US-based studies have reported that older blacks perform worse than older whites on cognitive tests and have higher risk of Alzheimer disease dementia (AD). It is unclear whether these findings reflect differences in cognitive decline.
The Chicago Health and Aging Project followed individuals, 65+ years old (64% black, 36% white), for up to 18 years. Participants underwent triennial cognitive assessments; stratified randomized samples underwent assessments for AD. We compared black and white participants’ cognitive performance, cognitive decline rate (N = 7,735), and AD incidence (N = 2,144), adjusting for age and sex.
Black participants performed worse than white participants on the cognitive tests; 441 participants developed AD. Black participants’ incident AD risk was twice that of whites (RR = 1.9; 95% CI, 1.4, 2.7), with 58 excess cases/1,000 occurring among blacks (95% CI, 28, 88). Among noncarriers of APOE ε4, blacks had 2.3 times the AD risk (95% CI, 1.5, 3.6), but among carriers, race was not associated with risk (RR = 1.1; 95% CI, 0.6, 2.0; P interaction = 0.05). However, cognitive decline was not faster among blacks: the black-white difference in 5-year change in global cognitive score was 0.007 standard unit (95% CI, −0.034, 0.047). Years of education accounted for a sizable portion of racial disparities in cognitive level and AD risk, in analyses using a counterfactual approach.
The higher risk of AD among blacks may stem from lower level of cognitive test performance persisting throughout the observation period rather than faster rate of late-life cognitive decline. Disparities in educational attainment may contribute to these performance disparities. See video abstract at, http://links.lww.com/EDE/B299.
From the aDepartment of Epidemiology, Boston University School of Public Health, Boston, MA; bRush Alzheimer’s Disease Center, Department of Neurological Sciences, Department of Behavioral Sciences, Rush University Medical Center, Chicago, IL; cDepartment of Epidemiology, University of Michigan School of Public Health, Ann Arbor, MI; and dDepartment of Internal Medicine, Rush Institute for Healthy Aging, Rush University Medical Center, Chicago, IL.
Submitted September 22, 2015; accepted August 28, 2017.
Data and statistical code sharing: The statistical computing code and data used to generate the findings are available on request. The text directly references the more novel components of the statistical code.
This investigation was supported by National Institutes of Health grant AG11101 (PI: Evans).
Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com).
Correspondence: Jennifer Weuve, Department of Epidemiology, Boston University School of Public Health, 715 Albany Street, T331E, Boston, MA 02118. E-mail: firstname.lastname@example.org.