Abstract
BACKGROUND: Alzheimer's disease (AD) disproportionately affects Black/African American (B/AA) older adults, yet this group remains underrepresented in research. Traditional neuropsychological assessments, often developed on predominantly White populations, may not be reliable for B/AA individuals. The Loewenstein-Acevedo Scales for Semantic Interference and Learning (LASSI-L) have been shown to effectively differentiate individuals with amnestic mild cognitive impairment (aMCI) from cognitively unimpaired (CU) individuals. This study examines the relationship between LASSI-L performance and plasma p-tau(217) levels to explore early detection methods for AD in B/AA populations. METHODS: Fifty-six older adults received clinical and cognitive evaluations and were deemed cognitively unimpaired (CU) and p-tau(217) negative (n = 35) or met criteria for amnestic mild cognitive impairment (aMCI) and p-tau(217) positive (n = 21). All participants were administered the LASSI-L to compare groups, but it was not used for group allocation to avoid circularity. RESULTS: After adjusting for age and MMSE score, the aMCI p-tau(217)+ group performed significantly worse than the CU p-tau(217)- group on both free recall on List B (Free B1 Recall) and frPSI (correct responses on Cued B2). These differences remained statistically significant after covariate adjustment (p < 0.001). In addition, four other outcomes remained statistically significant following covariate adjustment: the aMCI p-tau(217)+ group exhibited a higher percentage of intrusion errors (PIE) on both Cued B1 and Cued B2, along with poorer performance on maximal learning ability (Cued A2) and PSI (correct responses on Cued B1). However, after applying the Bonferroni correction, only PIE on Cued B2 remained statistically significant among these measures. Notably, performance on LASSI-L Free B1 Recall and PIE for List Cued B2 were significant predictors distinguishing aMCI p-tau(217)+ from CU p-tau(217)- groups, with high sensitivity (80%) and specificity (91.7%). ROC analysis of these predictors yielded an area under the curve of 0.872 (SE = 0.055; p < 0.001), with a 95% confidence interval ranging from 0.765 to 0.979. CONCLUSION: The study highlights the utility of the LASSI-L in conjunction with plasma biomarkers, particularly p-tau(217), for early AD detection in B/AA older adults. The LASSI-L demonstrated strong sensitivity to cognitive impairment, effectively differentiating between CU and aMCI groups based on plasma p-tau(217) levels. These findings suggest that combining cognitive assessments with plasma biomarkers can enhance early diagnosis and improve timely interventions, addressing health disparities in AD diagnosis and care.