Concordance Among 10 Different Anticholinergic Burden Scales in At-Risk Older Populations

The aim of the study was to evaluate the concordance among 10 anticholinergic scales for the measurement of anticholinergic drug exposure in at-risk elderly complex chronic patients in primary care. METHODS: An 8-month cross-sectional, multicenter study was carried out in a cohort of complex chronic patients older than 65 years in treatment with at least 1 drug with anticholinergic activity. Demographic, pharmacological, and clinical data were collected. Anticholinergic burden and risk were detected using the 10 scales included on the anticholinergic burden calculator (http://www.anticholinergicscales.es/). We used κ statistics to evaluated the concordance 2 to 2 (according to risk: high, medium, low or without risk) among the included scales. RESULTS: Four hundred seventy-three patients were recruited (60.3% female, median age of 84 years [interquartile range = 10]). Eighty was the total number of anticholinergic drugs with any scale (1197 prescriptions), with a median of 2 drugs with anticholinergic activity per patient (interquartile range = 2). The κ statistics comparing all the 10 scales ranged from −0.175 (Drug Burden Index versus Chew Scale) to 0.708 (Anticholinergic Activity Scale [AAS] versus Chew Scale). The best concordance was obtained between AAS and Chew Scale (κ = 0.708), followed by Clinician-Rated Anticholinergic Scale and Duran Scale (κ = 0.632) and AAS and Anticholinergic Cognitive Burden Scale (κ = 0.618), being considered substantial strengths of concordance. CONCLUSIONS: The agreement among the 10 scales in elderly patients with complex chronic conditions was highly variable. Great care should be taken when assessing anticholinergic drug exposure using existing scales because of the wide variability among them. The only scales that showed agreement were the AAS-Chew, Clinician-Rated Anticholinergic Scale–Duran, and AAS–Anticholinergic Cognitive Burden Scale pairs. In the rest of the cases, the scales are not interchangeable.