Differential prediction of vascular dementia and Alzheimer's disease in nondemented older adults within 5 years of initial testing

Abstract

To determine whether neuropsychological tests and the Hachinski Ischemic Score (HIS) can differentiate incident vascular dementia (VaD) from Alzheimer's disease (AD) in nondemented older adults within 5 years of initial testing. The Canadian Study of Health and Aging (CSHA) included three waves: CSHA-1 (1991-1992), CSHA-2 (1996-1997), and CSHA-3 (2001-2002). This analysis included participants of the CSHA who (a) underwent neuropsychological testing and clinical assessment at CSHA-2 and were determined to be nondemented, and (b) underwent diagnostic assessment at CSHA-3. The outcome measure was CSHA-3 diagnosis, classified as VaD (n = 22), probable or possible AD (n = 65), and all other diagnostic outcomes (n = 417). CSHA-3 diagnosticians were blinded to CSHA-2 test scores and diagnoses. Multinomial logistic regression with forward selection was used to determine the ability of the HIS and 15 CSHA-2 neuropsychological tests to predict CSHA-3 diagnostic outcome. The analysis was repeated after removing 15 AD cases with coexisting vascular disease. The HIS and four neuropsychological tests were significant predictors of CSHA-3 diagnostic outcome (χ2 (14) = 149.59, P < .001, R2 = 0.38). Relative to developing VaD, higher HIS (odds ratio [OR]: 0.70; 95% confidence interval [CI]: 0.57-0.86) and Rey Auditory Verbal Learning Test immediate verbal recall scores (OR: 0.77; 95% CI: 0.62-0.97) were associated with lowered odds of developing AD, whereas higher phonemic fluency scores (OR: 1.21; 95% CI: 1.02-1.17) were associated with increased odds of developing AD. Removing AD cases with vascular disease did not affect results. In an epidemiological sample of nondemented participants, the HIS and two neuropsychological tests contributed to the differential prediction of VaD and AD within 5 years of initial measurement.