New Directions and Challenges Ahead for Eating Disorders

By Mary K. Stein, Managing Editor
Reprinted from Eating Disorders Review
July/August 2007 Volume 18, Number 4
©2007 Gürze Books

At the International Eating Disorders Conference (ICED) in May, more than 1,000 clinicians and eating disorders experts from around the world examined the complexities, progress, and future of diagnosing and treating eating disorders.

Keynote Address: Changing the Status Quo

In a keynote address, Dr. Thomas Insel, Director of the National Institutes of Mental Health (NIMH), Bethesda, MD, told the ICED audience, “In eating disorders we need to move to an era where we diagnose with biomarkers and develop new treatments focused on the core biology.” Dr. Insel said that the current lack of this ability is not unique to the eating disorders field and that most psychiatric illnesses are diagnosed by the behaviors observed or patient symptoms, and are treated episodically even though the disorders are chronic. The 21st Century will be the century of chronic disease, but our healthcare system remains focused on acute care, he said.

The ultimate goal, he said, is personalized care, and the endgame is strategic prevention. To deliver personalized care, he added, those in medicine should be thinking more about evolutionary projects, new science, and new research to see that it actually informs practice. “Now is the time to do it for eating disorders,” he said.

A Spectrum of Brain Disorders

Dr. Insel said it is fundamental to recognize that the whole spectrum of eating disorders are brain disorders, and that it is now possible to study mental processes as brain processes. There are many ways to study the brain, he said, and this is happening all across biomedical research. As examples, he cited the Human Genome Project and the International Hapmap Project (, which is providing a way to study points of variations in genetic sequences. According to Dr. Insel, the crucial points of variation can be counted quickly, cheaply and well.

Larger Studies Are Needed

Research into the pathophysiology of eating disorders has suffered from the fact that research databases include small numbers of patients, Dr. Insel said. To establish well-defined phenotypes, DNA samples are needed from between 2,000 and 10,000 patients and an equal number of controls, he said, adding that in the NIMH database there are just 400 samples from anorexic patients. It is critical to get more information, and sharing databases is the way to do this, he said. An example is the Price Foundation’s efforts to build a large database, a place where data and information can be mined.

“We are not looking for genes for eating disorders,” he pointed out, “because genes don’t code for disease, but for proteins. We are looking for associations that increase the risk and susceptibility to eating disorders, and we want to use this as a gateway to the pathophysiology of these disorders.” Hopefully this will involve candidates that should be studied beyond serotonin and dopamine, and he stressed striving to get to the diseases as systems processes in the brain. “Genes are only part of the story,” he said. The genetic sequence can also be affected by early influences (“epigenics”), where very early experiences “can leave a footprint in the genome,” Dr. Insel pointed out.

Needed: A Change of Research Culture

Dr. Insel said that a real revolution is occurring in the way research is done. He added that in one sense it is becoming clear that the most exciting discoveries are coming from overlaps in the intersections of the basic sciences. As the research becomes more complex, he said there will be a need for “team science,” because the work will be too complex and vast for a single laboratory or research group. This also means getting away from science that is all about ownership, or moving from a proprietary-driven culture to one that is more about sharing. “Most families of patients with eating disorders don’t care who did the work, they just want answers,” said Dr. Insel.

The NIMH Director also turned to some of the negative factors that counterbalance the exciting possibilities in research. “We do have treatments that work for anorexia nervosa and for almost all the eating disorders and mental disorders but we don’t do them and most patients don’t have access to treatment. Health disparity is a problem because those who need the greatest care and intervention are probably the least likely to get it,” he added.

And, he noted, coordination of care is the area “where we have done the least careful thinking. He added, “Treatment of eating disorders is a great example of how badly we have done by parents by separating individuals from families and blaming families for the problem.” In fact, he added, “We need to ask for a day of atonement for past care, and we need to bring families into the picture and to make patient care much more individualized.” And, by defining mental illnesses as medical illnesses, we will have a better understanding and better access to treatment for patients, according to Dr. Insel.

Plenary Session Addresses Food Reward and Addiction

Strong parallels can be found between eating disorders and addiction, according to Dr. Caroline Davis, Professor of Psychology at York University, Toronto. Dr. Davis stressed that substance abuse, AN and BN are behavioral disorders, not lifestyle choices, and that research is clarifying the genetic vulnerability and physiologic consequences of disordered eating that sustain eating disorders. Eating disorders have characteristics resembling addictive disorders, including increasing salience of food, frequent relapses, functional impairment, and ambivalence toward treatment, she said.

