Explaining Racial Discrepancies in Verbal Memory Assessment with Joint Estimation of Life Course Social Inequalities and Measurement Bias

Decades of research in cognitive test performance have demonstrated that racial group differences appear on a wide variety of tests and samples. The development of new tests, statistical adjustments, and incorporation of additional sociocultural information have contributed to varying degrees of success in the resolution of these differences. At the same time, racially minoritized communities are affected by higher rates of dementia, and yet require, on average, greater levels of severity in cognitive impairment to be diagnosed. To improve the detection of dementia at earlier stages where interventions are more likely to be useful and to facilitate better research into treatments where cognitive test performance is the standard outcome of interest, solutions to measurement differences between racial groups need to be identified. This study proposes a novel theoretically informed psychometric approach to separate cognitive test performance into factors related to life course disparities and those related to measurement bias. Utilizing the Consortium to Establish a Registry for Alzheimer's Disease list learning test administered as part of the national Harmonized Cognitive Assessment Protocol (HCAP) study conducted in partnership with the Health and Retirement Study, the pervasive bias model is used to identify sources of measurement differences, quantify the remaining impact of social inequities on performance differences, and evaluate the potential clinical applications of the approach. Utilizing a sample of 2074 cognitively normal adults over the age of 65 who identified as either racially White or Black, the pervasive bias model revealed that race had minimal direct contributions to measurement bias. In contrast, educational attainment was found to cause measurement bias on the list learning test, and since there were significant differences between the racial groups' years of education, this effect explained apparent racial group differences. Importantly, accounting for education's effects on the measurement of memory not only removed evidence for racial group differences but provided positive evidence overall that the two racial groups had equivalent overall memory abilities. While promising as a new approach to understanding test data, additional evaluation, particularly with clinical samples, is needed before the potential of this novel approach is truly understood.

Author dc:creator

• Goette, William Foster

Contributors dc:contributor

• Lacritz, Laura H.
• Cullum, C. Munro
• Rossetti, Heidi
• Hynan, Linda S.
• Berry, Jarett

• Cognitive Dysfunction
• Health Inequities
• Neuropsychological Tests
• Neuropsychology
• Racial Groups
• Cognition

Language dc:language

en

Identifier

1566445484

OAI identifier oai:identifier

oai:utswmed-ir.tdl.org:2152.5/10757