Panel Tackles the Clinical, Research & Policy Implications Surrounding Eating Disorders Not Otherwise Specified

Eating disorders not otherwise specified (EDNOS) represent a spectrum of challenges and unanswered questions for clinicians, patients, and insurance companies alike. One of the greatest obstacles is the lack of a clear definition for this group of disorders, according to a panel of experts at the annual meeting of the Academy for Eating Disorders.

Marsha D. Marcus, PhD, Associate Professor of Psychiatry and Psychology and Chief, Eating Disorders Program, Western Psychiatric Institute, Pittsburgh, who moderated the panel discussion, told the audience that there are many unknown aspects of the variants of disordered eating, both at the subthreshold and the symptom level. In addition, the symptoms of disordered eating as well as the clinical disorder may also be associated with morbidity. She added, “EDNOS are more likely to affect women of diverse backgrounds, women of color, and overweight women—women who often aren’t included in clinical trials.”

Lack of a clear definition hampers research

Part of the problem with EDNOS, according to Ruth Striegel-Moore, PhD, Professor of Psychology at Wesleyan University, Middletown, CT, is that since most clinical cases do not meet the full diagnostic criteria for anorexia nervosa or bulimia nervosa, patients have great difficulty getting insurance coverage for treatment. She added, “EDNOS are not just important because of the numbers of persons affected but also because they are is associated with significant impairment.” High rates of depression, anxiety, and obesity are often found among individuals with EDNOS, she said.

The medical literature is also of little help because there is a lack of clarity in the language, she said, adding that other terms such as “subthreshold disorders” or “partial syndrome” may be used to refer to individuals who lack the core criteria for anorexia nervosa or bulimia nervosa. Because researchers and the media all use many different terms, it is difficult to draw conclusions about treatment and other facets of EDNOS.

Dr. Striegel-Moore reported that in her study of treatment utilization using a large database—4 million individuals over 1 year—people with EDNOS essentially received the same amount of treatment as individuals with bulimia nervosa, and they had comparable rates of treatment for other psychiatric disorders.

“If you look at the utilization of treatment for eating disorders, EDNOS is as serious as bulimia nervosa,” said Dr. Striegel-Moore.” In this health services use study, EDNOS was the most common diagnosis for patients receiving treatment for an eating disorder, and gender discrepancy was less pronounced in EDNOS than in anorexia nervosa or bulimia nervosa. EDNOS are clinically significant syndromes, she commented, as suggested by high rates of treatment, and the relatively comparable treatment duration.

Dr. Striegel-Moore noted that theoretical and empirical work is needed to clarify the definition of EDNOS. Working against this is the fact that most treatment studies thus far have focused on either anorexia nervosa or bulimia nervosa, leaving unanswered the question of whether evidence-based treatment will work for people with EDNOS, or if a different type of intensity of treatment is needed.

Body dissatisfaction and body mass index across ethnic groups

One of the areas of great concern is an increase in the numbers of overweight persons, particularly an epidemic increase among Hispanic and African-American women, compared to whites, said Marian Fitzgibbon, PhD, Associate Professor of Psychiatry and Preventive Medicine at Northwestern University Medical School, Chicago. Dr. Fitzgibbon pointed out that while overweight has become a problem among 34% of white women, it now affects 52% of African-American women and also about 52% of Hispanic women.

While genetics, diet and exercise, access to food, and recreational activities all have an effect on these numbers, part of the explanation also lies in differences in body image. Dr. Fitzgibbon explained that African-American women have a more positive body image than white women, so they feel more attractive even when overweight. The scant data available about Hispanic women are somewhat more controversial because their body image ideals are affected by acculturation. She explained that Hispanic women who came to the U.S. before age 17 have a body image similar to that of white women; those who immigrated to the U.S. after age 18 seem to have heavier ideal weights.