Dr. Davis noted that addiction was once a term applied only to a dependence on and abuse of substances injected, ingested, or inhaled, but more recently it has been expanded to include behavioral addictions, and some argue that food can also be addictive. Some types of food, particularly fat and sugar, have become more like conventional drugs in the way they affect the brain, she said. She also noted that more than 50% of the typical American diet includes added fat and sugar. She pointed out that over the last decade, the cost of refined sugar and vegetable oils has fallen, so they are used liberally by the food industry. In addition, more than 40% of the meals we eat are not prepared at home, and these are also processed and include lots of fat and sugar.

BED: A Good Model to Study

One phenotype well positioned for studying eating as an addictive disorder is binge eating disorder, or BED. BED involves eating not driven by hunger, and BED patients even refer to their binges in hedonistic terms, describing its taste, color, and texture (see table). In one study by Mitchell and colleagues, patients were actually very affectionate about their binges.

Dosage also is important, according to Dr. Davis. An increase in snacking behavior over the last generation and the fact that today’s snacks are high in fat and carbohydrates is teamed with portion sizes that have nearly doubled in the last decade. “Small isn’t in the lexicon of fast foods,” she said. A greater array of choices and easy access to food, as well as food additives, have only increased the problem.

Neurobiological parallels. Dopamine (especially via the mesocortical pathways) is an important neurotransmitter in the regulation of motivation and pleasurable activities. These dopamine pathways (along with other neurotransmitters) regulate and foster behaviors that are important to our survival as a species, according to Dr. Davis. The insidious part of drugs of abuse, according to Dr. Davis, is that they activate the same pathways, the common reward pathways, which react to natural reward and pharmacologic agents. In the well-modulated brain, there is an inhibitory component, but when we overtax that system with drugs, the brain adapts, and the neural adaptation insidiously fosters the very behaviors that caused the problem in the first place, and the cravings as well, she explained. “We are now getting evidence that excessive consumption of foods that our brains were not designed to cope with—high sugar and high fat—overtaxes these brain areas, contributing to a compulsion to have more,” she said. Dr. Davis pointed to a study by Dr. Bart Hoebel, another speaker in the session, in which laboratory rat pups given 25% more fat in their diets early in life, then put back on regular rat chow, were all obese as adults.

Environmental parallels. In the case of cocaine and heroin, availability and cost are directly related to the increase in addiction. Food is available 24 hours a day, 7 days a week, and is only a few steps away—it is also cheaper to buy a burger than a healthy, nutritious sandwich, said Dr. Davis.

Psychological parallels. Noting there is considerable human variability to sensitivity to reward and engaging in rewarding behavior, Dr. Davis pointed out that stress and environment can play a role in the development of addiction. Stress degrades the availability of dopamine, as does chronic use of addictive drugs. Low levels of dopamine and related neurotransmitters lead to anhedonic demeanor, she added, and a person may need a dopamine “fix.” Food can be seen as self-medication that will lead to more reward and more pleasure. At the other end of the syndrome, high sensitivity to reward can also increase risk—if a person is highly sensitive to the effects of food, they may engage in those activities more often.

Dr. Davis also described the results of a recent study of personality and eating disorders conducted with three groups of men and women, one with BED, a control group of obese persons, and a second control group of normal-weight persons. Both BED patients and obese controls had eating behaviors that could be described as “hedonic eating,” Dr. Davis told the audience. The BED group scored higher on snacking on sweets, and reported more emotional eating, boredom, depressions, haphazard meal planning, and ate because of sights, smells, and thoughts, not hunger itself.

Patients with AN, in contrast, show a pattern of willful self-starvation, and hyperactivity, and these behaviors can be thought of as an auto-addiction to circulating endorphins, said Dr. Davis. She added that both strenuous exercise and severe starvation increase circulating levels of beta endorphins and, just like morphine and heroin, can become addictive. When these behaviors become entrenched, they may take on a life of their own, a downward spiral of behaviors, just as seen in other addicts, she said.

How Genetics Can Help Unravel the Puzzling Symptoms of AN

Dr. Walter Kaye, Professor of Psychiatry at the University of Pittsburgh and the University of California, San Diego, described neurobiological research into the symptoms of AN.

