Cognition and the Predictive Utility of Three Risk Scores in an Ethnically Diverse Sample

Background: Various factors, such as age, cardiovascular concerns, and lifestyle patterns, are associated with risk for cognitive decline and Alzheimer's disease (AD). Risk scores model predictive risk of developing a disease (e.g., dementia, stroke). Many of these scores have been primarily developed in largely non-Hispanic/Latino (non-H/L) White samples and little is known about their applicability in ethno-racially diverse populations. Objective: The primary aim was to examine the relationship between three established risk scores and cognitive performance. These relationships were compared across ethnic groups. Methods: We conducted a cross-sectional study with a multi-ethnic, rural-dwelling group of participants (Mage = 61.6±12.6 years, range: 40-96 years; 373F:168M; 39.7% H/L). The Cardiovascular Risk Factors, Aging and Dementia (CAIDE), Framingham Risk Score (FRS), and Washington Heights-Inwood Columbia Aging Project (WHICAP) score were calculated for each participant. Results: All three scores were significantly associated with cognition in both H/L and non-H/L groups. In H/Ls, cognition was predicted by FRS: β= -0.08, p = 0.022; CAIDE: β= -0.08, p < 0.001; and WHICAP: β= -0.04, p < 0.001. In non-H/Ls, cognition was predicted by FRS: β= -0.11, p < 0.001; CAIDE: β= -0.14, p < 0.001; and WHICAP: β= -0.08, p < 0.001. The strength of this relationship differed between groups for FRS [t(246) = -4.61, p < 0.001] and CAIDE [t(420) = -3.20, p = 0.001], but not for WHICAP [t(384) = -1.03, p = 0.30], which already includes ethnicity in its calculation. Conclusion: These findings support the utility of these three risk scores in predicting cognition while underscoring the need to account for ethnicity. Moreover, our results highlight the importance of cardiovascular and other demographic factors in predicting cognitive outcomes.