ISBT Science Series (2016) 11, 125–131 ORIGINAL PAPER © 2016 International Society of Blood Transfusion Assessing the efficacy of a single-unit red blood cell transfusion policy at a multisite transfusion service using a computerized retrospective audit T. P. C. Covello, 1 J. G. Quinn, 2 A. Kumar-Misir, 3 S. Watson, 3 M. Almohammadi, 2 B. D. Crocker, 3 D. M. Conrad, 2 K. Tennankore, 4 I. Sadek, 2 E. Kahwash 2, * & C. K. Cheng 2,3 1 Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada 2 Division of Hematopathology, Department of Pathology and Laboratory Medicine, Dalhousie University, Halifax, NS, Canada 3 Pathology Informatics Group, Central Zone, Nova Scotia Health Authority, Halifax, NS, Canada 4 Division of Nephrology, Department of Medicine, Dalhousie University, Halifax, NS, Canadaa Background and Objectives In 2013, an Eastern Canada blood bank implemented a new policy to reduce red cell transfusion. The policy mandated clinical or labora- tory reassessment of stable, non-bleeding patients after each unit transfused, before issuing subsequent units. A computerized audit assessed the policy’s effectiveness. Materials and Methods This retrospective cohort study compares the policy’s effect on transfusion practice across three groups of adult inpatients: haematol- ogy, surgery and internal medicine. Compliance was inferred from increases in the proportion of single-unit red cell transfusions among all single- and double- unit transfusions. Outcome variables included transfusion intensity (red cell units administered per admission involving a transfusion) and pretransfusion haemo- globin levels. Results Each group had more transfusions issued as single units during the ten months following policy enforcement. In haematology patients, single-unit transfu- sions increased from 17% to 89% and transfusion intensity decreased (median: 2–2, Q1: 2–1, Q3: 6–4, P < 0Á001). Single-unit transfusions increased from 57% to 94% in medicine patients and from 63% to 87% in surgery patients. Transfusion inten- sity also decreased in surgical patients (median: 2–1, Q1: 1–1, Q3: 2–2, P < 0Á001) and in medicine patients (median: 2–1, Q1: 1–1, Q3: 2–2, P = 0Á008). No group showed a clinically significant change in pretransfusion haemoglobin levels. Conclusion The audit demonstrated significant compliance with a single-unit transfusion policy. Transfusion intensity decreased in all groups despite no clini- cally significant change in pretransfusion haemoglobin levels. Key words: computer audit, medical informatics, patient blood management, red blood cell transfusion, single-unit transfusion policy Introduction Unnecessary blood transfusion may contribute to blood shortages, increase costs and jeopardize patient safety. One potential strategy to reduce unnecessary red blood cell transfusion is a single-unit transfusion policy for stable patients who are not actively bleeding [1, 2]. Such a policy typically requires reassessment of the patient’s haemoglobin level and/or clinical status following each unit transfused to determine ongoing need for further red blood cell support. Historically, single-unit red cell trans- fusions have been avoided due to uncertainty of their benefit; however, much of the literature dissuading sin- gle-unit transfusions is based on older, methodologically limited studies involving extremely liberal transfusion Correspondence: Calvino K. Cheng, 205-5788 University Avenue, Halifax, NS, B3H 1V8, Canada. E-mail: calvino.cheng@nshealth.ca *Co-senior author. 125