Trajectories of depressive symptoms, metabolic syndrome, inflammation, and cardiometabolic diseases: A longitudinal Bayesian network approach.

Abstract

Introduction: Both cardiometabolic diseases (CMD) and depression carry high burden of disease and have a striking bi-directional comorbidity. Understanding mechanisms of this comorbidity is key in improving health outcomes. Through Bayesian network analysis and quantitative centrality assessments we disentangled longitudinal associational pathways connecting depressive symptoms with immuno-metabolic dysregulations and CMD.

Methods: Data are from the Netherlands Study of Depression and Anxiety (NESDA), an ongoing longitudinal cohort study. Subjects (N = 1059, 68 % female, mean age 42.4 ± 12.5) had a lifetime depression diagnosis at baseline, and data at baseline, 2-, 6- and 9-year follow-up. Variables included depressive symptoms, metabolic syndrome components, inflammation, diabetes and atherosclerotic disease. Individual changes over time, determined using generalised mixed models, were fed into a Bayesian network model, resulting in a directed acyclic graph (DAG). For centrality evaluation, indegree and outdegree of variables (nodes) were assessed.

Results: The DAG showed a path starting with the depressive symptom low energy, leading to appetite/weight alterations and hypersomnia, ultimately leading to the nodes of diabetes and markers related to dyslipidaemia and inflammation. Waist circumference was the node with highest centrality. This result remained robust in sensitivity analyses.

Discussion: The findings traced a pathway linking specific energy-related depressive symptoms (e.g. low energy, appetite/weight oscillations and hypersomnia) to inflammation, dyslipidaemia and diabetes. Depressive symptoms and biological markers connected in this identified pathway may provide a valuable target to reduce cardiometabolic risk related to depression.