Cerebrovascular disease and late-life depression: a latent-variable analysis of depressive symptoms after stroke.

OBJECTIVE: The author used a structural-equation modeling approach to examine depression symptom endorsement between geriatric stroke patients and general-medical geriatric patients. From the literature related to cerebrovascular disease and depression, he predicted that geriatric stroke patients would be more likely to endorse symptoms of withdrawal and affective flattening, and less agitation. METHODS: Data from 576 geriatric rehabilitation patients were utilized (N=176 with stroke). Confirmatory factor analysis of the Geriatric Depression Scale was conducted to determine the optimal factor structure for examining symptom endorsement differences between the two groups. MIMIC (Multiple Indicators, Multiple Causes) model analyses were incorporated to examine group differences in depression severity and symptom-cluster endorsement. RESULTS: Results indicate that stroke patients did not demonstrate greater depression severity. There were no differences in endorsement of depressive mood. However, stroke patients were more likely to endorse social withdrawal and less likely to endorse agitation symptoms. CONCLUSIONS: The current results suggest that geriatric stroke patients were more likely, after controlling for depression severity, to endorse greater social withdrawal and less agitation, but not depressed mood. These findings have implications for the types of symptoms evaluated in the context of geriatric stroke and depression.