Predicting the Onset of Anorexia Nervosa and Bulimia Nervosa

The etiologies of anorexia nervosa (AN) and bulimia nervosa (BN) are still uncertain. Although cross-sectional studies have produced useful hypotheses about risk factors, they haven’t been able to clarify whether the risk factors actually preceded development of an eating disorder or if they were consequences of it.

Researchers at four universities investigated prospective predictors of partial and full syndrome AN and BN among 157 young women (Int J Eat Disord 2002; 32:282). The women were first studied in the 7th through 10th grades (ages 12-16) and followed up 8 years later in young adulthood (ages 20-24). A telephone interview assessed eating habits, current dieting, weight and menstrual history, and feelings about food and weight. The Structured Clinical Interview for DSM-III-R (SCID) was used for screening for AN and the SCID was adapted to include open-ended questions about binge characteristics and concerns about shape and weight for BN.

Full and partial syndromes

Full-syndrome diagnoses were made according to DSM-IV criteria. For partial syndromes, the researchers used the category of eating disorder not otherwise specified. Partial syndrome AN was diagnosed when a participant reported a time when she was preoccupied with weight or dieting, at least 15% underweight, or others thought she was too thin. In addition, partial syndrome AN was diagnosed when a participant had any of the following: feeling that food controlled her life or compulsive eating habits, purging and/or amenorrhea.

Partial syndrome BN was diagnosed when the participant reported a time in which she engaged in regular binge eating (on average 3 times a month) and felt a lack of control over eating, as well as using compensatory purging. Frequency, duration and over-concern with shape and weight criteria for BN were not required for the partial syndrome.

What predicted the onset of AN and BN?

The most clear-cut risk for developing AN involved thinness and perfectionism. The 7th to 10th grade girls who went on to develop anorexia nervosa as young adults were initially thinner than the other girls. The measures of perfectionism used in the study mostly assessed the participant’s propensity for setting rigid, unrealistic standards for herself, striving to meet those standards, and equating lack of complete success in these goals with complete failure.

The authors comment that girls with such impossibly high standards are likely to have difficulty meeting the demands of adolescence, which include adjusting to a new body shape that often doesn’t conform to the excessively thin physique held up as the “feminine ideal.”

This characteristic also works against the flexibility and ability to take risks that are needed for developing a sense of mastery and for establishing close social relationships. Thus, such girls may shrink from the new demands posed by adolescence and their rigid perfectionist approach to dieting may allow development and maintenance of the very low body weight seen in this disorder. It also appeared that the initial thinness was not due to excessive dieting or eating problems because the initial EAT scores were not significantly higher in girls who developed the anorexic syndrome.

The predictors for bulimia nervosa were less clear-cut, according to the researchers. Negative emotion was a significant predictor and this negative affect might contribute to the development of the binge-purge cycle.