The optimal cut-off score of the Eating Attitude Test-26 for screening eating disorders in Japan

Purpose The Eating Attitude Test-26 (EAT-26) is a screening tool for eating disorders (EDs) in clinical and non-clinical samples. The cut-off score was suggested to be varied according to target population. However, no studies have examined the appropriateness of the originally proposed score of 20 for screening DSM-5 eating disorders in Japan. This study aimed to identify an appropriate cut-off score to better differentiate clinical and non-clinical samples in Japan for EDs.Methods The participants consisted of 54 patients with anorexia nervosa restricting type, 58 patients with anorexia nervosa binge-eating/purging type, 37 patients with bulimia nervosa diagnosed according to DSM-5 criteria, and 190 healthy controls (HCs). Welch's t test was used to assess differences in age, body mass index (BMI), and total EAT-26 scores between HCs and patients with EDs. Receiver operating characteristic (ROC) analysis was conducted to identify the optimal cut-off score.Results The HCs had significantly higher BMI and lower total EAT-26 mean scores than patients with EDs. The area under the ROC curve was 0.925, indicating that EAT-26 had excellent performance in discriminating patients with EDs from HCs. An optimal cut-off score of 17 was identified, with sensitivity and specificity values of 0.866 and 0.868, respectively.Conclusions The result supports the suggestions that optimal cut-off score should be different according to target populations. The newly identified cut-off score of 17 would enable the identification of patients with EDs who have been previously classified as non-clinical samples in the EAT-26 test. Level of evidence: III: evidence obtained from case-control analytic study.Conclusions The result supports the suggestions that optimal cut-off score should be different according to target populations. The newly identified cut-off score of 17 would enable the identification of patients with EDs who have been previously classified as non-clinical samples in the EAT-26 test. Level of evidence: III: evidence obtained from case-control analytic study.