Patient blood management data and metrics—two key components

Abstract

Effective and successful patient blood management (PBM) requires accessibility to accurate data and metrics. The use of computerized systems combining large amounts of donor data with electronic patient data, as well as big data and artificial intelligence (AI), has enhanced access to data and metrics. The application of AI and predictive models (PMs) in PBM offers significant potential for enhancing decision-making that will improve patient outcomes. Models include neural networks, support vector and gradient machines, retrieval-augmented generation (RAG), and random forest (RF). However, gaps in appropriate data access continue to remain an impediment. As healthcare systems have multiple electronic databases and sources of information, they pose challenges to the accurate reconciliation of data and the effective identification of actionable opportunities. PBM programs require a patient-centered, systematic, evidence-based approach to improve patient outcomes by managing and preserving a patient’s own blood, while promoting patient safety and empowerment. It is more than just optimal blood use (OBU) and programs should shift the focus from the blood product to the patient. Benchmarking ensures transparency concerning transfusion practices and enhances equity in blood product utilization, especially in vulnerable groups. Traditional metrics supporting effective PBM include inpatient red blood cell (RBC) transfusions per 1,000 patient days, transfusion thresholds, and rate of appropriate transfusion volumes (e.g., single unit RBC, single unit platelets, weight-based plasma volume). The importance of training and education to clinicians and advanced practice providers (APPs) cannot be overstated. This will serve to improve knowledge and adherence to transfusion guidelines and significantly reduce inappropriate transfusions. This manuscript will enhance the understanding of the role of data and metrics in a PBM program and best practices for designing and implementing it.