Transfusion, the most common inpatient procedure, carries risk of adverse events resulting in significant morbidity and mortality. Studies note that the passive reporting system in the United States underreport transfusion reaction rates and only produces population based data. Digitally enabled real time hemovigilance by a team of transfusion specialists allows surveillance for reactions during administration of blood products in our institution. This strategy improves transfusion reaction identification and reporting; it also provides rapid intervention for patients as well as robust data on transfusion reactions.
Our institution transfuses over 192,000 blood components annually. We built a digital health platform that extracts data from the electronic health record (e.g. increase in temperature, drop in blood pressure) and applies a weighted risk score to each patient receiving transfusion based on an algorithm utilizing the Centers for Disease Control hemovigilance protocol. The risk score prioritizes patients for real time chart reviews by transfusion specialist nurses. If a reaction is suspected, the team deploys an advanced practice provider (e.g. nurse practitioner) to evaluate and manage the patient. All touchpoints in this system are captured as discrete data fields for analysis. Using these data, we are building an artificial intelligence powered human/machine hybrid to aid in detecting transfusion reactions that could be implemented in other organizations. Our analyses currently include over 50,000 transfusions. Operationalizing this clinical program was achieved through collaboration between laboratory and nursing.
A patient based surveillance program can improve care beyond transfusion reactions if applied to sepsis early identification with rapid response or patient deterioration management. Our strategy has allowed a comprehensive and catalogued review of the transfusion process in our institution, which allows insight into transfusion practice.
Differentiate the difference between population based and patient based hemovigilance practices
Discuss the advantage that collaboration with nursing brings to laboratory initiated patient surveillance programs and outline the process of educating and establishing transfusion specialized nurses and nurse practitioners
Discuss the critical role that digital health plays in accomplishing patient based surveillance
Describe how patient specific transfusion reaction risk score based on patient specific data and CDC transfusion reaction definitions is computed and utilized in real time
Gather awareness on findings revealed from curation and analysis of over 50,000 transfusions