An international investigation into O red blood cell unit administration in hospitals: The GRoup O Utilization Patterns (GROUP) study

Michelle P. Zeller, Rebecca Barty, Astrid Aandahl, Torunn O. Apelseth, Jeannie Callum, Nancy M. Dunbar, Allahna Elahie, Henk Garritsen, Helen Hancock, José Mauro Kutner, Belinda Manukian, Shuichi Mizuta, Makoto Okuda, Monica B. Pagano, Ryszard Pogłód, Kylie Rushford, Kathleen Selleng, Claess Henning Sørensen, Ulrik Sprogøe, Julie StavesThorsten Weiland, Silvano Wendel, Erica M. Wood, Leo M G van de Watering, Maria van Wordragen-Vlaswinkel, Alyssa Ziman, Jaap Jan Zwaginga, Michael F. Murphy, Nancy M. Heddle, Mark H Yazer

Research output: Contribution to journalArticleResearchpeer-review

10 Citations (Scopus)


BACKGROUND: Transfusion of group O blood to non-O recipients, or transfusion of D- blood to D+ recipients, can result in shortages of group O or D- blood, respectively. This study investigated RBC utilization patterns at hospitals around the world and explored the context and policies that guide ABO blood group and D type selection practices. STUDY DESIGN AND METHODS: This was a retrospective study on transfusion data from the 2013 calendar year. This study included a survey component that asked about hospital RBC selection and transfusion practices and a data collection component where participants submitted information on RBC unit disposition including blood group and D type of unit and recipient. Units administered to recipients of unknown ABO or D group were excluded. RESULTS: Thirty-eight hospitals in 11 countries responded to the survey, 30 of which provided specific RBC unit disposition data. Overall, 11.1% (21,235/191,397) of group O units were transfused to non-O recipients; 22.6% (8777/38,911) of group O D- RBC units were transfused to O D+ recipients, and 43.2% (16,800/38,911) of group O D- RBC units were transfused to recipients that were not group O D-. Disposition of units and hospital transfusion policy varied within and across hospitals of different sizes, with transfusion of group O D- units to non-group O D- patients ranging from 0% to 33%. CONCLUSION: A significant proportion of group O and D- RBC units were transfused to compatible, nonidentical recipients, although the frequency of this practice varied across sites.

Original languageEnglish
Pages (from-to)2329-2337
Number of pages9
Issue number10
Publication statusPublished - Oct 2017
Externally publishedYes

Cite this