A study showing a significant increase in hospitalization for eating disorders is cause for alarm - but the reality is even more alarming.

The study from the federal Agency for Healthcare Research and Quality (AHRQ) shows that eating disorder-related hospital stays increased 18 percent from 1999-2000 to 2005-2006, with a total of 28,155 patients treated in hospitals. The study does not include binge-eating disorder, which is the most common eating disorder.

It is uncertain whether the study reflects an increase in the population with eating disorders, an increase in hospitalization among those with eating disorders or an increase in awareness of medical staff who are diagnosing eating disorders. It is likely that the study reflects all three to some degree.

The real news, though, requires an analysis of the statistics, as well as knowledge about issues that are not included in the study:

Eating disorders affect a cross-section of society. The widely held perception that eating disorders are a fad that affects only college-aged women is disproved by the AHRQ statistics.

The study shows that hospitalization of children under 12 with eating disorders more than doubled, increasing 119 percent during the six years of the study. In addition, the AHRQ found a 48 percent increase for patients 45 to 64 and a 37 percent increase for men. These statistics verify that eating disorders affect the whole population.

While children, men and mature adults still represent a small percentage overall of patients with eating disorders, that percentage is growing. Walden Behavioral Care, which is one of the few facilities in the country to treat young people and males, also has seen increases in treatment by these populations. However, young people and males as a group may be undertreated, because of the shortage of facilities to treat them.

Eating disorders cause serious health problems. The AHRQ data notes that eating-disorder patients are being treated for serious conditions, such as cardiac dysrhythmias, and acute renal and liver failure.

The reality is even worse. If left untreated, eating disorders can damage the brain, liver, kidneys, heart, GI tract, bones, teeth, skin and hair. They can result in serious medical conditions, such as osteoporosis, retarded growth, kidney problems, ulcers and heart failure.

In addition, most patients with eating disorders have other disorders. They may suffer from post-traumatic-stress disorder or depression, and they may have issues with substance abuse. Alcohol abuse is especially common among patients with anorexia.

Worse still, patients with anorexia have the highest rate of suicide of any psychological disorder. A 2003 Archives of General Psychiatry (Vol. 60, No. 2) study found that those with anorexia are 56 times more likely than their peers to take their lives.

A study just published in the British Journal of Psychiatry found that the mortality rates was six times that of the general population for the more than 6,000 Swedish women included in the study. The women included all received in-patient treatment for anorexia from 1973 through 2003. Suicide was cited as the cause in 20 percent to 30 percent of all deaths.

One out of every five people with anorexia eventually die of causes related to the disorder, according to The Monitor, which is published by The American Psychological Association.

Most patients with eating disorders are not treated for them. Only a minority of people with an eating disorder receive treatment specifically for their disorder, according to The Monitor.

Many patients with eating disorders try to hide their disorders. They may not want to be treated, they may feel shame or they may not recognize that they have a disorder.

In addition, medical staff may not recognize when patients have eating disorders. The AHRQ is based on diagnoses by primary care physicians, so many are recognizing when patients have eating disorders, but many are not. Medical professionals will certainly diagnose and treat medical conditions, such as cardiac dysrhythmias, but they may not recognize when patients also have eating disorders, which are psychiatric illnesses.

Eating disorder patients need specialized care. Given the medical complications associated with eating disorders, individuals who have advanced eating disorders need the 24-hour medical attention that hospitals provide - but they also need behavioral and nutritional care.

Eating disorders also require longer-term treatment than hospitals typically provide. Ideally, patients should advance along a continuum of care that may include residential treatment, partial hospitalization and intensive out-patient care. Patients frequently have setbacks along the way and treatment should be adjusted accordingly.

Few hospitals are equipped to understand and meet the treatment needs of eating-disorder patients. In fact, Walden Behavioral Care is one of only a few hospitals in the country to provide specialized in-patient care for patients with eating disorders.

A majority of patients with eating disorders are not classified appropriately. The AHRQ study comes at a time when work is under way to improve definitions of eating disorders. An Eating Disorders Work Group, an interdisciplinary group including psychiatrists and psychologists, has been organized to review and update criteria for eating disorders included in the Diagnostic and Statistical Manual of Mental Disorders. Revised criteria are due to be published in 2012.

This is significant, because DSM criteria are used by insurance companies to determine whether to cover patients with eating disorders.

Under the current standards, a majority of patients with eating disorders are classified as having "eating disorders not otherwise specified," or EDNOS. According to The Monitor, EDNOS accounts for 60 to 65 percent of eating-disorder diagnoses. EDNOS is typically not covered by insurers.

Binge-eating disorder is the most common eating disorder, yet it does not yet have its own official classification and is lumped under EDNOS. The Eating Disorders Work Group is considering whether to classify binge-eating as an eating disorder with its own criteria.

The criteria for anorexia nervosa and bulimia nervosa are also so confining that most patients do not qualify. For example, current criteria include amenorrhea as a requirement for diagnosing a patient with anorexia, but some women who have symptoms of anorexia do not stop menstruating. The criterion also doesn't take men into account or women using birth control pills.

How we define weight criteria is another key issue. Ideal body weight (IBW) is often used to determine whether a patient has an eating disorder. Several IBW formulas exist, but the commonly used Devine formula, which was developed for prescribing dosages, skews too low in determining an appropriate weight for short women. Using the Devine formula, an IBW for a woman who is five feet tall would be just 100 pounds. Given that 75% of IBW is a common criterion for diagnosing eating disorders, a woman who is five feet tall and weighs 80 pounds might not be classified as having anorexia.

A revised formula developed by J.D. Robinson, M.D. in 1983 is more accurate for women, but tends to be problematic for tall men.

More medical providers are relying on body mass index (BMI), which measures weight, adjusted for height. To calculate BMI, divide a person's weight by height in inches squared and then multiply by 703. A person with a BMI of 18.5 is considered underweight and a person with a BMI of 30 and above is considered obese, regardless of height or gender. A BMI of 19-24 is the accepted range for optimal health. BMI can also be used to identify discrepancies in IBW formulas.

When BMI is used for admission criteria, patients are admitted when BMI is less than 16. Insurance carriers tend to be open to paying for treatment of individuals who have eating disorders based on BMI, even when the individuals might fail to qualify based on IBW.

Flexibility is in everyone's best interest, because if treatment is delayed, it becomes more difficult and more expensive to treat the patient. When patients fail to meet the guidelines, they have a perverse incentive to lose weight or become medically unstable, so they can qualify for treatment.

Other criteria may also need to be updated and it is in everyone's best interest to do so. It is, of course, better to treat patients in the early stages of an eating disorder than it is to wait until the disorder is advanced. The prognosis for recovery is higher and the cost of treatment is lower if the patient is treated early.

The AHRQ study is merely a statistical review and was not meant to examine issues such as the adequacy of care for patients with eating disorders or whether criteria for identifying eating disorders is adequate. However, by raising awareness of the seriousness of eating disorders, it may accomplish a great deal more.

Stuart Koman, Ph.D. is President and CEO of Walden Behavioral Care in Waltham, Mass. and can be reached at skoman@waldenbehavioralcare.com. James Greenblatt, M.D. is Chief Medical Officer of Walden Behavioral Care in Waltham, Mass. and can be reached at jgreenblatt@waldenbehavioralcare.com