Global Vascular Risk Score and CAIDE Dementia Risk Score Predict Cognitive Function in the Northern Manhattan Study

Article type: Research Article

Authors: Rundek, Tatjana^{a; b; 1; *} | Gardener, Hannah^{a; b; 1} | Dias Saporta, Anita Seixas^{a; b} | Loewenstein, David A.^{a; b} | Duara, Ranjan^c | Wright, Clinton B.^d | Dong, Chuanhui^{a; b} | Levin, Bonnie^{a; b} | Elkind, Mitchell S.V.^e | Sacco, Ralph L.^{a; b}

Affiliations: [a] Department of Neurology, University of Miami Miller School of Medicine, Miami, FL, USA | [b] Evelyn F McKnight Brain Institute, University of Miami Miller School of Medicine, Miami, FL, USA | [c] Mount Sinai Medical Center, Miami, FL, USA | [d] National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA | [e] Department of Neurology, Columbia University College of Physicians and Surgeons, New York, NY, USA

Correspondence: [*] Correspondence to: Tatjana Rundek, Clinical Research Building, University of Miami Miller School of Medicine, 1120 NW 14th Street, Miami, FL, 33136, USA. Tel.: +1 305 243 7847; Fax: +1 305 243 7081; E-mail: trundek@med.miami.edu.

Note: [1] These authors contributed equally to this work.

Abstract: Background:Modifiable vascular risk factors (VRF) have been implicated in cognitive impairment. Objective:We compared the prediction of cognitive performance between the Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) dementia risk score, a validated tool to estimate dementia risk using VRF, and the Northern Manhattan Study (NOMAS) global vascular risk score (GVRS), created to predict vascular events. Methods:The CAIDE and GVRS scores were calculated based on baseline VRF among 1,290 stroke-free participants in the prospective population-based NOMAS MRI cohort (mean age 64±8 years, 60% women; 66% Hispanic, 17% Black, 15% White; 46% completed high school). Domain-specific Z-scores were derived for episodic and semantic memory, executive function, and processing speed, and averaged to calculate global cognition. Results:The CAIDE score was associated with worse global cognition at initial assessment (Beta per SD = –0.347, p < 0.0001), and with greater decline over time (Beta per SD = –0.033, p = 0.02). These associations were largely due to age and education, and the association with cognitive decline was not significant after adjusting for age, sex, and education. The GVRS was inversely associated with global cognition at initial testing (Beta per SD = –0.247, p < 0.0001) and greater decline over time (Beta per SD = –0.127, p < 0.0001), which persisted after adjusting for sociodemographics. The associations for both scores with initial cognitive performance were driven by executive function and processing speed, and the GVRS was associated with decline in episodic memory and processing speed. Conclusions:The GVRS was a stronger predictor of cognitive decline than the CAIDE in a multi-ethnic urban cohort. The inclusion of glucose and smoking in the GVRS, which are absent in CAIDE, likely explains the better performance of the GVRS.

Keywords: Cognition, dementia, epidemiology, risk score, vascular risk factors

DOI: 10.3233/JAD-190925

Journal: Journal of Alzheimer's Disease, vol. 73, no. 3, pp. 1221-1231, 2020