Study Objectives: Given the established racial disparities in both sleep health and dementia risk for African American populations, we assess cross-sectional and longitudinal associations of self-report sleep duration (SRSD) and daytime sleepiness with plasma amyloid beta (A beta) and cognition in an African American (AA) cohort.Methods: In a cognitively unimpaired sample drawn from the African Americans Fighting Alzheimer's in Midlife (AA-FAiM) study, data on SRSD, Epworth Sleepiness Scale, demographics, and cognitive performance were analyzed. A beta 40, A beta 42, and the A beta 42/40 ratio were quantified from plasma samples. Cross-sectional analyses explored associations between baseline predictors and outcome measures. Linear mixed-effect regression models estimated associations of SRSD and daytime sleepiness with plasma A beta and cognitive performance levels and change over time.Results: One hundred and forty-seven participants comprised the cross-sectional sample. Baseline age was 63.2 +/- 8.51 years. 69.6% self-identified as female. SRSD was 6.4 +/- 1.1 hours and 22.4% reported excessive daytime sleepiness. The longitudinal dataset included 57 participants. In fully adjusted models, neither SRSD nor daytime sleepiness is associated with cross-sectional or longitudinal A beta. Associations with level and trajectory of cognitive test performance varied by measure of sleep health.Conclusions: SRSD was below National Sleep Foundation recommendations and daytime sleepiness was prevalent in this cohort. In the absence of observed associations with plasma A beta, poorer self-reported sleep health broadly predicted poorer cognitive function but not accelerated decline. Future research is necessary to understand and address modifiable sleep mechanisms as they relate to cognitive aging in AA at disproportionate risk for dementia.Clinical Trial Information Not applicable.

Sleep abnormalities may play a prominent role in the pathogenesis or exacerbation of Alzheimer’s disease (AD). Sleep is considered a modifiable target for prevention and treatment. Racial sleep health disparities exist, and may contribute to the disproportionately increased risk for developing AD associated with being African American. Yet, the relationships between sleep health and AD pathophysiology in African American individuals are under-studied. This investigation was designed to address research gaps by evaluating the associations of self-report sleep duration, sleep disturbances, and daytime sleepiness with change in plasma-derived amyloid β 42/40 ratio (Aβ42/40) in an African American sample. Archival data from African American participants enrolled in the Wisconsin Registry for Alzheimer’s Prevention study and the Wisconsin Alzheimer’s Disease Research Center were utilized. Data included the Medical Outcomes Study Sleep Scale (MOS-Sleep), Epworth Sleepiness Scale (ESS), and demographic information. Sleep duration was determined by MOS-Sleep #2, sleep disturbances were captured by a related MOS-Sleep subscale (SLPD4), and ESS assessed daytime sleepiness. Participants provided plasma samples across multiple timepoints. PrecivityAD TM assays quantified Aβ42/40. Aβ42/40 rate of change (Aβ42/40_ROC) was computed for follow-up collections to serve as the primary outcome measure. Aβ42/40_ROC was computed as the difference in Aβ42/40 between follow-up and first collection divided by years between collections. Linear mixed-effects models determined associations with Aβ42/40_ROC. Observations were nested within participant. Fully adjusted models included age at collection, gender, BMI, education, parental history of dementia, and APOEe4 carrier status. Data included 339 follow-up collections from 143 participants. Women provided 70.8% of collections, with average age at collection = 66.6 ± 8.84 years, Aβ42/40_ROC = -0.0005 ± 0.003, and sleep duration = 6.35 ± 1.00 hours. Sleep duration significantly associated with Aβ42/40_ROC in the fully adjusted model (β = -0.0005; SE = 0.0002; p = 0.02) (Figure 1). Associations between SLPD4 and ESS with Aβ42/40_ROC were nonsignificant. Longer self-report sleep duration associated with a steeper reduction rate for plasma Aβ42/40, a biomarker for amyloid pathology. Further research is necessary to clarify the relationships between sleep health components and changes in AD-related biomarkers within African American adults to identify focal risk factors and interventional strategies.