Abstract:Objective: To examine the relationships between post-stroke depression and cognition using network analysis. In particular, we identified central depressive symptoms, central cognitive performances, and bridge components that connect these 2 constructs. Design: An observational study. We applied network analysis to analyze baseline data to visualize and quantify the relationships between depression and cognition. Setting: Home and Community. Participants: 202 participants with mild-to-moderate stroke (N=202; mean age: 59.7 years; 55% men; 55% Whites; 90% ischemic stroke). Intervention: Not applicable. Main outcome measures: Patient Health Questionnaire (PHQ-8) for depressive symptoms and the NIH Toolbox Cognitive Battery for cognitive performances. Results: Depressive symptoms were positively intercorrelated with the network, with symptoms from similar domains clustered together. Mood (expected influence=1.58), concentration (expected influence=0.67), and guilt (expected influence=0.63) were the top 3 central depressive symptoms. Cognitive performances also showed similar network patterns, with executive function (expected influence=0.89), expressive language (expected influence=0.68), and processing speed (expected influence=0.48) identified as the top 3 central cognitive performances. Psychomotor functioning (bridge expected influence=2.49) and attention (bridge expected influence=1.10) were the components connecting depression and cognition. Conclusions: The central and bridge components identified in this study might serve as targets for interventions against these deficits. Future trials are needed to compare the effectiveness of interventions targeting the central and bridge components vs general interventions treating depression and cognitive impairment as a homogenous clinical syndrome.

Post-stroke Depressive Symptoms and Cognitive Performances: A Network Analysis