Background and objectives: The Sakata's Eating Behavior Questionnaire (SEBQ) was developed to grasp distortions and habits in eating behavior. This study aimed to validate the factor structure and psychometrics properties of the Chinese version of the SEBQ, and examine the relationship between eating behavior and hypertension.
Methods and study design: Individuals were recruited from a community in Xuhui district in Shanghai. The 30-item SEBQ was used to measure eating behavior. Confirmatory factor analysis was performed to evaluate the dimensional structure, logistic regression analysis and mediating effect analysis were conducted to evaluate the relation between SEBQ with obesity and hypertension.
Results: Cronbach's alpha coefficient was 0.81 revealed good internal reliability. The values of root mean square error of approximation [90% confidence intervals (CI)], comparative fit index and Tucker-Lewis index were 0.053 (0.050, 0.056), 0.90 and 0.89 for the original version, and 0.062 (0.045, 0.079), 0.91 and 0.86 for the 7-item short form respectively. SEBQ can be used as a predictor of general obesity [odds ratio (OR): 5.14, 95%CI: 2.91-9.08], central obesity (OR: 2.87, 95%CI: 1.86-4.42) and abdominal obesity (OR: 2.59, 95%CI: 1.67-4.01). Its effect on hypertension was mediated by obesity (β=0.018, p=0.027), and the percentage of mediated effect by obesity between eating behavior and hypertension was 43.9%.
Conclusions: This study shown that SEBQ is a valid and reliable measurement tool to define obesity in Chinese as well and the associations between eating behavior and hypertension was mediated by obesity. Further validation among different populations is needed in the future study.