Abstract
BACKGROUND: The prospective associations between metabolites and depression and anxiety risk are limited. Therefore, we conducted this cohort study to estimate the associations of depression and anxiety risk with metabolites, metabolic risk score (MRS), and the combinations and interactions of MRS with lifestyles.
METHODS: This study included 30,127 participants with plasma metabolites data and mental health assessments in the UK biobank. XG-Boost and LASSO-COX regression were used to select metabolites. Cox regression was used to construct MRS and estimate the associations between metabolites, MRS, the combinations and interactions of MRS with lifestyles and depression and anxiety risk. Restricted cubic splines were used to evaluate non-linear relationships.
RESULTS: During a median follow-up of 94.03 months, 1581 depression and 1807 anxiety were identified. Totally 148 and 42 metabolites were significantly associated with depression and anxiety risk, respectively. Intermediate and high MRS were significantly associated with increased depression risk (HR = 1.88[1.43-2.46] and 3.18[2.37-4.27], P < 0.001) and anxiety risk (HR = 1.34[1.03-1.73], P = 0.027 and 1.67[1.24-2.24], P < 0.001) with a non-linear pattern (Pnonlinear < 0.001). Intermediate and high MRS combined with lifestyle, BMI, sleep duration, physical activity, diet, and smoking were significantly associated with depression and anxiety risk. High MRS combined with favorable lifestyle was significant on the risk of depression and anxiety. Significant additive interaction was identified between MRS and lifestyles on depression risk.
CONCLUSIONS: Plasma metabolites and MRS were significantly associated with depression and anxiety risk. Maintaining normal metabolite levels is crucial for depression and anxiety prevention, even among individuals with healthy lifestyles.