The authors examined the vascular depression hypothesis by comparing the frequency of post-stroke depression and clinically-defined vascular depression and by examining the relationship between vascular burden and depression. Data from 670 geriatric rehabilitation patients were incorporated to compare the frequency of depression in three patient groups: 1) those with no evidence of vascular disease or stroke, 2) those with cerebrovascular risk factors (CVRFs) but no evidence of stroke, and 3) patients with stroke. They examined the unique relationship between CVRFs and depression by use of logistic-regression analysis. Although the frequency of depression was not significantly different between stroke (36.4%), CVRF (35.2%), and non-vascular patients (28.7%), there was a significant increase in the frequency of depression in patients without stroke as CVRF burden increased. This effect was not observed among stroke patients. CVRF burden predicted depression among patients without stroke even after controlling for general medical comorbidity, cognitive functioning, and ADL limitations. These findings provide empirical support for the vascular depression hypothesis and also indicate that the rates of clinically-defined vascular depression and post-stroke depression are similar in geriatric rehabilitation patients.