Eating behavior and metabolic syndrome over time.

ABSTRACT

PURPOSE: We evaluated the longitudinal associations between eating behaviors (EB) and risk of metabolic syndrome (MetS). METHODS: We obtained complete data on EB, assessed using the Dutch Eating Behavior Questionnaire, and MetS components at baseline and follow-up. Participants included 1876 individuals (704 men, 1172 women; mean age, 45.0 ± 12.8 years) from those participating in the Korean Healthy Twin study. A generalized estimating equation model was applied, with sociodemographic factors, health-related factors, follow-up interval, and EB (baseline and changes over time) as independent factors. RESULTS: MetS at baseline was 21.5%, while incident MetS and persistent MetS were 12.0% and 66.6%, respectively, at the 3.13 ± 1.38 years follow-up period. In men, baseline restrained EB had positive associations with concurrent MetS (odds ratio [95% confidence interval] per 1 point increase in the score, 1.55 [1.33-1.81]) and persistent MetS (1.53 [1.16-2.01]); baseline external EB and change in external EB had positive associations with persistent MetS (1.56 [1.04-2.33], 1.37 [1.01-2.22], respectively). In women, baseline restrained EB had a positive association with concurrent MetS (1.14 [1.01-1.30]); baseline external EB had an inverse association with persistent MetS (0.71[0.52-0.98]); baseline emotional EB had positive associations with concurrent, incident, and persistent MetS (1.23 [1.01-1.50], 2.14 [1.50-3.06], and 1.92 [1.40-2.64], respectively); and change in emotional EB had positive associations with incident and persistent MetS (1.50 [1.05-2.15], 1.62 [1.14-2.29], respectively). CONCLUSION: Higher restrained and external EB, and an increase in external EB in men; and higher restrained and emotional EB, and an increase in emotional EB in women may be associated with increased risk of concurrent, incident, or persistent MetS. LEVEL OF EVIDENCE III, cohort study.