RESUMO
BACKGROUND: The purpose of this study was to search for a more effective transfusion-monitoring system than the existing system of retrospective peer review. STUDY DESIGN AND METHODS: This research used a study-control, preintervention and postintervention design, to evaluate the effectiveness of a prospective physician self-audit transfusion-monitoring system that functioned without the direct involvement of transfusion service physicians. This research also evaluated the effectiveness of issuing to physicians a memo with transfusion guidelines. Three process indicators were used to assess physician behavior at various stages of the blood-ordering process: 1) the number of crossmatches ordered per admission, 2) the transfusion-to-crossmatch ratio, and 3) the number of blood units returned to the laboratory after physician self-auditing. The study used two outcome indicators to reflect overall blood utilization: 1) the percentage of patients who received red cell transfusions and 2) the number of blood units transfused per recipient each month. RESULTS: The prospective physician self-audit system implemented at the study hospital did not reverse physician transfusion decisions, and the process of issuing to physicians a memo with transfusion guidelines at the control hospital failed to reduce blood usage. However, a transient reduction in blood utilization was observed at the study hospital. CONCLUSION: The reduction was hypothesized to be due to a Hawthorne effect, in which observed behavior is affected by the subject's awareness of the research study.
