Threshold for a Reduction in Anticholinergic Burden to Decrease Behavioral and Psychological Symptoms of Dementia

Background: A high anticholinergic burden (AB) is associated with the occurrence of behavioral and psychological symptoms (BPSDs), which are frequent in dementia.

Objectives: Our aim was to determine the threshold for a reduction in AB that would lead to a clinically significant improvement in BPSDs (in terms of frequency, severity, and disruptiveness).

Design: A single-center prospective study.

Settings: Dedicated geriatric care unit specializing in the management of patients with dementia.

Participants: The study involved older patients with dementia, hospitalized for management of BPSDs.

Methods: One hundred forty-seven patients were included (mean age = 84.1 ± 5.2 years). The AB was assessed using 3 scales, namely, the Anticholinergic Drug Scale (ADS), the Anticholinergic Cognitive Burden scale (ACB), and the Anticholinergic Risk Scale (ARS). A clinically significant improvement in BPSDs was defined as a reduction of 4 points in the frequency × severity (F×S) score of the Neuropsychiatric Inventory-Nursing Home (NPI-NH) questionnaire. The threshold for a reduction in AB that corresponded to a clinically significant improvement in BPSDs was determined by multiple linear regression.

Results: One hundred forty-seven patients were included (mean age = 84.1 ± 5.2 years). Using the ADS, a reduction of 2 points in AB in patients with moderate-intensity BPSDs was associated with a clinically significant improvement in the F×S score of the NPI-NH [6.34, 95% confidence interval (CI) 4.54-8.14], and a reduction of 3 points was associated with a clinically significant improvement in the occupational disruptiveness score (4.26, 95% CI 3.11-5.41).

Conclusions/implications: In older patients with dementia presenting BPSDs, the risk-benefit ratio of anticholinergic drugs is debatable and, where possible, drugs with a lower AB would be preferable. Because BPSDs are a frequent cause of hospitalization, a standardized approach to analysis and reduction of AB is warranted in this population. Depending on the scale used to assess anticholinergic burden (AB), a small reduction in AB is associated with a decrease in frequency, severity, and disruptiveness of moderate-intensity BPSDs. Drugs with a high AB should be avoided where possible in older patients with dementia, and drugs with a lower AB would be preferable. Heterogeneity between the assessment scales for AB precludes generalization of the impact of a reduction in AB on BPSDs.