(The New York Times News Service) -- San Francisco -- Too many young people with serious eating disorders are being given "catch-all" diagnoses that could underestimate the severity of their illness and prevent them from getting the best possible treatment, a Stanford researcher says.

Between 1 percent and 5 percent of Americans suffer from anorexia or bulimia, according to the National Institutes of Health. But many more people -- roughly two-thirds of all people diagnosed with an eating disorder -- fall into the hazy category known as "eating disorder, not otherwise specified" because they don't meet all the criteria for anorexia or bulimia.

The result is that these patients may not get the medical attention they need, although in many cases they suffer the same health problems, including risk of heart failure and severe malnutrition, as people with the more well-known illnesses, said Dr. Rebecka Peebles , an instructor of adolescent medicine at Stanford University School of Medicine.

"We know that people who meet the criteria for anorexia and bulimia are quite ill. The question is whether there are other people we are missing who are the walking wounded," Peebles said.

In a paper being published today in the journal Pediatrics , Peebles reports that she and her colleagues studied 1,310 girls, ages 8 to 19, who were treated for an eating disorder between 1997 and 2008. Only 25 percent were diagnosed as anorexic and 12 percent as bulimic -- the remaining 63 percent fell into the "not otherwise specified" category, known as EDNOS.

The risk of lumping patients into the unspecified category is that many people, including some doctors who don't regularly treat eating disorders, assume the diagnosis is less serious than anorexia or bulimia.

Perhaps equally important, many insurance companies will only pay to treat patients who get the official anorexia or bulimia diagnosis, say doctors who regularly treat patients with eating disorders.

"There are insurance companies that don't pay for this diagnosis (of not otherwise specified). They're just waiting for people to fall off the edge," said Dr. Don Mordecai , director of mental health and chemical dependency services for Kaiser Permanente Northern California.

The Stanford paper comes at a convenient time, just as the psychiatric community is in the middle of what could be a major overhaul in the way eating disorders are diagnosed. The changes are expected to appear in the next edition of the American Psychiatric Association's diagnostic manual , which is widely used by doctors and therapists nationwide.

Doctors don't want to over-simplify the symptoms and have their patients be inappropriately labeled as anorexic or bulimic, said Dr. Evelyn Attia , a member of the committee that is reevaluating the eating disorder category in the diagnostic manual. But she said it's clear that the diagnostic criteria need to be refined so that more patients who currently fall into the "not otherwise specified" category get the proper treatment.

The Stanford paper, Attia said, is notable for highlighting some of the problems with the current diagnoses. In particular, the criteria may be especially tough to apply to children and teenagers.

For example, one symptom doctors look for in diagnosing anorexia in a female patient is whether her menstrual cycle has stopped. If a 12-year-old girl hasn't started menstruating, it would be impossible to apply that symptom to an anorexia diagnosis.

"A younger patient can be very difficult to evaluate, and that has contributed to a lot of people winding up in this limbo category rather than receiving the (anorexia or bulimia) diagnosis," said Attia, who is director of the eating disorders program at Columbia University Medical Center.

In the Stanford study, girls with unspecified eating disorders had rates of serious health problems like irregular heartbeats, low blood pressure and low nutrient levels that were similar to girls with anorexia or bulimia, Peebles said. In some cases, girls with the unspecified diagnosis actually had more severe health problems.

Girls who had lost more than 25 percent of their weight in a short period of time, but were still a "normal" weight, for example, wouldn't be considered anorexic -- yet they sometimes had more serious health problems than girls who were anorexic and looked alarmingly thin.

Doctors who are familiar with treating eating disorders might recognize rapid weight loss as a sign of a significant mental illness, Peebles said. But a pediatrician or primary care doctor might not -- and, in fact, might congratulate a patient on his or her weight loss.

"A lot of these kids, they've been patted on the back, and it's not until this goes on awhile that people realize there's something wrong," Peebles said. "We're learning more and more that you can't tell by looking at someone if they have an eating disorder.

It raises the question, when do we take someone seriously?"

Copyright 2010 The New York Times News Service. All rights reserved.