Association of metabolic syndrome with depression in US adults: A nationwide cross-sectional study using propensity score-based analysis

Abstract

Background: The association of metabolic syndrome (MetS) with depression has been previously reported; however, the results are ambiguous due to imbalanced confounding factors. Propensity score-based analysis is of great significance to minimize the impact of confounders in observational studies. Thus, the current study aimed to clarify the influence of MetS on depression incidence in the U.S. adult population by using propensity score (PS)-based analysis.

Methods: Data from 11,956 adults aged 20-85 years from the National Health and Nutrition Examination Survey (NHANES) database between 2005 and 2018 were utilized. Using 1:1 PS matching (PSM), the present cross-sectional study included 4,194 participants with and without MetS. A multivariate logistic regression model and three PS-based methods were applied to assess the actual association between MetS and depression incidence. Stratified analyses and interactions were performed based on age, sex, race, and components of MetS.

Results: After PSM, the risk of developing depression in patients with MetS increased by 40% in the PS-adjusted model (OR = 1.40, 95% confidence interval [CI]: 1.202-1.619, P < 0.001), and we could still observe a positive association in the fully adjusted model (OR = 1.37, 95% CI: 1.172-1.596, P < 0.001). Regarding the count of MetS components, having four and five conditions significantly elevated the risk of depression both in the PS-adjusted model (OR = 1.78, 95% CI: 1.341-2.016, P < 0.001 vs. OR = 2.11, 95% CI: 1.626-2.699, P < 0.001) and in the fully adjusted model (OR = 1.56, 95 CI%: 1.264-1.933, P < 0.001 vs. OR = 1.90, 95% CI: 1.458-2.486, P < 0.001). In addition, an elevation in MetS component count was associated with a significant linear elevation in the mean score of PHQ-9 (F =2.8356, P < 0.001). In the sensitivity analysis, similar conclusions were reached for both the original and weighted cohorts. Further interaction analysis revealed a clear gender-based difference in the association between MetS and depression incidence.

Conclusion: MetS exhibited the greatest influence on depression incidence in US adults, supporting the necessity of early detection and treatment of depressive symptoms in patients with MetS (or its components), particularly in female cases.

Keywords: cross-sectional study; depression; metabolic syndrome; positive association; propensity score-based analysis.