The eating attitudes test: twenty-five years later

Abstract

This manuscript reviews the literature involved with the Eating Attitudes Test (EAT), first developed in the late 1970s as a self-report, indicative of the symptoms of eating disorders. The EAT has good psychometric properties of reliability and validity, and reasonable sensitivity and specificity for the eating disorders, but very low positive predictive value because eating disorders are relatively uncommon. In addition they exist on a continuum, because of denial and social desirability, the results of a self-report instrument may be affected. A very large literature has documented the use of the EAT in a variety of cultures. It is used to screen eating disturbances in general as the first part of a two-part diagnostic screen, as an ability to compare across groups and to measure change between groups and over time.