(P-TS-4) A Resource-Conscious Solution for Enhancing Transfusion Reaction Detection: Development of an Automated EMR Alert

Transfusion services are required to establish and adhere to policies and procedures for monitoring and identification of transfusion reactions.  According to the 2021 National Blood Collection and Utilization Survey, the overall incidence of adverse reactions to transfusion was 0.27%.  Perceived failure to meet requirements can result in citations from regulatory bodies and possible discontinuation of transfusion services if not addressed.   Many transfusion services rely on a passive surveillance system, which utilizes bedside nurses to recognize and report transfusion reactions.  An active surveillance program--in which a person(s) or computerized surveillance system actively seeks out potential reactions­­--requires resources, time, and personnel that many transfusion services cannot provide.  In a resource-limited setting, we developed a Best Practice Alert (BPA) to better support bedside nurses in order to adhere to the standards and more effectively surveil transfusions and identify reactions.

Study

Design/Methods:

An automated electronic medical record (EMR) BPA was implemented to notify the bedside nurse when a change in the temperature, oxygen saturation, heart rate, or blood pressure might indicate an adverse transfusion event.  An alert is triggered when there is an increase in temperature of 1.0°C or 2.0°F and above 38°C or 100.3°F, oxygen saturation of 90% or less on room air, a change in heart rate of 10 BPM, and a change in blood pressure of 30 mmHg from the most recent documented pre-transfusion vital signs. The alert gives the bedside nurse the option to cancel the BPA, accept the BPA (no action required), or activate transfusion reaction work up.

Results/Findings:

At our institution, the overall reported incidence of transfusion reaction in the year (2021) prior to implementation was 0.31%.  In the year following implementation of the new BPA (2023), the overall reported incidence was 0.40%.  Although, not yet statistically significant (p = 0.0786) the implementation did improve reporting and nursing awareness as well as increase discussion and education with bedside providers. Preliminary data suggested that most alerts were cancelled, leading to review of parameter effectiveness.

Conclusions: The automated BPA improves reporting and more effectively monitors and identifies potential transfusion reactions compared to a passive surveillance system. Although the difference in reported reactions was not statistically significant, the system prompts nurses to consider transfusion reaction workups that may have been underrecognized and underreported prior to its implementation.