While Parkinson’s disease (PD) is traditionally known as a movement disorder, cognitive decline is one of the most debilitating and common non-motor symptoms. Cognitive profiles of individuals with PD are notably heterogeneous (Goldman et al., 2018). While this variability may arise from the disease itself, other factors might play a role. Greater anticholinergic medication use has been linked to worse cognition in those with PD (Fox et al., 2011, Shah et al., 2013). However, past studies on this topic had small sample sizes, limited ranges of disease duration, and only used cognitive screeners. Thus, this study aimed to examine this question within a large, clinical sample, using a more comprehensive neuropsychological battery. We hypothesized that higher anticholinergic medication usage would relate to worse cognitive performance, particularly memory.

Participants included 491 nondemented individuals with PD (m=64.7, SD=9.04 years old; education m=15.01, SD=2.79; 71.9% male; 94.3% non-Hispanics white) who underwent a comprehensive neuropsychological assessment at the UF Fixel Institute’s movement disorders program. Medications at the time of the neuropsychological evaluation were identified from chart review and scored based on anticholinergic properties using the Magellan Anticholinergic Risk Scale (Rudolph J.L., et al, 2008); each medication was scored from 0 (no load) to 3 (high load). The neuropsychological battery included measures across 5 cognitive domains: (1) executive function (Trails B, Stroop Interference, Letter Fluency), (2) verbal delayed memory (WMS-III Logical Memory and Hopkin’s Verbal Learning Test-Revised delayed recalls), (3) language (Boston Naming Test-II, Animal Fluency), (4) visuospatial skills (Judgment of Line Orientation, Face Recognition Test), and (5) attention/working memory (WAIS-III Digit Span Forward and Backward). The published normative scores for each task were converted into z-scores and averaged into a domain composite. Due to non-normality of Magellan scores, Spearman correlations examined the relationship between each cognitive domain composite score and Magellan scores.

As predicted, higher Magellan scores were significantly associated with worse memory (r=-0.11, p=0.016), with a small effect size. There were no significant relationships between Magellan scores and the remaining cognitive domains (EF, language, visuospatial, attention).

We found that greater anticholinergic burden was associated with worse performance on memory, but not other neuropsychological domains, in a large cohort of nondemented individuals with PD who underwent comprehensive assessment. This finding corresponds to previous literature in smaller PD cohorts. Though the effect size was low, this finding highlights the importance of monitoring anticholinergic burden in PD patients in order to minimize detrimental effects of medications on memory function. Future work should examine whether greater anticholinergic burden predicts future progression of memory decline.

Acknowledgement: Supported in part by the NIH, T32-NS082168