The prevalence of depression in Alzheimer's disease and vascular dementia in a population sample

Background: Previous studies have shown that depression is more prevalent i n vascular dementia than Alzheimer's disease (AD). Subjects for these studi es were either psychiatry or neurology patients, raising the issue of wheth er factors leading to treatment might have introduced sampling bias. Method s: Data for the present study came from the Canadian Study of Health and Ag ing (CSHA, 1994), a population-based prevalence study of dementia. AD was d iagnosed using NINCDS-ADRDA criteria (McKhann et al., 1984), vascular demen tia was diagnosed using draft ICD-IO criteria (World Health Organization, 1 987) and the Ischemic Scale (Hachinski et al., 1975), major depression was diagnosed using an algorithm based on DSM-III-R criteria (American Psychiat ric Association, 1987). The sample for the present study consisted of 481 s ubjects with AD and 140 with vascular dementia. Results: The weighted preva lence rate of major depression was 3.2% for AD and 21.2% for vascular demen tia, giving a crude odds ratio of 8.2 (95% Confidence Interval: 1.7-40.2). This finding was confirmed by a logistic regression analysis which adjusted for age, sex, place of residence (community, institution), self-reported h ealth, severity of cognitive impairment, and antidepressant or beta-blocker use. Limitations: Data on depressive symptoms were mon often missing in su bjects with dementia resulting in differential loss of potential study subj ects. Data on depressive symptoms were not sufficiently detailed to permit DSM-III-R criteria to be implemented rigorously. The method of diagnosing v ascular dementia was subject to misclassification. Conclusions: This study confirms in a population sample that depression is more prevalent in vascul ar dementia compared to AD. (C) 1999 Elsevier Science B.V. All rights reser ved.