Measures of Anticholinergic Drug Exposure, Serum Anticholinergic Activity, and All-cause Postdischarge Mortality in Older Hospitalized Patients with Hip Fractures

Objectives: To assess possible associations between anticholinergic drug exposure and serum anticholinergic activity (SAA) and their capacities to predict all-cause mortality in older hospitalized patients. Setting: Academic medical center. Participants and Measurements: Data on clinical characteristics, full medication exposure, SAA, and 4 anticholinergic drug scoring systems (ADSSs: Anticholinergic Risk Scale [ARS], Anticholinergic Drug Scale, Anticholinergic Burden scale, and anticholinergic component of the Drug Burden Index) were collected in 71 older hospitalized patients (age 84 +/- 6 years) awaiting surgical repair after hip fractures. Results: The median (range) SAA was 2.8 (1.1-4.9) pmol/mL. Age (rho = 0.25, p = 0.03), Katz Index of Independence in Activities of Daily Living score (rho = 0.39, p = 0.001), in-hospital delirium (rho = 0.29, p = 0.01), preadmission cognitive impairment (rho = 0.31, p = 0.01), and the number of nonanticholinergic drugs (n-NA, rho = -0.27, p = 0.02) were associated with SAA. No significant associations were detected between ADSSs and SAA. Cognitive impairment (beta = 2.1, 95% confidence interval [CI]: 0.7 to 2.5, p = 0.005) and n-NA (beta = -0.3, 95% CI: -0.5 to -0.03, p = 0.03) were independently associated with SAA. Cognitive impairment (hazard ratio [HR]: 6.7, 95% CI: 1.1 to 40.3, p = 0.04) and higher ARS scores (HR: 2.2, 95% CI: 1.2 to 3.7, p = 0.006) independently predicted 3-month mortality whereas in-hospital delirium (HR: 3.6, 95% CI: 1.3 to 10.3, p = 0.02), living at home (HR: 0.2, 95% CI: 0.0 to 0.9, p = 0.03), and length of hospital stay (HR: 1.1, 95% CI: 1.0 to 1.2, p = 0.004) independently predicted 1-year mortality after adjustment for age, gender, and Charlson comorbidity index. Conclusions: Cognitive impairment and n-NA, but not ADSSs, are independently associated with SAA in older hospitalized patients. The ARS score, together with cognitive impairment, in-hospital delirium, place of residence, and length of hospital stay, predicts all-cause mortality in this group.