Jared DeFife Ph.D.

The Shrink Tank

Eating Disorders On The Rise, Part II: Weighing The Evidence For Psychotherapy

Psychotherapy for eating disorders: weighing the evidence

Posted Jun 06, 2009

As a social culture, we can't seem to find a healthy balance when it comes to weight. On one hand, we're an increasingly obese nation with all of the added weight of health complications. On the other hand, we're either heralding or demonizing celebrities for drastic shifts in body weight.

We're as perplexed as the characters in Kafka's short story A Hunger Artist:

"I always wanted you to admire my fasting," said the hunger artist.
"But we do admire it," said the supervisor obligingly.

"But you shouldn't admire it," said the hunger artist.
"Well then, we don't admire it," said the supervisor...

Given our perpetual fascination with and raging envy of thinness, the prevalence of eating disorders should come as no surprise. Communities have emerged in celebration of Pro-Ana or Pro-Mia lifestyles, promoting "thinsporation" and advice. But eating disorders are not fashionable and glamorous. They are not lifestyle choices or spiritual experiences. They are devastating illnesses with severe consequences.

In a previous post, we examined methods that researchers use to evaluate whether or not psychotherapy "works". By nature, psychotherapy is not highly conducive to rigorously controlled experimental methods. Experimental studies of psychotherapy, known as RCTs, tend to be conducted with very brief, tightly manualized interventions with patients who demonstrate a very discrete symptomatic pattern. Then again, is a tightly controlled experiment really the most useful method for studying the complex, pervasive, and persistent phenomena seen in psychopathology and treated in psychotherapy? More observational studies of psychotherapy as practiced in real-world treatment settings over time have greater difficulties identifying the specific causal factors of change.

First, one obstacle for investigating treatment of eating disorders, as one of my perceptive readers commented on to part I, is diagnosis. The exceptionally paltry Eating Disorders section of the DSM-IV-TR lists three diagnostic categories: Anorexia Nervosa (which has a hallmark of a body weight less than 85% of expected for an individual's height/age); Bulimia Nervosa (identified by a cycle of binging and purging behavior); and Eating Disorder Not Otherwise Specified/NOS (in which disordered eating patterns cause significant physical/psychological distress, but with symptoms that do not match any existing diagnostic category).

Not surprisingly, an excessive number of cases fall into the NOS diagnostic category. The recent report from the Agency for Healthcare Research and Quality identified that hospitalizations for atypical forms of eating disorder are most sharply on the rise. Without a focused definition of a disorder, a controlled experimental study of treatment is more difficult to conduct.

Let's look at some evidence from psychotherapy Randomized Control Trials, which are allegedly the "gold standard"* for identifying "Evidenced Based Treatment". What do they tell us about psychotherapy for eating disorders?

Unfortunately, there appears to be little to no repeated RCT evidence of a consistently and lasting beneficial psychotherapy for the treatment of Anorexia Nervosa. The American Psychiatric Association identifies a few sparse trials with very small samples. One study found 20 sessions of nonspecific clinical management to be as effective as either Cognitive-Behavioral Therapy (CBT) or Interpersonal Therapy (IPT). Another study found one-year of CBT to be more effective than one-year of nutritional counseling. In a third trial, patients in psychoanalytic psychotherapy, family therapy, and cognitive-analytic-therapy (an adaptation of psychodynamic psychotherapy) were found to achieve greater improvement over a group with low-contact treatment-as-usual. Yet, in each of these trials, improvements were minimal and gains were only observed in a small percentage of the patient sample. In terms of what patients found helpful about treatment: "support, understanding, and empathic relationships were rated as critically important, psychological approaches were rated as the most helpful, and medical interventions focused exclusively on weight were viewed as not helpful".

It seems that Anorexia may well be one of, if not the single, most difficult forms of mental illness to treat. It's a particularly entrenched disorder in its own right, and when one adds the serious medical/physical complications involved, the problem can quickly become very life-threatening. As an aggregate whole the field has not yet demonstrated robust efficacy for psychotherapy of anorexia, but this does NOT mean that certain individuals with anorexia can't be helped, and perhaps provided life-saving help, from psychotherapy.

Evidence from RCTs for the treatment of bulimia is better, but still very limited. Meta-analyses (research which compiles results from multiple studies) demonstrate that the brief interventions provided in RCT studies do result in statistically significant improvements of eating disorder symptoms and improved levels of functioning.

The bad news is this: "two thirds of BN [Bulimia Nervosa] patients who receive individual psychotherapy with CBT-the most efficacious treatment studied to date-either drop out or fail to recover by termination, and patients who do not recover tend to retain symptom levels surpassing the DSM-IV criteria for the disorder" (Thompson-Brenner and Westen, 2005, p. 573).

I'm reminded of a saying I've heard many many times having grown up with the Cleveland Browns as my local football team: "sure, the team sucks now, but at least we're better than last season". That mentality just isn't good enough. Does that mean that there's little hope for psychotherapy to help? Not necessarily so. In the next post, we'll look at what happens when we take our investigation of psychotherapy outside the research lab and into the real-world.


Weigh-in with your thoughts, concerns, or questions by commenting below.

*The term "gold standard" actually applies to an economic theory that no country currently practices and fell apart decades ago in real-world implementation. Furthermore, the longer a country adhered to a gold standard, generally the greater was its economic severity and time to recover from the Great Depression.  As applied to the canonization of RCT methodology for guiding clinical practice, this may be an apt metaphor.

Thompson-Brenner, H., & Westen, D. (2005). A naturalistic study of psychotherapy for bulimia nervosa, Part 1: Comorbidity and therapeutic outcome. Journal of Nervous and Mental Disease. 193(9):573-84.

Email correspondence to jareddefifept@gmail.com

More Posts