A common assumption in clinical neuropsychology is that cerebrovascular risk is adversely associated with executive function, while Alzheimer’s disease (AD) primarily targets episodic memory. The goal of the present study was to determine the cross-sectional and longitudinal validity of these assumptions using validated markers of cerebrovascular and AD burden.

19271 longitudinally-followed participants from the National Alzheimer Coordinating Center (NACC) database (Mean age= 72.25; SD age= 10.42; 58% women; 51.6% CDR= 0, 33.7% CDR= 0.5, 14.7% CDR> 1) were included. Cognitive outcomes were a composite memory score and an executive function composite score (UDS3-EF; Staffaroni et al., 2020). Baseline presence of cerebrovascular disease was indexed by the presence of moderate to severe white matter hyperintensities or lacunar infarct on brain MRI (yes/no), while baseline AD pathology was indexed by the presence of a positive amyloid PET scan or elevated CSF AD biomarkers (yes/no). We used linear mixed effect models to assess the effects of baseline cerebrovascular disease, baseline AD pathology, and their interactions with time in study (years post baseline) controlling for baseline age, sex, education, and baseline MoCA score.

Baseline cerebrovascular disease was significantly associated with a lower intercept on executive functioning (between-person effect) (p < -0.001, 95% CI -0.37, -0.14) but not memory, while presence of AD biomarkers was associated with a lower memory intercept (p < -0.001, 95% CI -0.52, -0.39) but not executive function. However, only presence of AD pathology at baseline was associated with faster longitudinal decline on both memory and executive functioning over time. Baseline cerebrovascular disease did not independently relate to rate of cognitive decline.

Consistent with widely held assumptions, our between-person analyses showed that MRI evidence of cerebrovascular disease was associated with worse executive functioning but not memory, while biomarker evidence of AD pathology was associated with worse memory but not executive function. Longitudinally, however, AD is the primary driver of decline in both executive and memory function. These results extend our understanding of how pathology impacts cognition in aging cohorts and highlight the importance of using longitudinal models.