Quality Improvement Project to Reduce Anticholinergic Burden in Older Patients: Impact on Readmission, Delirium, Length of Stay

Abstract ID
3756
Authors' names
M Drelciuc1; R Chatterjee1; L Shakeshaft1; C Burns1; D Robson1; G Hollywood1; N Feeney1; C Cullen1.
Author's provenances
1. Acute Inpatient Frailty Unit, Royal Liverpool University Hospital
Abstract category
Abstract sub-category
Conditions

Abstract

Introduction: Anticholinergic medications are widely prescribed to manage pain, urinary incontinence, allergies. Patients with high frailty scores are more susceptible to anticholinergic adverse effects such as falls, cognitive impairment, urinary retention. The Anticholinergic Burden Score (ACB) is a tool used to quantify the cumulative anticholinergic effect of patients' medications. A score of 3 or more is associated with an increased risk of mortality and worse cognitive function. This quality improvement project aims to quantify and reduce ACB scores of patients admitted to the Acute Integrated Frailty Unit (AIFU) with a view to reduce hospital readmissions and overall mortality.

Methods: The medical team used the ACB score database to create an automated Excel calculator, which identifies each drug in a patient's medicine reconciliation list, assigns its database score, and calculates a cumulative ACB score. For cycle 1, we analysed ACB scores and patient data for 100 retrospective patients admitted to AIFU. Following cycle 1, we enlisted the help of frailty specialist nurses, who admitted 100+ patients to AIFU during in-reach ward rounds, used the calculator to obtain pre-admission ACB scores, and recorded them in patients' electronic notes. Following cycle 2, we created a poster for AIFU and presented the project at our local audit meeting as to provide structured education in ACB score calculation, medication alternatives for score reduction, and deprescribing strategies. Following the poster and presentation, for cycle 3, the medical and nursing team will collect data for another 100 patients admitted to AIFU and calculate three-month re-admission rates for our cycle 2 cohort. Our aim is the reduction of discharge ACB scores and of the overall 3-month readmission rate.

Results: In the retrospective data collection, 28% (28/100) of AIFU patients had an ACB score of 3 or more. At discharge, 27% (27/100) of these still had ACB scores of 3 or more. 33% of these patients had been re-admitted at 3 months after discharge. In cycle 2, 33.7% (27/80) of admitted AIFU patients had ACB scores of 3 or more. With the involvement of specialist frailty nurses and recording the ACB scores in patients' notes, only 30% (24/80) of these still had ACB scores of 3 or more at discharge.

Conclusion: The involvement of specialist frailty nurses and recording ACB scores in patients' notes saw a small improvement in ACB score reduction for AIFU inpatients at discharge. Going forward, we aim to quantify the effect of the educational presentation and poster on ACB scores, calculate 3-month re-admission rates for these patients, and identify the barriers to prescribing alternative low ACB score medications or deprescribing altogether. We aim to implement this practice on other geriatric medicine wards in the hospital and trust with a view to reduce readmissions in the clinically frail patient population.

Comments

This QIP provides strong evidence that frailty teams can successfully intervene to improve medication safety and potentially reduce adverse outcomes like falls, delirium and re-admission rates.

Submitted by drannjojy_46709 on

Permalink

Reducing anticholinergic burden is really important for patients, particularly amongst this cohort. Did you investigate the impact of this intervention on clinical outcomes, other than re-admission?

Submitted by rg3g22@soton.ac.uk on

Permalink

We did look at the link between admission ACB scores and markers such as length of stay and 4AT scores - with no significant association found. Therefore we did not explore the link between ACB reduction at discharge and length of stay and 4AT score. 

I think it would be useful to look at in future cycles especially if we decide to increase our cohort size for data collection. Especially so because of how our first cycles have already increased commitment to reviewing ACB scores regularly and documenting a decision on reducing/reasons for not reducing (which, as a side remark, is a great outcome of this QIP and our main aim initially). I think looking at individual clinical outcomes (which all feed into readmission to an extent) would be a good way to break things down to assess the clinical impact of our intervention before we look at how we influence readmission rates any further.