This study analyses the efficacy of two electronic tools to improve physicians’ adherence to transfusion guidelines and restrictive triggers for packed red blood cell transfusion in orthopaedic surgery. Over three 5-month periods, each tool was tested separately and then both tools were used in combination. During these 3 periods, 36, 31 and 25 patients, respectively, received 85, 59 and 60 red blood cell transfusions.
The first software tool, named decision support system (DSS), provides electronic help for the clinician’s decision to transfuse in a specific individual by matching the patient’s characteristics to a computerised knowledge base. The resulting recommendations did not improve guideline adherence during the first study period, when this tool was tested alone. The authors concluded that the lack of concurrent education and feedback to users who ignore alerts or bypassed the system may explain this negative result.
The second tool was the safe and controlled release of compatible blood (including electronic crossmatch) from a refrigerator remote from the laboratory to ease the access to blood when needed. This is meant to speed up the delay from blood order to transfusion and might decrease the number of ordered and transfused units. However, this tool failed to alter transfusion habits when tested alone.
During the first two periods, neither the number of transfused blood units nor the administration of double units was reduced. By a combination of both tools, however, compliance with transfusion guidelines, i.e. the percentage of patients with pre-transfusion haemoglobin levels below 8 g/dL, increased and the number of ordered and returned units decreased significantly.
It is difficult to find an explanation for the efficacy observed only by a combination of two electronic tools. However, this study demonstrates the need to improve blood use. In addition, it shows that electronic solutions for administration safety may exist but that their effect is poorly understood.
– Thomas Frietsch
Discuss this article on the Nataonline forum