Internal consistency and validity of the BITE for the screening of bulimia nervosa in university students, Colombia

OBJECTIVE: To establish the validity and internal consistency of the bulimic investigatory test, Edinburgh (BITE) in a non-clinical population of university students in Colombia. METHODS: A validation study with a cross-sectional sampling was designed. Two-hundred and sixty-one students were evaluated independently and blindly with BITE and the Composite International Diagnostic Interview. Kuder-Richardson's test, sensitivity, specificity, predictive values, likelihood values, Cohen kappa, area under the ROC curve for BITE's symptoms scale and Cronbach's alpha for severity scale were computed. RESULTS: The Kuder-Richardson's test was 0.86, two factors were identified which explain the 22.68% and 7.12% of variance. Cronbach's alpha for severity scale was 0.52. Area under ROC curve was 0.9736. The best cutoff point was 11 with sensitivity of 94.59%, specificity of 91.52%, positive predictive value of 64.81%, negative predictive value of 99.03%; with a concordance of 0.7226. CONCLUSIONS: The BITE's symptom scale is useful in identifying clinically binge-eaters in university students.