We previously found adverse effects of anticholinergic (AC) and sedative medications on cognition in community-dwelling adults at risk for cognitive decline.  The present study extends this work in older patients seen for outpatient neuropsychological evaluation for a question of mild cognitive impairment (MCI) or dementia. We hypothesized that AC/sedative drug exposure would be associated with worse attention/processing speed (AP), executive functioning (EF), and memory.

This cross-sectional chart review assessed 392 patients (mean (M) age=72±7.7 years, range=54-91; 60.5% women; 93% White; education M=14±3 years; WRAT-4 Reading M=104.3±15.1). Patients self-reported current medications and medical comorbidities and were administered a comprehensive test battery. Medications were considered AC if they had a score ≥1 on the Anticholinergic Cognitive Burden Scale (ACB).  Sedative medications were categorized based on a previously published approach that utilizes US Medicare Part D drug formularies. Relationships between AC/sedative exposure and polypharmacy (≥5 daily medications) were elucidated with chi-square analyses. Demographically-adjusted composites were calculated for AP, EF, and memory using clinical norms. Bivariate Pearson correlations assessed relationships between medication exposure and cognitive domains. Multivariate linear regressions further evaluated significant medication-cognition associations, controlling for total medications, medical comorbidities, and estimated premorbid functioning (WRAT-4 reading).

Most patients (70%; n=275) used ≥1 sedative medication. Over half (63%; n= 248) used ≥1 AC drug; however, overall ACB scores were low, with 74% (n=292) of the sample having a total ACB score ≤2 (i.e., minimal anticholinergic burden). Polypharmacy was common (80%; n=314).  The vast majority (78%) of patients with polypharmacy used ≥1 sedative medication, compared with 38.5% of the non-polypharmacy group (χ2=46.7,p<.001). AC medication use was also more prevalent in the polypharmacy group (70.1% vs. 35.9%; χ2= 31.4, p<.001). Total sedative medications was negatively correlated with AP (r =-.134, p=.008) and EF (r=-.105, p=.04). Although number of AC medications was not correlated with AP or EF, ACB scores were negatively correlated with AP (r=-.106, p=.037). There were no significant associations with memory. Sedative medications and the a priori covariates significantly predicted AP (R2=.127, p<.001); more sedative medication use was uniquely associated with worse AP (β=-.426, p =.049).  An analogous model significantly predicted EF (R2=.069, p<.001), but the unique relationship with sedative medications attenuated to non-significance. A model with ACB scores and the same covariates significantly predicted AP (R2=.119, p <.001), but ACB was not a unique predictor.

AC/sedative medication exposure was prevalent in this clinical sample of older adults, most of whom dealt with polypharmacy. Though both drug classes had negative relationships with AP and EF, sedative medications had a particularly negative effect on AP. Whereas memory was not associated with sedative or AC medications, anticholinergic burden was low and may reflect referring prescribers’ awareness of AC risks in older adults. Regardless, AP and EF deficits may masquerade as memory problems. Though replication in a more diverse sample is needed, our findings suggest that sedative/AC medications are a modifiable risk factor for cognitive impairment in older adults with a question of MCI/dementia.