Background: The present study reports on the first translation and use of the Cornell Scale for Depression in Dementia (CSDD) (Alexopoulos, Abrams, Young, &amp
Shamoian, 1988) among poststroke patients (n = 101) in Japan. Objectives: The study had three main purposes: 1. To examine the factor structure of the CSDD among Japanese poststroke patients
2. To compare this with the factor structure identified for Anglo-American Alzheimer's (AD) patients
and 3. To examine the prevalence and covariates of depressive symptoms among the Japanese stroke survivors. Methods: Poststroke patients and their caregivers (n = 202), at a random sample of neurological hospitals in western Japan, were interviewed using the study instruments. Data was also collected from patient charts. All subjects at each site who met the study criteria participated in the study. Results: The four-factor solution for poststroke subjects was analogous to that found among AD patients with 2 main exceptions. In contrast to AD patients: 1. Physical complaints were unrelated to depressed mood in stroke patients
and 2. Agitation and psychosis loaded with depressed mood in stroke patients rather than as a separate unique factor as in AD patients. However, in the exploratory 5-factor model, agitation and suicidal ideation comprised a unique factor. Using standard cutoff scores for the CSDD, 58.2% of poststroke patients had scores suggesting possible depression. CSDD scores were not related to functional ability, or stroke characteristics such as aphasia or right or left-sided paralysis. However, scores were significantly higher among subjects ≤ 2 years poststroke. Feelings of irritability, anxiety, sadness, and sleep problems were most prevalent. Discussion: Despite the prevalence of depressive symptoms, none of the subjects were currently receiving any mental health treatment. Findings suggest that symptoms differ by poststroke duration, which may necessitate different treatment approaches. Copyright © 2002 John Wiley &amp
Sons, Ltd.