Antithrombotic prescribing in atrial fibrillation: application of a prescribing indicator and multidisciplinary feedback to improve prescribing

Background: atrial fibrillation is common in older people, and is associated with an increased risk of ischaemic stroke. Antithrombotic therapy reduces stroke-risk, but is known to be under-prescribed. Objectives: to use an evidence-based indicator to audit antithrombotic prescribing for older hospital inpatients with atrial fibrillation, and to assess whether feedback of audit results to hospital staff increases antithrombotic use. Design: cross-sectional notes-based audits, before and after feedback. Setting: six Aged Care and three General Medicine units at nine Australian public teaching hospitals between September 1998 and May 1999. Subjects: 1416 hospital inpatients aged 65 years and over (median age 81). Methods: medication charts were reviewed to identify patients prescribed digoxin or amiodarone. Presence of atrial fibrillation was confirmed by review of the patients' medical notes. To be considered appropriate, patients with atrial fibrillation had to be receiving either warfarin or aspirin (or both), or have documented contraindications to both agents. Feedback of audit results was provided to medical, pharmacy and nursing staff at multidisciplinary meetings. Changes in antithrombotic prescribing 4-8 weeks and 6 months after feedback were assessed. Prescribing 8 weeks prior to feedback was assessed retrospectively. Results: appropriateness of the decision to prescribe (or not prescribe) antithrombotic therapy increased from 81/112 (72%) immediately prior to feedback to 97/105 (92%) 4-8 weeks later (P<0.0001). Six months after feedback, appropriateness of prescribing declined slightly, to 85% (p=0.36). Over the 8 weeks prior to feedback, appropriateness of prescribing did not change (74% versus 77%, p=0.80). Increased aspirin prescribing accounted for most of the improvement in antithrombotic use after feedback, while warfarin continued to be under-used. Conclusions: antithrombotics were under-prescribed for older patients with atrial fibrillation. Audit and multidisciplinary feedback resulted in increased antithrombotic prescribing. The intervention had a greater impact on aspirin prescribing compared with warfarin.