Epidemiological Studies of Eating Disorders in Gifted Dance Students and Ordinary High School Students:Prevalence Rate, Risk Factors and 1-Year Outcome (1st Year).

This is the first-year part of a two-phase prospective study. The aims of this study are to investigate the prevalence rate of eating disorders in Taiwan, it's correlated psychosocial risk factors and establish the validity index of BITE. The study subjects consist of Taiwanese gifted dance high school students and comparable ordinary class students. Parental informed consent for nonpartipation was delivered 3 days before study started.All students completed questionnaires(Eating Altitude Test (EAT-26), Bulimic Investigatory Test, Ediburgh (BITE), Body Shape Questionnaire (BSQ), Physical Appearance Related Teasing Scale, Pubertal Develoment Scale, and Brief Symptom Rating Scale (BSRS) at first phase. All subjects who are above threshold (EAT-26 score≧20 and BMI≦17.5Kg/m2; the symptom scale score of BITE≧15 and/or severity scale≧5),and 10% of the sub-threshold cases need to receive second-phase interview by two senior psychiatrists. The contents of interview included SCID-IP, and risk factors assessment for eating disorders. The known risk factors included both personal factors (history of obesity, psychiatric, menstruation, teasing, or sexual or physical abuse, etc.) as well as family factors (parental problems, history of obesity, dieting, psychiatric illness etc.).

Statistical analyses were conducted using SPSS 10.0 for Windows/PC. Categorical data were analyzed by nonparametric test (Chi-square test). ANOVA or Student's t-test was used for continuous data analyses. Reliability is assessed by internal consistency and test-retest reliability. Inter-rater reliability was assessed by Kappa statistics. The sensitivity and specificity for eating disorders were calculated for individual screening instrument.ROC analyses were undertaken to evaluate the overall performance of the BITE +/- EAT in detecting eating disorder cases. Regression analysis was applied to determine predictors of disordered eating or eating disorder status. A value of P< 0.05 was considered statistically significant.