Establishment of a Japanese version of the Sick, Control, One Stone, Fat, and Food (SCOFF) questionnaire for screening eating disorders in university students.

OBJECTIVE: This study aimed to validate the Sick, Control, One stone, Fat, and Food (SCOFF) questionnaire in relation to the Eating Disorders Examination Questionnaire (EDE-Q) and to examine the appropriateness of a question concerning weight loss among Japanese university students. The psychometric properties of the two Japanese versions were determined among 649 Japanese college students. The original version (SCOFF-O) employed the original item 3, whereas the revised version (SCOFF-2.5) modified the item to "Have you recently lost more than 2.5 kg within three months?" Validity was tested relative to EDE-Q. RESULTS: The test-retest reliabilities of SCOFF-O and SCOFF-2.5 were 0.52 and 0.57, while the correlations of SCOFF-O and SCOFF-2.5 with EDE-Q were r = 0.53 and r = 0.56. The sensitivity and specificity of SCOFF-O were 65.2 and 89.7, and those of SCOFF-2.5 were 69.5 and 86.5, respectively. There were significant correlations between the question concerning losing 2.5 kg and the EDE-Q subscales. The Japanese version of SCOFF-2.5 is an appropriate tool for the early screening of eating disorders among Japanese university students.