Working Toward a Better Definition for EDNOS

The current version of the Diagnostic and Statistical Manual for Mental Disorders IV Text Revision (DSM-IV-TR) includes definitions of three eating disorders—anorexia nervosa (AN), bulimia nervosa (BN), and eating disorder not otherwise specified (EDNOS; see table). As the fifth edition of the DSM is being written and refined by eating disorders experts,  particular attention is being devoted to the category of EDNOS, according to Kamryn T. Eddy, PhD, and colleagues at Massachusetts General Hospital, Harvard University, and the University of Chicago.

More than 50% of adults seeking treatment for eating disorders are diagnosed with EDNOS, according to Dr. Eddy. Despite this, according to the authors, the diagnosis needs to be clarified for adults and adolescents. (J Am Acad Child Adolesc Psychiatry. 2008; 47:156). Dr. Eddy and colleagues noted few studies have been done to address EDNOS among adolescents.

A study to clarify the diagnosis among teens

To better define EDNOS among adolescents, Dr. Eddy and her co-workers designed a study among 281 consecutive adolescents referred for evaluation and treatment for eating disorders at the University of Chicago Hospitals between October 2001 and February 2007. A total of 259 females (92.2%) and 22 males (7.8%), from 12 years to 19 years of age, participated in the 5-month study. 

When the DSM-IV-TR diagnoses were evaluated, the researchers found that 59.1% of the teens had EDNOS, about 20% had AN, and 20% had BN. About half of the teens with EDNOS reported one or more binge-eating episodes and one or more purging episodes per month. About  half also reported engaging in excessive compensatory exercise, and more than a fourth of the students reported engaging in 24-hour fasts. Laxative and diuretic abuse was far less common among the youths (6.0 % and 2.45%, respectively).

When compared to teens with AN or BN, those with EDNOS were younger than those with BN and endorsed eating disorders pathology and general psychopathology that fell somewhere between those reported for either AN or BN.

At first, the authors classified the teens into 4 subtypes: subthreshold anorexia nervosa (SAN), subthreshold bulimia nervosa (SBN), EDNOS purging, and EDNOS, binge eating. Even with these categories, the symptoms of 20% of the teens did not resemble those reported for the other types, so a fifth category (“other”) was created.

When Dr. Eddy and her colleagues compared the teens in the 5 EDNOS categories, they identified between-group differences in demographics, eating disorder pathology, and depressive symptoms. These findings suggested there may be meaningful differences among youths who are diagnosed with EDNOS based on their eating disorders symptoms.

Two-thirds of the youths diagnosed with EDNOS were below 85% ideal body weight (IBW), but missed the amenorrhea criterion for AN. There were no differences in demographics, ED pathology (except for body weight), depressive symptoms or self-esteem between youths with AN and those with SAN. The lack of differences added support for the suggestion that the strict diagnostic criteria for AN be relaxed. Some investigators have suggested that the amenorrhea criterion for a diagnosis of AN be dropped entirely because menstruation is not a meaningful indicator—some individuals with low weight continue to menstruate while some who are at normal weight become amenorrheic.

One-fifth of those in the EDNOS group narrowly missed meeting the full criteria for BN, and thus were categorized as having SBN. Most of the teens missed meeting the full criteria for BN based on objective binge eating occurring less than the twice-weekly frequency threshold for BN; a few adolescents missed meeting the criteria based on infrequent purging or binge eating and purging.  Compared with those with BN, adolescents with SBN were younger, had fewer objective binge and purge episodes, lower Eating Disorder Examination scores, and higher levels of self-esteem.

DSM-V modifications needed

Just as in adults with eating disorders, the majority of these teens met the diagnostic criteria for EDNOS. Lack of differences between the subtypes AN and SAN suggested that the strict criteria for AN could be relaxed, and differences between BN and bulimic variants did not support combining these patients into one group. The authors’ findings underscore the importance of increasing research into EDNOS, and support modifying the current criteria for AN and BN in the DSM-V.