Leveraging technology to improve care: a real-time electronic dashboard within an anticoagulation stewardship service to optimise anticoagulant safety

Anticoagulants are high-risk medications associated with significant morbidity and mortality when used inappropriately. Electronic medical records, laboratory results, and medication data can be combined into electronic “dashboards”, providing a novel way to screen patients prescribed anticoagulants for potential errors. To describe and evaluate the impact of an efficient electronic dashboard within an Anticoagulation Stewardship service able to identify and facilitate proactive intervention for at-risk patients prescribed anticoagulants. Retrospective audit of adult inpatients receiving anticoagulants between 8th November 2022 and 31st January 2024 in a large multicentre tertiary hospital network. The dashboard was used daily to screen all hospital inpatients prescribed anticoagulants. Prescribing errors and bleeding risk factors were identified, and recommendations were made to treating teams. Cases were subclassified as “high-risk” or “standard-risk” based on the prescribing error and likelihood of harm. Combined clinically relevant non major bleeding (CRNMB) and major bleeding (MB) at 30 days post-intervention. Approximately 10,500 prescribing events were screened, 718 (6.9%) requiring a change to therapy in 611 patients. Of these patients, 292 (48%) were sub-classified as “high-risk”, and the remainder were “standard-risk” (n = 319, 52%). “High-risk” patients demonstrated higher rates of MB/CRNMB compared to those deemed “standard-risk” (7.2% vs. 3.1%, p = 0.022). This real-time dashboard enables rapid screening and identification of inpatients at risk of anticoagulant-related bleeding, allowing for timely intervention to reduce harm. A real time, electronic anticoagulation dashboard embedded within an Anticoagulation Stewardship (ACS) service enabled rapid screening of high patient numbers to identify and intervene on potential prescribing errors. Nearly half of cases requiring intervention had high-risk prescribing errors, and demonstrated significantly higher 30 day clinically relevant bleeding rates (7.2% vs 3.1%, p = 0.022). Timely ACS interventions prevented administration of inappropriate anticoagulant doses in ~30% of cases, with high clinician acceptance of advice. Future directions: Expansion of the dashboard to incorporate medication interaction checks and dynamic tracking of changes in key laboratory results could further enhance predictive accuracy and allow development of targeted alerting systems.