The Eating Disorders Work Group for DSM-5 was faced with two major issues. DSM-IV recognized only two specific eating disorders, anorexia nervosa and bulimia nervosa. However, a large minority, or in some studies a majority, of individuals with clinically significant eating problems who were presenting for care did not meet criteria for either of these disorders and were therefore assigned to the residual category “eating disorder not otherwise specified.” Second, the decision to eliminate the DSM-IV chapter describing disorders usually first diagnosed in infancy, childhood, or adolescence led to our Work Group's being responsible for pica, rumination disorder, and feeding disorder of infancy or early childhood. Grappling with these issues led to virtually all of our recommended changes, which have been incorporated into DSM-5 in the chapter “Feeding and Eating Disorders.”
Undoubtedly, the major change is the official recognition of binge-eating disorder. Although its inclusion was considered for DSM-IV, in the end it was decided that insufficient data were available about the characteristics, course, and treatment of individuals with this syndrome, and criteria were provided for further study in an appendix. Extensive research followed the promulgation of these preliminary criteria for binge-eating disorder, leading to the publication of more than 1,000 articles over the ensuing years. These data documented the clinical utility and validity of the proposed syndrome (
1). For example, individuals with this disorder experience substantially more anxiety and mood disturbance than do comparison subjects, and they respond well to treatments that incorporate methods to address these symptoms. In addition, data are emerging that the presence of binge-eating disorder in adolescents portends a greater risk of weight gain. The only significant difference in the DSM-5 criteria from those provided in the DSM-IV appendix is that the minimum average frequency of binge eating required is once weekly over the last 3 months, identical to the frequency criterion for bulimia nervosa in DSM-5.
The second major change involves the DSM-IV category “feeding disorder of infancy or early childhood.” This disorder has been renamed “avoidant/restrictive food intake disorder” and the criteria have been expanded. There are two major reasons for these changes. First, the DSM-IV category was rarely used, and limited information is available on the characteristics, course, and outcome of children with the DSM-IV disorder. Second, a number of individuals, primarily but not exclusively children and adolescents, substantially restrict their food intake and experience significant associated physiological or psychosocial problems, but they do not meet criteria for any DSM-IV feeding or eating disorder (
2). For example, after an unpleasant episode of gastrointestinal illness accompanied by vomiting, some youngsters may avoid eating solid food and develop significant nutritional problems. Similarly, although idiosyncratic preferences for foods of a certain taste or consistency are very common during childhood development and are typically self-limited and require no intervention, in rare cases the preferences are so extreme and persistent as to lead to clinically significant problems. Avoidant/restrictive food intake disorder is a broad category intended to capture this latter range of presentations.
The core diagnostic criteria for anorexia nervosa are conceptually unchanged from DSM-IV with one exception: the requirement for amenorrhea has been eliminated in DSM-5. In DSM-IV, this requirement was waived in a number of situations (for example, for females taking contraceptives and for males). In addition, the clinical characteristics and course of females meeting all DSM-IV criteria for anorexia nervosa except the requirement for amenorrhea closely resemble those of females meeting all DSM-IV criteria (
3). As in DSM-IV, individuals with this disorder are required by criterion A to have a significantly low body weight for their developmental stage. The wording of this criterion has been changed from DSM-IV for clarification, and guidance regarding how to judge whether an individual’s weight is significantly low is provided in the text. In DSM-5, criterion B has been expanded to include not only expressed fear of weight gain but also persistent behavior that interferes with weight gain, as a number of individuals do not report this fear (
4,
5).
The only change in the criteria for bulimia nervosa is a reduction in the required minimum average frequency of both binge eating and inappropriate compensatory behavior from twice to once weekly. The clinical characteristics and outcome of individuals meeting this slightly lower threshold are similar to those meeting the DSM-IV criteria (
6). In addition, the scheme distinguishing purging and nonpurging subtypes was eliminated, since data did not support the clinical utility of this distinction (
7). Other small changes were made to the criteria for pica and rumination disorder for clarification and to permit these diagnoses to be assigned to individuals of any age. The section “other specified feeding or eating disorder” lists several presentations, such as purging disorder and night eating syndrome, that have been described in the literature; insufficient information is currently available about these conditions to merit full recognition as a disorder (
8,
9). Studies carried out by a number of investigators during the development of DSM-5 indicate that the changes recommended will significantly reduce the need for the use of this residual category relative to the use of the comparable “not otherwise specified” category in DSM-IV.
Finally, as was the case in DSM-IV, obesity is not included in DSM-5 as a mental disorder. Obesity results from the long-term excess of energy (calorie) intake relative to energy expenditure. Genetic, physiological, behavioral, and environmental factors that vary across individuals contribute to the development of obesity; thus, obesity per se is not considered a mental disorder. However, there are robust associations between obesity and a number of psychiatric disorders (for example, binge-eating disorder, mood disorders, and schizophrenia) as an associated feature of the disorder or its treatment. Therefore, while obesity is not a psychiatric disorder, it is a highly prevalent condition important for psychiatrists and other mental health professionals to consider in managing their patients (
10).