To Drs. Christopher Fairburn and Zafra Cooper, of Warneford Hospital at Oxford University, the classification system proposed in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) does not reflect the clinical reality of treating patients with eating disorders (Br J Psychiatry 2011; 198:8). According to the British researchers, the main problem is that in adults the current classification system recognizes only two conditions, anorexia nervosa (AN) and bulimia nervosa (BN). As a result, according to Fairburn and Cooper, at least half the cases of eating disorders seen in clinical practice become relegated to the residual diagnosis of eating disorder not otherwise specified, or EDNOS. And, according to the authors, changes proposed in the new DSM-V will only partly correct this.

Apart from patients with binge eating disorder, almost all individuals with EDNOS have clinical features seen in AN and BN, although in differing degrees and different combinations. Most have the characteristic overvaluation of shape and weight and its expressionsuch as repeated body-checking or body avoidance, extreme feelings of fatness, sustained attempts to follow demanding dietary reels, and unhealthy weight control behaviors.

Some possible solutions

Dr. Fairburn notes that the least extreme solution would be to adjust the current DSM-IV classification system, as opposed to making any radical changes. This would relax the diagnostic criteria for AN and BN. As long as the core diagnostic concepts were not materially altered, the criteria could be relaxed. Thus, for example, the criterion for the presence of amenorrhea could be eliminated, and the weight threshold raised. As for BN, the main proposal has been to lower threshold of frequency of binge eating and purging. The British researchers recommendations are in accord with the proposal being made by the DSM-V working group to recognize BED as a distinct disorder, separate from AN and BN.

A few more strategies

Other strategies might be to accept the DSM-V proposals as reasonable, and to reassign the remaining EDNOS cases. Remaining cases could be reclassified as an additional eating disorder, perhaps termed “mixed eating disorder.”

Drs. Fairburn and Cooper note that one of the challenges to classifying eating disorders is that they are not stable, and symptoms change with age and duration of the disorder. Also, they write that eating disorder diagnoses are like “snapshots in the course of an eating disorder.” While some individuals maintain strict control over their eating and thus remain underweight, they are but a small minority of all patients seen. Most patients have a course that changes over time, from AN to BN, and then to a mixed state, or to an EDNOS.

Finally, the authors note that, with DSM-6 in mind, comprehensive transdiagnostic samples need to be studied, with data on the patients’ current state, course, and response to treatment.