Cover Story

Long before the Human Genome Project was a twinkle in the eye of a geneticist, researchers were conducting twin studies that found eating disorders probably have a genetic component. New studies by psychologists show the value such efforts still hold for the field.

Using statistical analysis of similarities and differences among twins, the studies suggest the degree to which genetic and environmental influences play a role in the development of eating disorders, and point to what those influences might be.

"Findings from these studies can underscore that, yes, a certain condition is genetically influenced," says Cynthia Bulik, PhD, a psychologist and eating disorders expert at Virginia Commonwealth University known for her studies on twins. "Once we have that information, we can go in and look for specific genes related to that area."

Twins studies examine the relative contributions of genes and the environment by noting statistical differences between identical twins, who share 100 percent of their genes, and fraternal twins, who share 50 percent of their genes.

"If two members of [identical] twin pairs have a disorder more commonly than [fraternal] twin pairs, that starts to give us evidence for a genetic component to a given illness," Bulik explains. "Using statistical analysis, we can start to put numbers on those differences."

Psychologist Tracey Wade, PhD, of Flinders University in Australia, Bulik and colleagues have laid some basic groundwork on anorexia genetics in a study of anorexia in 2,163 female twins. In their study, published in the March 2000 issue of the American Journal of Psychiatry (Vol. 157, No. 3), they found that genes accounted for 58 percent of the variance in twins who had anorexia nervosa, while unique environmental factors--environmental factors experienced by only one of the twins--made up the remaining 42 percent.

In another study on bulimic twins by Bulik, Wade and colleagues, the team showed possible reasons why one member of a pair of identical twins may develop bulimia nervosa and the other doesn't. Twins who developed bulimia nervosa, they found, were more likely to report generalized anxiety disorder and to be described by their mothers as anxious and fearful as children. They also showed lower mastery, optimism and self-esteem than their unaffected twins, and more obsessive and compulsive symptoms.

All of these factors point to possible genetic factors for future study, the authors note. In the realm of possible unique environmental effects, the twins who developed bulimia nervosa also recalled greater discord in their families than the unaffected twins. The findings are reported in the January 2001 issue of the International Journal of Eating Disorders (Vol. 29, No. 1).

In a related set of twin studies using data from the Minnesota Twin Family Study, psychologist Kelly Klump, PhD, of Michigan State University, has also been finding a strong genetic component to eating disorders in girls and young women. In a study reported in the May 2000 issue of the Journal of Abnormal Psychology (Vol. 109, No. 2), Klump examined developmental differences in genetic influences on disturbed eating patterns in 340 pairs of 11-year-old twins and 301 pairs of 17-year-old twins. The analysis showed no genetic influence on disordered eating in the 11-year-olds but a 55 percent contribution from genes in the 17-year-olds, suggesting that puberty plays a role in the onset of eating pathology.

In her next study, Klump examined disordered eating patterns in the same set of 11-year-old twins, this time divided into a pre-pubertal group and a pubertal/post-pubertal group. That study, now in press in the Journal of Abnormal Psychology, found no genetic effect for eating pathology in the pre-pubertal sample, but a genetic contribution of about 50 percent for the pubertal girls.

Klump now has pending a National Institute of Mental Health grant to examine the relationship between eating pathology and hormones activated during puberty, such as estrogen, progesterone and cortisol. It will be the first study to look directly at this possible link.

"Eating disorders are interesting because they're the epitome of the biopsychosocial model," Klump comments. "What's generally been neglected in the past, and what these studies are helping to address, is the biological part."