The Chinese Eating Disorder Brief Screening Scale (CEDBS): Development and Psychometric Validation in General and Clinical Populations

AI Summary

CEDBS developed and refined through expert evaluation, cognitive interviews, pilot study, and validated in large nonclinical and clinical Chinese samples.
Ten items form a unidimensional scale with high internal consistency (McDonald’s ω = 0.95) and acceptable test retest reliability (ICC = 0.81).
High screening accuracy, AUC = 0.88, sensitivity 0.75 and specificity 0.84; practical integer cutoff recommended at 17 to identify probable ED cases.

Int J Eat Disord. 2026 May 26. doi: 10.1002/eat.70130. Online ahead of print.

ABSTRACT

OBJECTIVE: Despite growing recognition of eating disorders (EDs) as a public health concern in China, validated instruments suitable for efficient large-scale screening remain limited. This study aimed to develop and validate the Chinese Eating Disorder Brief Screening Scale (CEDBS).

METHOD: Scale development and validation were structured across two studies. Study 1 involved item generation and refinement through expert evaluation, cognitive interviews, and a pilot study (N = 146). Study 2 examined psychometric properties in large non-clinical (N = 3573) and clinical (N = 325) samples. Analyses included exploratory and confirmatory factor analysis for construct validity, McDonald’s ω and intraclass correlation coefficients (ICC) for reliability, Pearson correlations for convergent and discriminant validity, and receiver operating characteristic (ROC) analysis for screening accuracy.

RESULTS: The final 10-item CEDBS demonstrated a unidimensional structure, high internal consistency (ω = 0.95), and acceptable test-retest reliability (ICC = 0.81). CEDBS scores correlated strongly with established ED-specific measures (eating disorder inventory subscales: rs = 0.51-0.87; eating disorder examination-questionnaire global score: r = 0.74; eating attitudes test-26: r = 0.81) and showed a moderate correlation with the anxiety scale (r = 0.35), supporting discriminant validity. ROC analysis yielded an area under the curve of 0.88, with an optimal cutoff of 16.5 (sensitivity = 0.75, specificity = 0.84).

DISCUSSION: The CEDBS demonstrates preliminary psychometric support as a brief screening tool for identifying probable ED cases in Chinese populations. An integer cutoff score of 17 may be considered for practical use.

PMID:42192220 | DOI:10.1002/eat.70130

Document this CPD