Acta Psychiatr Scand 1998: 97: zyxwvutsrqponm 358-343 Printed in zyxwvutsrqponm UK zyxwvutsrqponm - all rights reserved CoDvriehf zyxw 0 Munksgaard 1998 ~ ', zyx ACTA PSYCHIATRICA SCANDINAVICA ISSN 0001-690X zyx Impact of a history of physical and sexual abuse in eatine disordered and asymptomatic"subj ects Favaro A, Dalle Grave R, Santonastaso €? Impact of a history of physical and sexual abuse in eating disordered and asymptomatic subjects. Acta Psychiatr Scand 1998: 97: 358-363. 0 Munksgaard 1998. The present study aimed to explore the impact of sexual and/or physical abuse among eating disordered patients (ED) and asymptomatic subjects. A total of 86 patients with anorexia nervosa, 69 patients with bulimia nervosa and 81 asymptomatic subjects were assessed. Among ED, we did not find a significant association between abuse experiences and the severity of the eating disorder, or between abuse and dissociative symptoms. Among ED, self-destructive behaviour appears to be the most important predictor of a history of sexual and/or physical abuse. In contrast, in the asymptomatic group, the score on the Dissociation Questionnaire is the only significant predictor of reported abuse experiences. zyxwvut Introduction Abuse experiences during childhood and early adolescence may have important consequences for adult psychopathology (1, 2). Although some studies have suggested that experiences of child- hood sexual abuse may contribute to the develop- ment of eating disorders, the recent literature has not confirmed the existence of a specific causal link between sexual abuse and eating disorders (3). Among victims of childhood abuse, the risk of developing an eating disorder appears to be no greater than that of developing any other psychi- atric disorder (4). High levels of dissociative symptoms have been observed among subjects with a history of sexual abuse, in both non-clinical (5, 6) and clinical sam- ples (7, 8). DSM-IV (9) describes dissociative disorders as a disruption of the usually integrated functions of consciousness, memory, identity or perception of the environment. When a history of trauma is present, dissociation has been interpreted as a defence against unbearable emotional turmoil. In fact, it has been observed that traumatized people may develop dissociative symptoms, such as memory lacunas, a disturbed self-perception, depersonalization and derealization, not only as an acute reaction, but also as a form of adaptation to trauma (10). A. Favaro', R. Dalle Grave', P. Santonastaso' 'Department of Neurological and Psychiatric Sciences. University of Padova, Padova and *Casa di cura Villa Garda, Garda. Verona, Italy Key words physical abuse, sexual abuse, dissociation. eating disorders Paolo Santonastaso. Dipartimento di Scienze Neurologiche e Psichiatriche. Via Giustiniani 3, Accepted for publication July zyxw 25. 1997 ~ 35128 Padova. Italy In eating disordered samples, the origins and role of dissociative symptoms are unclear. Various studies have reported high levels of dissociative symptoms in eating disordered patients, especially in patients with bulimia nervosa (11-13). More- over, dissociation appears to be correlated with measures of eating psychopathology among both bulimic and normal females (13). Bulimic patients often describe binge-eating as being associated with dissociative states (14), and some authors have postulated that binge-eating could be a form of dissociation, i.e. a way of inducing dissociative states and an alternative means of reducing the awareness of intolerable feelings following the abuse experience (15). Attempts to verify this hypothesis have yielded conflicting findings (16). In fact, some studies have found a significant relationship between dissoci- ation and the presence of sexual abuse (17), while in others dissociation appeared to be linked to the presence of physical, rather than sexual, abuse (18). Moreover, various studies of bulimic samples have found that dissociation is significantly associated with self-mutilation (19) and depression zy (12), both of which are typical of bulimic patients. However, no study to date has sought to establish by means of a multivariate statistical procedure which clinical z 358