This talk is concerned with an integration of analytic ideas from latent variable modeling and survival analysis (time-to-event, event-history or duration analysis) in order to examine aspects of the vascular depression hypothesis (VDH), which has been receiving increased attention in psychological gerontology, aging research, and geriatrics over the past 15 years or so (Alexopoulos, 1997). Discrete-time survival analysis provides some tentative evidence in favor of the VDH, using data from the ongoing Health and Retirement Study (HRS). The approach specifically addresses the unique predictive power of depression as a potential warning signal for an impending stroke, beyond widely used medical correlates in stroke studies. Suggested policy implications and related hypotheses to examine in future research are indicated in conclusion.
Read More