Cognitive differences across ethnoracial category, socioeconomic status across the Alzheimer's disease spectrum: Can an ability discrepancy score level the playing field?

An ability discrepancy (crystallized minus fluid abilities) might be a personally relevant cognitive marker of risk for Alzheimer's disease (AD) and might help reduce measurement bias often present in traditional measures of cognition. In a large national sample of adults aged 60-104 years (N = 14,257), we investigated whether the intersectionality of group characteristics previously shown to pose a risk for AD including ethnoracial category, socioeconomic status, and sex (a) differed in ability discrepancy compared to traditional neuropsychological tests and (b) moderated the relationship between an ability discrepancy and AD symptom severity. In cognitively normal older adults, results indicated that across each decade, fluid and memory composite scores generally exhibited large group differences with sex, education, and ethnoracial category. In contrast, the ability discrepancy score showed much smaller group differences, thus removing much of the biases inherent in the tests. Women with higher education differed in discrepancy performance from other groups, suggesting a subgroup in which this score might reduce bias to a lesser extent. Importantly, a greater ability discrepancy was associated with greater AD symptom severity across the AD continuum. Subgroup analyses suggest that this relationship holds for all groups except for some subgroups of Hispanic Americans. These findings suggest that an ability discrepancy measure might be a better indicator of baseline cognition than traditional measures that show more egregious measurement bias across diverse groups of people.

Linear and nonlinear relationships between cognitive subdomains of ability discrepancy and Alzheimer's disease biomarkers.

OBJECTIVE: Substantial research indicates that fluid and crystallized abilities are highly correlated throughout the adult life span. However, recent proposals suggest that a large discrepancy between these two abilities, defined as crystallized performance minus fluid performance, indicates heightened risk for Alzheimer's disease (AD). METHOD: In 266 cognitively healthy older adults, the present study tested linear and quadratic relationships between an ability discrepancy score and early AD neuropathology indexed via in vivo measures of beta-amyloid deposition and cortical thickness in AD-vulnerable regions. We also tested the extent that alternative forms of this ability discrepancy measure (e.g., subdomain discrepancies, verbal-visual discrepancies) and an episodic memory composite might also be sensitive markers of early AD pathology. RESULTS: An overall ability discrepancy was linearly and positively correlated with beta- amyloid. A quadratic relationship was found between the overall ability discrepancy score and cortical thickness such that a small positive correlation was found at lower discrepancy levels (fluid > crystallized), but at higher discrepancy levels (crystallized > fluid) a negative relationship was found (i.e., an inverted-U pattern). Similar patterns were found across each subdomain of cognition, but the effects were weaker than the overall ability discrepancy score. Importantly, inclusion of episodic memory (the gold standard) did not alter any of the effects, suggesting that an ability discrepancy confers unique predictiveness of AD biomarkers. CONCLUSIONS: These findings replicate previous findings and increase the confidence in their usefulness to predict AD biomarkers. Longitudinal validation is needed to clearly relate an ability discrepancy to specific stages of preclinical AD. (PsycINFO Database Record (c) 2020 APA, all rights reserved).