RESUMO
BACKGROUND: Adverse events in blood administration frequently involve the identification of transfusion recipients or components. This report details the results of an investigation of the efficacy of direct observation and that of a hospital-wide incident-reporting system in detecting standard operating procedures (SOPs) for deviations in blood administration. STUDY DESIGN AND METHODS: A process-driven audit form targeting 19 blood administration steps was developed for direct observation monitoring of blood administration. Over 18 months, 202 transfusions were observed in selected hospital locations. Data from this audit were compared with data collected from the incident reporting system. RESULTS: Through direct observation, 334 events were identified for a rate of 1.65 SOP deviations per transfusion. The incident reporting system identified 52 adverse events. Deviations were categorized as being related to the patient or component information, transfusion, patient monitoring, record documentation, and ordering or delivery of the component. Fifty-five percent of the events detected with direct observation related to identification of the patient or component, compared with 17 percent of incident reports. Using direct observation, 9 percent of transfused patients had wristband identification deviations. Such SOP deviations were not detected with the incident reporting system. Transfusion SOP deviations represented 15 percent of direct observation reports and 38 percent of incident reports. Direct observation identified deviations in monitoring practices and record documentation not detected by incident reporting. CONCLUSION: Direct observation appears to be an effective means for identifying deviations related to patient identification, patient monitoring, and record documentation.
