The development and validation of a new instrument to measure eating disorders and dissociative symptoms.

摘要

The necessity to have an objective assessment instrument that screens for dissociative symptoms in eating-disordered patients as well as eating problems in dissociative patients has been evidenced by the proliferation of patients diagnosed with eating disorders and dissociative disorders. Since there is, at present, no known single and validated instrument designed to examine both eating disorders and dissociative processes, the EDDS-Q will eliminate the need to administer several different instruments in order to study the relationship between the two.;The purposes of this study were to (a) develop and validate a psychometric instrument, titled "EDDS-Q" (Eating Disorders and Dissociative Symptoms Questionnaire), (b) examine the relationships and differences between the EDDS-Q and other relevant psychosocial measures, and (c) to examine specific group comparisons to assess the predictive capability of the EDDS-Q.;The instrument is a 91-item, self-administered, paper and pencil questionnaire. The EDDS-Q score is based on a 5-point, multiple choice, experiential format. Subscales include: eating behaviors (EB), food-related experiences with dissociation (FRED), childhood memories (CM), childhood memories with dissociation (CMD), dissociative symptoms (DS), and self-abuse (SA). Supplemental indexes were created to measure childhood traumata, and autohypnosis and hypnotic anesthesias.;The scale was administered to 386 subjects comprising five groups: (1) normal control (NQ university students, (2) individuals diagnosed with eating disorders (ED), (3) individuals diagnosed with dissociative disorders (DD), (4) individuals diagnosed with both eating disorders and dissociative disorders (EDDD), and (5) individuals diagnosed with other psychiatric disorders (OD).;Good test-retest reliability was demonstrated on five of six subscales. The results also showed that five of the EDDS-Q subscales had good internal consistency. The EDDS-Q correlated highly with other relevant psychosocial measures revealing major similarities and differences among groups, which demonstrated a high degree of construct, convergent and discriminant validity. Results demonstrate that the EDDS-Q accurately classified subjects into one of several groups which supports a high level of predictive validity. Research hypotheses were formulated to determine the nature and relevance of the predictions. These findings provide strong initial support for the reliability and validity of the EDDS-Q.