Dr. Fitzgibbon reported the results of a study that sought to correlate the level of body mass index (BMI) and body dissatisfaction. Among 389 women, differences in age, education, and BMI did not correlate with body dissatisfaction. However, there were differences in the rate and the level of increase of body satisfaction as a function of ethnicity. White women experienced body dissatisfaction at a lower BMI than the other groups, while Hispanic women did not experience body dissatisfaction until they were overweight, and African-American women did not express body dissatisfaction until they were obese.

Dr. Fitzgibbon urged clinicians to be aware of differences in strategies of weight loss and weight gain in different ethnic groups of men and women.

Family studies of Binge-Eating Disorder (BED)

Although binge-eating disorders have not traditionally been viewed as heritable diseases, Dr. Lisa Lilenfeld, PhD, Assistant Professor of Psychology at Georgia State University, Atlanta, and her colleagues found that BED does seem to run in families. Dr. Lilenfeld and her colleagues recently studied 300 first-degree relatives of patients with BED.

Dr. Lilenfeld identified a potential relationship between alcohol abuse and BED. Substance abuse is more common in female relatives of BED patients than female relatives of non-BED patients. The results also suggest the possibility that substance abuse and BED may share a common etiology, specifically among women.

A common shared familial vulnerability factor may explain the elevation of substance use disorders in the female relatives of women with BED. There may be some common vulnerability factor that is manifested in some women as BED and in some others as substance use problems, she said.

Dr. Lilenfeld noted that family study data can be used to search out potential etiologic factors that are important in the development of BED. For example, she and her colleagues found elevated rates of anxiety disorders in relatives of the BED women. One explanation for this may be an etiologic link between BED and anxiety disorders. Anxiety disorders are also common in the families of BED women.

Health implications of EDNOS

The serious health implications of EDNOS are clearly reflected in conditions like functional hypothalamic amenorrhea (FHA), according to Dr. Marsha Marcus. FHA is cessation of menses in women who previously have had normal menstrual function and where there is no identifiable cause for the amenorrhea, she explained.

Dr. Marcus, Sarah L. Berga, MD, and Tammy Loucks postulated that stress alone or undernutrition alone do not cause FHA; instead, it is due to a combination of energy deprivation from mild calorie restriction, poor nutrition or exercise and stress in the form of unrealistic goals or performance pressure. The combination causes hypothalamic alterations and disrupts gonadotropin-releasing hormone (GnRH ) pulsatility.

In Dr. Marcus’s study, women with FHA (n=16) were compared with women with organic amenorrhea/anovulation (n=19) and eumenorrheic/cyclic women (n=15) on levels of eating disorder symptoms. Women with current or past eating disorders were screened out of the study.

FHA women scored higher on the Eating Disorders Inventory than either the cyclic women or the amenorrheic women in symptoms of bulimia, drive for thinness, and interoceptive awareness (a measure of sensitivity to internal cues associated with emotion, hunger, and satiety). Dr. Marcus noted that this was intriguing to the researchers because this was a group of women who were specifically screened to exclude frankly disordered eating.

On the bulimia test (BULIT-R), FHA women reported significantly more symptoms of bulimia nervosa than the two other groups. One-fourth of the women with FHA had scores above 85 on the BULIT-R test, while neither of the other two groups had scores in this range.

The symptoms of eating disorders but not those of depression discriminated FHA women from those with organic amenorrhea and control subjects. “Thus,” Dr. Marcus pointed out, “for some women, FHA may reflect the adaptation to mild nutritional compromise or the attitudes and behaviors associated with disordered eating.”

Dr. Marcus told the audience that psychological intervention designed to ameliorate subthreshold symptoms may promote restoration of ovulation in women with FHA. Her group is currently conducting a randomized control trial of cognitive behavior therapy for subthreshold eating disorders symptoms in the treatment of FHA. This may provide an alternative to expensive and risky infertility treatments. It may also help ameliorate serious side effects of amenorrhea, such as osteoporosis and heart disease, she said.

Finally, Dr. Marcus said, “We need to help our physician colleagues become more sophisticated in how they talk to young women about their behaviors and their body eating, weight, and shape concerns.”