Depressive Symptomatology and Functional Status Among Stroke Survivors: A Network Analysis.

ABSTRACT

OBJECTIVE: To (1) characterize poststroke depressive symptom network and identify the symptoms most central to depression and (2) examine the symptoms that bridge depression and functional status. DESIGN: Secondary data analysis of the Stroke Recovery in Underserved Population database. Networks were estimated using regularized partial correlation models. Topology, network stability and accuracy, node centrality and predictability, and bridge statistics were investigated. SETTING: Eleven inpatient rehabilitation facilities across 9 states of the United States. PARTICIPANTS: Patients with stroke (N=1215) who received inpatient rehabilitation. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The Center for Epidemiologic Studies Depression Scale and FIM were administered at discharge from inpatient rehabilitation. RESULTS: Depressive symptoms were positively intercorrelated within the network, with stronger connections between symptoms within the same domain. "Sadness" (expected influence=1.94), "blues" (expected influence=1.14), and "depressed" (expected influence=0.97) were the most central depressive symptoms, whereas "talked less than normal" (bridge expected influence=-1.66) emerged as the bridge symptom between depression and functional status. Appetite (R2=0.23) and sleep disturbance (R2=0.28) were among the least predictable symptoms, whose variance was less likely explained by other symptoms in the network. CONCLUSIONS: Findings illustrate the potential of network analysis for discerning the complexity of poststroke depressive symptomology and its interplay with functional status, uncovering priority treatment targets and promoting more precise clinical practice. This study contributes to the need for expansion in the understanding of poststroke psychopathology and challenges clinicians to use targeted intervention strategies to address depression in stroke rehabilitation.