Abstract
BACKGROUND: The dose response relationship of nine-year cumulative anticholinergic exposure and dementia onset was investigated using the Korean version anticholinergic burden scale (KABS) in comparison with the Anticholinergic Cognitive Burden Scale (ACB). We also examined the effect of weak anticholinergics in the prediction of dementia. METHODS: A retrospective case-control study was conducted comprising 86,576 patients after 1:2 propensity score matching using the longitudinal national claims database. For cumulative anticholinergic burden estimation, average daily anticholinergic burden score during the 9 years prior to dementia onset was calculated using KABS and ACB and categorized as minimal, < 0.25; low, 0.25-1; intermediate, 1-2; and high, ≥ 2. Adjusted odds ratio (aOR) between cumulative anticholinergic burden and incident dementia was estimated. RESULTS: Patients with high exposure according to KABS and ACB comprised 3.2 and 3.4% of the dementia cohort and 2.1 and 2.8% of the non-dementia cohort, respectively. Dose-response relationships were observed between anticholinergic burden and incident dementia. After adjusting covariates, compared with minimal exposure, patients with high exposure according to KABS and ACB had a significantly higher risk for incident dementia with aOR of 1.71 (95% confidence interval (CI) 1.55-1.87) and 1.22 (CI 1.12-1.33), respectively. With the exclusion of weak anticholinergics, the association became stronger, i.e., 1.41 (CI 1.14-1.75) with ACB whereas the association became slightly weaker with KABS, i.e., 1.60 (CI 1.38-1.86). CONCLUSION: This study confirmed the dose response relationship for cumulative anticholinergic burden measured using the Korean specific anticholinergic burden scale with incident dementia.