Exploring Physician-Related Transfusion Errors Reported to the Transfusion Error Surveillance System (TESS) from 2016-2023: A Single Centre Study

Introduction: Blood transfusions involve a multi-step process with an interdisciplinary team. Errors, defined as any deviation from established policies and standard operating procedures, can occur in any step of this process and may lead to adverse transfusion events. The web-based Transfusion Error Surveillance System (TESS) was implemented in Canada in 2005 and funded by the Public Health Agency of Canada to anonymously report and track error events. Previous studies have examined transfusion-related error types and frequency; however, an analysis of physician-related transfusion errors has not been performed. The purpose of this study was to describe physician-related transfusion errors in a tertiary care academic center, their harms and consequences, and how the COVID-19 pandemic may have influenced error trends.

Methods: All transfusion error events are investigated, detailed and reported through TESS by trained blood bank laboratory technologists. These errors are categorized as those performed by the clinical or transfusion service. This was a retrospective study of errors reported to TESS from 2016 to 2023, focusing on clinical service transfusion errors and errors where the primary individual involved was a physician. Clinical service errors were categorized as involving sample collection (SC), sample handling (SH), product request (PR), request for pickup (RP), or unit transfusion (UT). SC errors are events related to sample collection; SH errors occur during the collection process but do not involve the sample itself; PR errors involve the incorrect ordering of blood or blood products for transfusion; RP errors involve the request to pick up blood or blood products from the transfusion service; and UT errors occur outside of the transfusion service and involve the storage, selection and administration of blood or blood product. The consequence of an error is classified as harm if the error leads to an adverse event including a transfusion reaction, delayed transfusion or under-transfusion. For error rates, the following denominators were used: SC and SH errors used total number of specimens received; PR errors used total number of blood components (BC) and fractionated products (FP) requested; and RP and UT errors used total number of BC and FP issued. Error rates were reported per 1000 of their respective denominator. Descriptive statistics were used to determine overall error rate trends across the years, types and consequences of errors, and details of patient harm cases.

Results: Between January 2016 and December 2023, 29852 total errors were reported to TESS. 16155 (54%) were clinical service errors while 13697 (46%) were transfusion service errors. 2448 (8%) of the events reached the patient level, with 32 (0.1%) of these resulting in patient harm. SC had the highest error rate across all 8 years, ranging from 23 to 37 per 1000. In 2020, error rates for SC, SH, and PR increased. SC error rates decreased in 2021, while SH and PR decreased in 2022. Error rates for UT remained steady. RP errors increased in 2021 and remained consistently higher thereafter, increasing from 6 to 13 per 1000. While physician-related errors made up 9% (n=2672) of total errors, they accounted for 91% (29/32) of patient harm cases. Of the 29 patient harm cases, 27 (93%) were PR errors and 2 (7%) were UT errors. The consequences included 20 (69%) patients who had transfusion-associated circulatory overload (TACO), 3 (10%) who had febrile non-hemolytic transfusion reactions, 1 (3%) who had an allergic reaction, 1 (3%) who was hypotensive and 4 (14%) who had other adverse outcomes. Three patients died within 30 days of discovering the error; two deaths were not deemed attributable to the error.

Conclusion: Physician-related transfusion errors have serious consequences. Despite making up only 9% of total errors, physician-related transfusion errors are responsible for 91% of cases that resulted in patient harm. Preliminary analysis suggests that harmful transfusion events such as TACO, could have been prevented (ex. by ordering diuretics or decreasing the rate of transfusion to high-risk patients). Further exploration into the patient harm cases will be conducted to identify areas of improvement for physicians. Using the TESS database to examine physician-related transfusion errors highlights the potential harm and preventable nature of these errors and offers opportunities to improve patient care.

Vandermeulen:CSL Vifor: Consultancy. Lin:Canadian Blood Services: Research Funding; Octapharma: Research Funding; Pfizer: Honoraria. Gupta:University of Toronto: Research Funding; Canadian Blood Services: Research Funding; Rallybio: Research Funding.