Anorexia Nervosa: Curious Past, Hopeful Future

by Arnold E. Andersen, MD, University of Iowa School of Medicine, Iowa City, Iowa


Anorexia nervosa (AN) has a history as interesting as the Cheshire cat's smile: now you see it, now you don't. Morton's first generally accepted account of AN appeared in the English language in 1689. Over the next 140 years AN was sequentially compressed into a medical category ("postpartum pituitary necrosis"), wedged into psychoanalytic theory, made an exclusively female disorder, relegated to a forme fruste of several other disorders (schizophrenia, depression), and fought over regarding etiology, pathogenesis, and treatment. Recently it has spread like a virus to developing countries. Males have been reunited into the AN camp in the last 40 years.

AN is a prototype of disorders of motivated behavior, all of which serve as a final common pathway for a variety of developmental, familial, intrapsychic, and societal conflicts. At its core, AN arises from a conflict between the individual's neurobiological forces, which regulate weight stability, and social norms, which mandate thinness. The probabilistic nature of a single person developing AN from multiple risk facts may mean no single causative factor will ever be found.


A Syndrome Waiting to Happen

AN can be appreciated as a syndrome that blossoms when predisposing features are acted upon by specific precipitating factors, which then sustained by a combination of biomedical and psychosocial factors. Occasionally AN begins inadvertently, without a flame, but always with kindling ready to light. Recently, the list of predisposing factors has been shown to include genetic vulnerabilities, primarily of the serotoninergic system. The best estimates of the heritable vs. acquired elements hover around 50% for each, with wide differences suggested by different studies.

In contrast to other disorders that are present or absent, subsyndromal eating disorders and negative attitudes toward weight and shape place a qualitative burden on the everyday lives of a majority of adolescent and adult women who do not meet the full criteria for an eating disorder. The most widely held norm among many young women in our society is not honesty, virtue, or materialism, but the desire for thinness (already present in 50% of 5th graders), a belief that is not merely abstract, but one with serious behavioral and emotional consequences. Males are equally dissatisfied with weight and shape, but divided between a desire for increased or decreased weight. Epidemiologic studies suggest that 0.5% to 1.0% of young women develop full AN, with most studies indicating increased prevalence over the past decades. Approximately 5% of young women have a mild version of AN. One problem is that the diagnostic criteria keep changing, largely because of the need to impose categories on a continuum but also because the criteria are the result of a compromise between scientific and political concerns. The varying requirement for the weight to qualify for AN constantly alters the epidemiology.

What's New in Diagnosis?

Three new areas in diagnosis include: stable diagnostic criteria (albeit practical and usefully arbitrary); the demonstration that 3 months of amenorrhea is not necessary for a diagnosis of AN; and the description of reverse anorexia (muscle dysmorphia) in males. Still unsettled is the question of whether binge eating disorder is a separate disorder, and where it belongs. I believe it should be categorized as a third component of bulimic disorders (bulimia nervosa with purging, bulimia nervosa with other compensations, and bulimia nervosa with no compensations, also called binge eating disorders).

Gender bias and problems with definition. Unfortunately, there is continued gender bias. There is no endocrine or hormonal abnormality for males, so the diagnosis for men should either match the female criterion or the female criterion should be eliminated. The eating disorders not otherwise specified (EDNOS) category is much too broad—most are subsyndromic varieties of AN or BN, but don't fit into narrow research criteria. Finally, there is a failure to broaden the definition of AN to a more meaningful definition. Here is a suggestion: AN should be defined as any decrease from a healthy set point driven by desire for thinness that leads to functional biological and psychological impairment; i.e., a person who is normally 20 % above population mean, and then loses weight to 10% below, has no periods, is cold, decreased cognition, etc., has true AN. We are working on this, and hope to publish a proposed revision of EDNOS soon. Several groups at risk of AN are also often overlooked, including minorities, males, and older women.

What's New in Treatment?

Improvements in treatment have included elimination of most single-modality treatments (nutrition alone, psychotherapy alone) in favor of validated multidisciplinary treatment. There is an increasing consensus about the validity and utility of CBT as the core of most AN psychotherapies. Another improvement is a move away from extreme medicalization of AN treatment. It is the impression of some of those in the field that the death rate has decreased with good initial care plus good follow-up over several years. We have learned that there is a lack of benefit from treatment with antidepressants during the acute treatment phase, and a possible benefit during relapse prevention.

Better care available. Another positive treatment trend includes growth of the spectrum of care programs: inpatient, partial hospital, outpatient, and outreach. Other positives include increasing recognition of gender-specific needs of males and females, and that anorexia nervosa coexists with specific comorbid disorders, such as obsessive-compulsive disorders, borderline personality disorder, and with substance abuse

Pessimism despite progress. An unduly pessimistic attitude about treatment lingers in some quarters despite the robust evidence that cure (defined as cultural normality) is regularly achieved for most patients who remain in experienced treatment for 3 to 5 years. There is increasing agreement on the need for multidisciplinary treatment, including scientific nutritional rehabilitation, psychological change via cognitive-behavioral and interpersonal therapy, behavioral relearning, and occasionally, individualized psychopharmacology, more for the comorbidity symptoms than for the core eating disordered thinking. As with cancer treatment, immediate improvement must be matched by sustained improvement on global measures at a five-year mark. Acute treatment, followed by a relapse prevention program run by experts, is more economical in the long run as well as exceedingly more effective than revolving-door, short-term stabilization.

Uncertainties and changes. Some areas of treatment are still uncertain. For example, what are the conditions for transition from inpatient to day (partial) programs? (Am J Psychiatry 1999; 156:1697). There is also debate about normal food eaten normally vs. liquid supplements, and the validity of setting target weight ranges. Other uncertain areas include the degree of restoration of ventricular dilatation and decreased cortical mass, gray vs. white.