BACKGROUND: A higher anticholinergic risk score (ARS) is associated with an increased risk of anticholinergic adverse effects in elderly patients. It is unknown whether factors other than the use of anticholinergic drugs determine the ARS. METHODS: A comprehensive medical record review was conducted in 155 consecutive hospitalised patients (median age 79.0 years, interquartile range 66.0-86.0). Information was collected on: demographics; clinical characteristics (including medications and their doses); history of anticholinergic-induced adverse effects; and biochemical markers of hepatic and renal function (serum albumin concentrations and estimated glomerular filtration rate, eGFR). The ARS was calculated for each patient using a standard scoring approach and Poisson regression was used for identifying variables associated with the ARS. RESULTS: Patients with an ARS >or= 3 had a lower eGFR (p = 0.012) and were receiving more non-anticholinergic drugs (p <0.001) than patients with an ARS <3. In addition to being prescribed more anticholinergic drugs, patients with ARS >or= 3 were prescribed high doses of these drugs more often than patients with ARS <3 (41.3% vs. 26.9%, p = 0.034). A higher number of non-anticholinergic drugs (p <0.001), a lower serum albumin concentration (p = 0.014), and a lower eGFR (p = 0.012) were independently associated with a higher ARS. CONCLUSIONS: Polypharmacy, hypoalbuminaemia and low eGFR are independently associated with the ARS. Patients with a higher ARS are also prescribed higher doses of anticholinergic medications than those with lower ARS.