Christoph Buchta1, Wim Coucke2, Wolfgang R Mayr1,3, Mathias M Müller1, Günther F Körmöczi1,3. 1. ÖQUASTA, Austrian Association for Quality Assurance and Standardization of Medical and Diagnostic Tests, Vienna, Austria. 2. Sciensano, Quality of Laboratories, Brussels, Belgium. 3. Department of Blood Group Serology and Transfusion Medicine, Medical University of Vienna, Vienna, Austria.
Abstract
BACKGROUND: As some errors in pretransfusion testing remain unrecognized, error rates and the resulting need for corrective measures are probably underestimated. External quality assessment (EQA) schemes could provide valuable input for identifying error-prone laboratory tests because they are designed to monitor test performance and errors. So far, however, there are only limited published data on error rates in such schemes. METHODS: The types and incidence of incorrect results in an EQA scheme for red cell immunohematology with 187 participating laboratories were examined. The results of 58 distributions between 1999 and 2017 were evaluated, considering also the employed determination methods. RESULTS: Out of a total of 58,726 results, 563 (0.96%) were incorrect. Error rates were 5.45% for antibody identification, 1.39% for Rh phenotyping, 0.83% for serologic cross-match, 0.60% for direct antiglobulin test, 0.20% for Kell phenotyping, 0.16% for antibody screening, and 0.14% for ABO phenotyping. During the observation period, 53 participants reported error-free results, while 37 reported one incorrect result and 97 repeatedly reported incorrect results for one or more analytes. Error rates obtained by manual methods significantly surpassed those obtained by automated methods (1.04 vs. 0.42%). The introduction of double testing with two different systems reduced error rates in Rh phenotyping from 1.55 to 0.50%. CONCLUSION: Risk assessment should consider that error rates in pretransfusion test results vary. These data delineate the error risk potential of individual laboratory tests and thus should aid in tailoring appropriate improvement measures.
BACKGROUND: As some errors in pretransfusion testing remain unrecognized, error rates and the resulting need for corrective measures are probably underestimated. External quality assessment (EQA) schemes could provide valuable input for identifying error-prone laboratory tests because they are designed to monitor test performance and errors. So far, however, there are only limited published data on error rates in such schemes. METHODS: The types and incidence of incorrect results in an EQA scheme for red cell immunohematology with 187 participating laboratories were examined. The results of 58 distributions between 1999 and 2017 were evaluated, considering also the employed determination methods. RESULTS: Out of a total of 58,726 results, 563 (0.96%) were incorrect. Error rates were 5.45% for antibody identification, 1.39% for Rh phenotyping, 0.83% for serologic cross-match, 0.60% for direct antiglobulin test, 0.20% for Kell phenotyping, 0.16% for antibody screening, and 0.14% for ABO phenotyping. During the observation period, 53 participants reported error-free results, while 37 reported one incorrect result and 97 repeatedly reported incorrect results for one or more analytes. Error rates obtained by manual methods significantly surpassed those obtained by automated methods (1.04 vs. 0.42%). The introduction of double testing with two different systems reduced error rates in Rh phenotyping from 1.55 to 0.50%. CONCLUSION: Risk assessment should consider that error rates in pretransfusion test results vary. These data delineate the error risk potential of individual laboratory tests and thus should aid in tailoring appropriate improvement measures.
Authors: Christoph Buchta; Wim Coucke; Wolfgang R Mayr; Mathias M Müller; Reinhard Oeser; Christian R Schweiger; Günther F Körmöczi Journal: Clin Chem Lab Med Date: 2018-11-27 Impact factor: 3.694
Authors: Mario Plebani; Laura Sciacovelli; Mariela Marinova; Jessica Marcuccitti; Maria Laura Chiozza Journal: Clin Biochem Date: 2012-12-05 Impact factor: 3.281
Authors: P H Pinkerton; D E Wood; K L Burnie; K C Carstairs; B E Croucher; S C Ezer; D C Ley; J A McBride; M C Quantz; E D Zuber Journal: Clin Lab Haematol Date: 1981