Heightened accuracy of self-reported weight in bulimia nervosa: a useful cognitive "distortion".

OBJECTIVE

This study was designed to test two competing hypotheses regarding bias in self-report of weight and height in bulimia nervosa.

METHOD

General population samples of 102 young women with bulimia nervosa and 204 age and social class-matched healthy control women were recruited. Subjective and measured values of height and weight were obtained and compared within and between the groups.

RESULTS

The healthy control subjects reported that they were lighter and taller than they actually were, and the degree of this discrepancy was related both to their actual weight and, more strongly, to the difference between their actual and desired weight. In contrast, the bulimia nervosa subjects showed little reporting bias.

DISCUSSION

The results support the hypothesis that the intense interest of bulimia nervosa subjects in their weight is expressed by heightened precision of their self-report. Epidemiological studies of bulimia nervosa can capitalize on this cognitive "distortion."