An unusual union of disciplines resulted in a new possibility in diagnosing and treating eating disorders.

The study, published Monday in Rapid Communications in Mass Spectrometry, was the brainchild of three Brigham Young University professors. The research looked at the differences in a person's hair if that person has anorexia or bulimia. At the completion of the study, researchers had an equation with an 80 percent success rate in diagnosing people with eating disorders.

"They're looking for an objective physiological test," said Michael Smart, a spokesman for the university.

Currently there is no test that can definitively diagnose an eating disorder; the diagnosis is made based on questionnaires, interviews and symptoms. Often the diagnosis can be missed because women who have eating disorders are in denial or won't admit they have the disease, so they aren't completely honest with their doctors. This test could change that.

Between the scientific discovery and the experience of working with people in different fields, the researchers called it a success. The idea started when Stephen Nelson, a geologist at BYU, was talking with Steve Thomsen, a communications professor who has studied the effect of print media on young women and if it affects the incidence of eating disorders. Nelson then connected Thomsen with Kent Hatch, an assistant professor of integrative biology who was researching how carnivorous black bears were based on the makeup of their hairs.

"That's kind of where the whole genesis of the project came from," Thomsen said. "We just kind of wondered out loud together."

Hatch's experiment looked at the eating habits of bears, which could be determined by looking at the types of nitrogen in their hairs. He's seen these trends in other animals as well.

"I figured you could probably do that with humans as well," he said.

So they did. The study included hairs from a number of people diagnosed with eating disorders and a control group of students. People with eating disorders generally go through a certain progression, where they cut out meat, then cheese and eggs, then continue moving toward a high-fiber, low-protein diet, until eventually the body starts eating itself to get enough energy.

The eating habits are mirrored in the nitrogen and carbon in each hair. A person who eats meat will have more nitrogen-15 than a person who doesn't, while a person who isn't getting energy from outside sources will see an increased level of nitrogen-15, which indicates the body is getting protein from itself. The carbons also shift in the hairs, Hatch said.

So they looked at the hairs, did several tests, went back to the laboratory, ran some statistics and did several more tests. Then they came up with an equation that takes into account the difference between nitrogen and carbon isotope ratios at the tip and the base of a strand and the variance along each hair.

"It can do better than I can," Hatch said. "When you plug that in, you're accurate about 80 percent of the time."

This kind of research is important, Thomsen said, because an eating disorder is usually diagnosed based on what a patient tells her doctor. If she is in denial about her condition, doesn't want to admit she has a problem or is unsure about her eating habits, the diagnostic test can indicate scientifically if the patient is sick.

It also can be beneficial in treating eating disorders, in case patients aren't following the treatment regimen on their own. Often the psychological aspect of eating disorders can cause women to not speak up or give total information about their problems.

"I've felt that to some degree whenever you study disordered eating you tend to get underreporting," he said.

The next step for the researchers is finding everything that can be wrong with the tests and ways to fix it. Hatch said other factors could result in the altered hair makeup, and he wants to look at how the test could be affected if the subject was on a diet, was a vegan or an athlete who had to maintain a certain weight.

"Those are the things we need to look at next," he said.

In addition to Nelson, Hatch and Thomsen, associate professor of statistics Dennis Eggett, professor of integrative biology Beverly Roeder, and Morgan Crawford and Amanda Kunz, both undergraduate students at BYU at the time of the study, worked on the project.

Heidi Toth can be reached at 344-2543 or This e-mail address is being protected from spam bots, you need JavaScript enabled to view it