Initial Validation of the Cantonese Eating Disorder Assessment for DSM ‐5 in Hong Kong Adults With Binge Eating

OBJECTIVE: This study aimed to refine and culturally adapt the Eating Disorder Assessment for DSM-5 (EDA-5) into Cantonese and evaluate its diagnostic validity, feasibility, and acceptability in Hong Kong adults with recurrent binge eating. METHOD: Sixty-seven community-recruited Cantonese-speaking adults (91% female; mean age = 30.3 years) with recurrent binge eating in Hong Kong completed both the Cantonese EDA-5 interview and the clinician-administered SCID-5 interview Eating Disorder modules in a counterbalanced order. EDA-5 diagnoses were compared with clinician-assigned DSM-5-TR diagnoses established through the SCID-5 interviews using Cohen's kappa (κ), sensitivity, specificity, and predictive values. Administration time and feasibility/acceptability ratings were compared using Wilcoxon signed-rank tests. RESULTS: EDA-5 diagnoses matched the clinician-assigned DSM-5-TR diagnoses in 83.6% of cases. Agreement for any ED versus no diagnosis was substantial (κ = 0.82; sensitivity = 0.98; specificity = 0.81; accuracy = 0.93). Agreement was substantial for BN (κ=0.87) and BED (κ=0.81), whereas OSFED showed weaker agreement (κ = 0.36). The Cantonese EDA-5 required significantly less time than the SCID-5 (M = 10.6 vs. 26.0 min; p < 0.001), with comparable feasibility and acceptability ratings. DISCUSSION: The Cantonese EDA-5 demonstrated substantial concordance with clinician-assigned DSM-5-TR diagnoses established through SCID-5 interviews, high sensitivity for ED case identification, and substantially reduced administration time, while maintaining high participant acceptability. These findings provided preliminary support for the Cantonese EDA-5 as a feasible and acceptable DSM-5-TR-aligned diagnostic tool for eating disorder assessment in Cantonese-speaking populations.