Development and Implementation of Audit and Feedback for Patient Blood Management

BACKGROUND: Red blood cell (RBC) transfusions, a common perioperative procedure, are overused nationwide¹ despite their association with numerous adverse postoperative outcomes² and often unnecessary administration. Patient blood management (PBM) programs respond to these trends by promoting responsible and restrictive transfusion to reduce unnecessary transfusions and overuse. LOCAL PROBLEM: RBC transfusion is an overused procedure in cardiac surgery at a major academic medical center. Sources of overuse include the use of transfusions for avoidable indications and variability in transfusion practice between physicians within service lines. METHODS: Physician surveys and interviews were conducted to understand the current state of transfusion practice and identify metrics of meaning for a blood utilization audit and feedback system, a potential future component to a PBM program. Retrospective review of cardiac and noncardiac thoracic surgical cases were conducted to establish baseline RBC transfusion rates. Following the development and implementation of an audit and feedback system for cardiovascular and thoracic anesthesiology and surgery, analysis was conducted to detect any effect on the population at hand. INTERVENTIONS: The development of an audit and feedback system regularly reporting departmental and physician-specific trends in RBC transfusion practice aimed to inspire constructive self- evaluation and group discussion on areas of improvement. RESULTS: Anemia and RBC transfusion are highly prevalent in cardiac surgery and are associated with the increased incidence of adverse postoperative outcomes. Variability exists among service line physicians in terms of adherence to evidence-based restrictive transfusion guidelines. Physician feedback supports an audit and feedback system and strongly advocates for risk-adjusted peer comparisons and granular feedback regarding transfusion trends. Though the implementation of the audit and feedback system did not have a significant effect on various process and outcomes measures, it may be associated with an increase in single-unit transfusion orders. CONCLUSIONS: Data-driven audit and feedback, developed with physician collaboration and support, may be able to reduce avoidable RBC transfusions and improve perioperative transfusion practice by promoting thoughtful reflection and constructive conversation about current departmental trends and peer comparisons. However, such an effect may only be possible when the site at hand has enough capacity and infrastructure to support a widespread initiative.