Anticholinergic Burden and Cognitive Impairment in Nursing Homes: A Comparison of Four Anticholinergic Scales.

Abstract

Medications with anticholinergic effects are commonly used in nursing homes, and their cumulative effect is of particular concern for the risk of adverse effects on cognition. The relation between cognitive function and anticholinergic burden measured with four scales, the Anticholinergic Cognitive Burden (ACB) Scale, the Anticholinergic Risk Scale, the German Anticholinergic Burden Scale, and the CRIDECO Anticholinergic Load Scale, is assessed according to the hypothesis that a higher anticholinergic burden is associated with reduced cognitive performance. This retrospective cross-sectional multicenter study was conducted in a sample of Italian long-term-care nursing homes (NH). Sociodemographic details, diagnosis, and drug treatments of each NH resident were collected using medical records four times during 2018 and 2019. Cognitive status was rated with the Mini-Mental State Examination (MMSE). The prevalence of anticholinergic use and its burden were calculated referring to the last time point for each patient. A longitudinal analysis was done on NH residents with at least two MMSE between 2018 and 2019 to assess the relation between the anticholinergic load and decline in MMSE. The relationship between drug-related anticholinergic burden and cognitive performance was analyzed using Poisson regression model theory. Multivariate analyses were adjusted according to the known risk factors of reduced cognitive performance available [age, sex, history of stroke or transient ischemic attack (TIA), and number of non-anticholinergic drugs] and for cholinesterase inhibitors. In view of the high number of subjects with an MMSE score = 0 among residents with dementia, for this group a zero-inflated Poisson regression model was used to give more consistent results. The association of anticholinergic burden with mortality was examined from each patient's last visit using a multivariate logistic model adjusted for age, sex, and Charlson Comorbidity Index (CCI). Among 1412 residents recruited, a clear direct relationship was found between higher anticholinergic burden and cognitive impairment only for the Anticholinergic Cognitive Burden Scale. Residents taking an anticholinergic who scored 5 or more had 2.5 points more decline than those not taking them (p < 0.001). Among residents without dementia there was a trend toward direct relationship for the Anticholinergic Cognitive Burden Scale and the Anticholinergic Risk Scale. Residents with higher scores had about 2 points more decline than residents not taking anticholinergic drugs. No relation was found between anticholinergic burden and cognitive decline or mortality. The cumulative effect of medications with modest antimuscarinic activity may influence the cognitive performance of NH residents. The anticholinergic burden measured with the ACB scale should help identify NH residents who may benefit from reducing the anticholinergic burden. A clear direct relationship between anticholinergic burden and cognitive impairment was found only for the ACB Scale.