Development of a psychological test for eating disorders which has screening and onset-predictive functions

点击翻译按钮获取中文摘要

1. Reliability of existing screening tests for eating disorders (ED)(1)EAT-26 : First-year university students with more than 15 points on EAT-26, and/or with menstrual abnormality were structurally interviewed. 29 were diagnosed as an AN-group (AN n=2 ; past AN n=2 ; EDNOS-1 n=4), 8 as a BN-group (BN n=7 ; EDNOS-3 n=1), 9 as BE. Screening ability of a menstrual questionnaire for ED was higher compared to that of EAT-26.(2) SCOFF (BMJ 319 : 1467-1468, 1999) : Female undergraduate students marking "yes" to more than two items of SCOFF questions, were structurally interviewed. 7 were diagnosed as EDNOS-1, 9 as a BN-group, 10 as a BE-group. Two 7s of them were also diagnosed as seasonal affective disorders (SAD), and as suspected SAD, respectively. These data suggested the followings : First, screening ability of SCOFF for ED was higher than that of EAT-26. Second, a questionnaire about menstrual abnormality was needed to find AN. Third, co-morbidity of SAD was higher in patients with bulimia than in those with anorexia.2. Reliability of UPI and SDS as a prediction test for the onset of EDThe subjects were undergraduate students who developed ED during school period. The number of students with AN, BN, and BE were 20, 17, and 4, respectively. Their points of UPI, SDS, and each factor item at admission and after the onset were compared to each other, and to control groups. There was no significant difference among points of different groups at admission. Nevertheless, points of any subtypes of ED had increased in amount after the onset, yielding a significant difference. These data suggested the followings : First, it was impossible to predict the onset of ED using these tests. Second, personality characteristics including introversiveness, obsessiveness, dependency, impulsiveness, and low-graded self-esteem, often seen in patients with ED, was state dependent.