Abstract
OBJECTIVE: Glucose metabolism is impacted by circadian disruption. Dinner-bedtime interval (DBI) was an accessible indicator to reflect the alignment between dinner time and circadian clock. We aimed to investigate the association of DBI with type 2 diabetes mellitus (T2DM). METHODS: 7676 adult subjects from the Henan Rural Cohort were included. Their demographic information including dinner time and bedtime was collected. Fasting venous blood samples were collected for biochemical determinations. Generalized linear regression model was used to analyze the factors influencing DBI. Furthermore, logistic regression incorporated with restricted cubic spline model was applied to evaluate the association between DBI and T2DM. RESULTS: The results of multiple linear regression model showed that age (β: -0.018, 95% CI: -0.021, -0.015) was negatively correlated with DBI. Female (β: 0.311, 95% CI: 0.229, 0.393), junior high school education (β: 0.246, 95% CI: 0.187, 0.306), high school education or above (β: 0.346, 95% CI: 0.259, 0.433), average monthly income with 1000-1999 CNY(0.102, 95% CI: 0.032, 0.171), average monthly income ≥ 2000 CNY (β: 0.164, 95% CI: 0.076, 0.251), moderate physical activity (β: 0.134, 95% CI: 0.071, 0.197), current smokers (β: 0.214, 95% CI: 0.118, 0.309), current drinkers (β: 0.099, 95% CI: 0.008, 0.190) were positively correlated with DBI. Furthermore, DBI was significantly associated with T2DM (adjusted OR: 0.910, 95%CI: 0.845-0.979, P = 0.012). DBI longer than 3 h was associated with decreased risk of T2DM (adjusted OR: 0.773, 95%CI: 0.648-0.921, P = 0.004). CONCLUSIONS: DBI larger than 3 h is beneficial to T2DM prevention. Further investigation is required to verify the association.