Sexual Abuse and Eating Disorders: A Test of a Conceptual Model

Abstract

To provide a better conceptual understanding and to stimulate further research, A. Kearney-Cooke and R. H. Striegel-Moore (1994) proposed several theoretical models concerning the relationship between sexual abuse and eating disorders. In this study, we tested the model that hypothesized an indirect relationship between sexual abuse and eating disorders, with the effects of sexual abuse being mediated through bodily shame and body disparagement. Three hundred thirty female undergraduates from a large southwestern public university participated. The women represented the social class of the university, which is predominantly middle to upper-middle class. Race/ethnicity of the participants was 69.8% Caucasian, 15.9% African American, 5.8% Latino/Mexican American, 4.9% Asian American, and 0.9% Native American; the remainder did not indicate their race/ethnicity. Of the total sample, 60% reported having been sexually abused at some point in their lives; 21% had experienced abuse prior to age 14. Almost 8% were categorized as having a diagnosable eating disorder, whereas another 72.7% were symptomatic. To test the hypothesized model, structural equation modeling with LISREL 8.3 (K. G. Joreskog & G. Sorbom, 1999) was used. Results indicated that the model fit the data well and the hypothesized relationships among the variables were in the expected directions. Sexual abuse predicted higher levels of bodily shame that, in turn, predicted increases in body disparagement. As expected, only body disparagement directly predicted eating disorder symptoms. Directions for future research are discussed.