Reliability and validity of a Korean version of the children's eating behavior questionnaire in anorexia context

Although childhood anorexia is a common clinical disorder, there is no established tool for evaluating it. The Children's Eating Behavior Questionnaire (CEBQ) is a parent-reported measure designed to assess the eating behavior of children. We aimed to investigate the reliability and validity of the Korean version of the CEBQ (K-CEBQ) for children with anorexia.

Parents of children with anorexia aged between 2 and 9 years participated in a survey conducted twice at 1-month intervals. The general characteristics and K-CEBQ scores of the children were recorded. To assess the reliability of the K-CEBQ, the internal consistency and test-retest methods were used. Furthermore, correlation analysis was performed for each item and factor, and the discriminant validity was determined through comparison with the normal group. Optimal cut-off scores, which are based on the maximum area under the curve of the receiver operating characteristic curve, were calculated in two categories.

A total of 336 participants responded to the first survey, and the responses from 171 participants from the second survey were included in the analysis. The K-CEBQ showed relatively high internal consistency reliability (Cronbach's alpha = 0.738), and the retest demonstrated sufficient temporal stability. The relationship between each item of the K-CEBQ and the factor to which the item belongs showed a high correlation. There were significant differences between the anorexia and normal groups in two categories of the K-CEBQ: 'food approach' (p = 0.0063) and 'food avoidant' (p < 0.0001). The optimal cut-off values for the 'food approach' and 'food avoidant' category scores were 39.50 and 58.50 points, respectively.

This study demonstrated that the K-CEBQ may be a reliable and valid tool for assessing the eating behavior of children with anorexia.