Imaging studies are helping researchers understand reward pathways, said Dr. Kaye. One area under investigation is a disturbance of hedonic stimuli among these patients with AN. He noted that people with restricting-type AN can be described as anhedonic. These anxious, perfectionist people are fidgety, restless, and prone to over-exercise, all of which are related to alterations in dopamine. In this group, unlike the BED patients described earlier in the plenary session by Dr. Caroline Davis, immediate rewards are not as important. In addition, Dr. Kaye noted that researchers are taking a fresh approach to understanding such behaviors that includes studying so-called behavioral economics, or how experience affects behavior.

“We’ve been dealing with the syndrome but it doesn’t tell us what the brain mechanism is,” said Dr. Kaye. He added, “We need to move way from observing symptoms to finding out what is happening in the brains of these patients.” Powerful new brain-imaging techniques, including positron emission tomography (PET), are helping researchers see how the brain changes in AN and allows researchers to follow activity in the brain.

“We see lots of disturbed things in AN, even in those patients who are recovered and have been at a healthy weight for more than a year,” he said. For example, Dr. Kaye and colleagues found that anorexic patients who had been ill and were recovered had reduced levels of dopamine, suggesting that they had reduced levels in the brain. Dopamine encodes information in many different ways, and with increased dopamine there is increased binding, he said. AN people starve themselves to death, he said, and if eating releases dopamine, perhaps AN patients need a way to reduce this and in essence make dopamine go away, he said.

Dr. Kaye described a recent study conducted by his group that tested for reward and loss, a monetary choice task that activated the striatum. The study involved 13 patients who had recovered from AN and 13 control women. All the women looked at a computer screen, and guessed an answer; if they were correct, they received a dollar. The game was designed so that the women would win about $40. Among the control women who won, there was a change in neural activity from baseline, and then the activity went back to normal. However, among the women with AN, there was a blunted response in neural activity whether they won or lost. Those with AN did have an exaggerated response in the caudate nucleus. In the planning circuits of the brain these patients have an overconcern about future consequences, while the control women just made a guess, he said. Dr. Kaye added that the women with AN were trying to get it right each time, and to figure out a strategy to do so. The information-processing neurocircuits dealing with planning were very active among the AN patients.

Dr. Kaye said that the positive aspects of what is being learned about the neurobiological circuits of the brain and women with AN is that different people have different balances in the brain that handle emotional response and consequences. In the AN temperament, there is a different balance, he said, which explains why such patients can focus on planning and consequences and why they are achievement-oriented, obsessive and detail-oriented. Long ago the increased ability to plan ahead may have meant surviving or not surviving, he said, and environment may have shaped certain tendencies, he added.

Current treatments have borrowed from other disorders and there is a huge amount of emphasis on treatment but little on research, and much more is needed, he said. Dr. Kaye said that finding biomarkers in AN nervosa is very important for one more reason: to demonstrate to insurance companies that there are underlying physiological and biological reasons for the eating disorders.

An Animal Model of Binge Eating and Bulimia Nervosa

Animal models of rats eating to stimulate dopamine release can provide much information about binge eating and BN, according to Bart Hoebel, PhD, Professor of Psychiatry at Princeton University. Dr. Hoebel noted that while many researchers do not like the term “sugar addiction,” studies of increased feeding of glucose in rats has provided some helpful clues to binge eating in humans.

Dr. Hoebel said that all drugs of abuse, including nicotine, cocaine, and the opiates, can increase extracellular dopamine, and food might also release dopamine. Aberrant sugar intake, a model of sugar addiction, is also a model of the eating disorders, he said.

Dr. Hoebl noted that in the animal model, if an animal fasts 12 hours a day every day for a month, it self-escalates its intake of glucose; in effect, it learns to binge. Among animals that are given intermittent sugar, a huge meal in the first hour and as much in 12 hours as in 24 hours, opiate bingeing is increased and the D1 receptor intake increases.

Why does sugar take on addiction-like properties? Dr. Hoebel explained that bingeing on sugar can release dopamine regularly. According to Dr. Hoebel, the reason that people and rats binge may be that dopamine is released and they are self-medicating to alleviate discomfort, anxiety, and depression. And, just as with any addictive drug, they need more sugar to accomplish this.

Mary K. Stein
Mary K. Stein

Managing Editor

No Comments Yet

Comments are closed