why I cannot agree that all that is “evidence-based” is gold.  There is a lot of so-called “evidence” out there that is misleading and even demonstrably false.  And an even larger pile of “evidence” that has been erroneously interpreted.  This article is a perfect example of the latter: drawing the wrong conclusions from the data, probably a result of bias in concept to begin with.  This research represents a randomized controlled study (RTC), considered to be the gold standard for evidence.  I say:  caveat emptor!  Let the buyer beware!