Impact of the New Criteria for Diagnosing Eating Disorders
Past and Current Definitions
For roughly two decades prior to 2013, the basic guidelines for diagnosing eating disorders in the U.S. came from the fourth edition of an American Psychiatric Association publication called the Diagnostic and Statistical Manual of Mental Disorders. This edition of the manual (known widely as DSM IV) contained only two fully defined forms of disordered eating: anorexia nervosa and bulimia nervosa. People with significant eating-related problems who did not qualify for a diagnosis of either of these conditions were grouped together under a diagnostic heading called eating disorder not otherwise specified (EDNOS), even though any two individuals with this diagnosis might have widely varying symptoms and eating-related issues. Many people diagnosed with EDNOS had symptoms of a then-unofficial disorder called binge-eating disorder.
When it published the fifth edition of the Diagnostic and Statistical Manual (DSM 5) in May 2013, the APA grouped the eating disorders category with another category of conditions formerly known as feeding and eating disorders of infancy or early childhood. This reorganization, which was made from a desire to improve the diagnosis and treatment of eating-related problems, resulted in a combined category of conditions known simply as feeding and eating disorders. Critically, this category now includes binge-eating disorder as a separately diagnosable disorder. It also now includes a condition called avoidant/restrictive food intake disorder or ARFID. Together with anorexia and bulimia, binge-eating disorder and ARFID now form the four conditions in the feeding and eating disorders category viewed as the primary eating-related disorders. Other conditions in the category include rumination disorder, pica, “unspecified” feeding or eating disorder and “other” specified feeding or eating disorder.
Impact on Eating Disorder Diagnosis
In the study published in the Journal of Adolescent Health, researchers from eight universities and the Cleveland Clinic made two separate evaluations of the symptoms of 215 teenagers seeking treatment for disordered eating for the first time. During the first evaluation, the researchers used the criteria for diagnosis contained in the now-outdated fourth edition of the Diagnostic and Statistical Manual. During the second evaluation, they used the current criteria for diagnosis contained in DSM 5. Almost 89 percent of the study participants were girls; this gender distribution reflects the real-world rates for eating disorders among males and females.
When using the criteria from DSM IV, the researchers diagnosed 30 percent of the study participants with anorexia. They also diagnosed 7.3 percent of the participants with bulimia, while fully 62.3 percent of the participants received a diagnosis of eating disorder not otherwise specified. Anorexia was diagnosed in 40 percent of the study participants when the researchers examined them under the DSM 5 criteria, while bulimia was diagnosed in 11.8 percent of the participants. Only 32.6 percent of the participants qualified for an EDNOS (technically, an “unspecified” or “otherwise” specified) diagnosis under the criteria contained in DSM 5. Roughly 14 percent of the study participants met the criteria established for diagnosing cases of avoidant/restrictive food intake disorder.
The authors of the study in the Journal of Adolescent Health undertook their efforts because no other research team had fully addressed the potential impact of the new eating disorder guidelines on teenagers. They note that the rise in the number of anorexia and bulimia diagnoses under the DSM 5 guidelines almost certainly stems from small adjustments in the definitions of those conditions that statistically increase the number of affected people. They also note that the drop in EDNOS diagnoses stems from both the establishment of binge-eating disorder as a separately diagnosable condition and a related chance for doctors to avoid the known difficulties of accurately diagnosing EDNOS in teen populations. Finally, the authors note that the study underscores the practical usefulness of adding avoidant/restrictive food intake disorder as a diagnosable eating-related condition.