Differential cognitive and functional performance in racial and ethnic groups when stratified by consensus diagnosis: Findings from the NACC dataset
Giovanna Pilonieta, David S. Geldmacher- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Recent studies comparing non‐Hispanic Whites (NHW) to non‐Hispanic Blacks (NHB) with cognitive impairment found differences in associations between neuropsychological performance and ratings of daily function. Evaluating differences in these relationships by race and ethnicity may be important to interpreting disparities in clinical research outcomes for people with Alzheimer’s disease and related disorders.
Method
We compared demographic and performance variables across consensus diagnostic groups (CN, impaired‐not‐MCI, MCI, or dementia) at baseline in NHB (N = 1,609; 15.62%), Hispanic (N = 706, 6.86%) and NHW (N = 7,984,77.52%) adults in the NACC cohort, using data from Version 3.0 of the UDS, collected at the 37 NIA‐ADRCs from March 2015 to June 2022 (alz.washington.edu). Spearman’s rank correlation examined the associations between CDR‐GS, MoCA, and FAS scores by race and ethnicity within consensus diagnostic groups.
Result
Among 10,299 participants, 5,964 (57.91%) were women, mean age at baseline was 69.34 (±9.80), and mean education was 16.02 (±2.81) years. Significant differences in gender proportions between race and ethnic groups were observed. NHB had a higher proportion of female participants (χ2 = 233.39 <.0001); and reported the lowest FAS scores (Z = 200.67, P < .001). Hispanic participants had completed fewer years of education (Z = 3999.92, p = <.0001); and reported higher depressive symptoms (Z = 39.81, p = <.0001). Significant correlations were observed between all assessments by race and ethnic groups. When stratified by consensus group, in the NHW and NHB subsamples, CDR and MoCA scores associations were significant except in the “Impaired not MCI” groups. In the Hispanic subsample, significant correlations were only observed in the dementia group. After adjusting for the effects of GDS scores, correlations retained significance except in the NHB MCI group. Significant correlations between CDR and FAS scores were observed except in the “Impaired not MCI” group in the Hispanic subsample. These associations were retained after controlling for depressive symptoms except in the Hispanic MCI group.
Conclusion
Associations between cognitive and functional performance measures varied by racial and ethnic group when stratified by consensus groups in the NACC dataset. Specific ethnic, racial and socio‐cultural factors, as well as implicit/explicit examiner bias that may drive differences in performance warrant further study.