Assessing inpatient platelet ordering practice: evaluation of computer provider order entry overrides. Issue 6 (5th December 2020)

• Record Type: Journal Article
• Authors: O'Brien, Kerry L.; Chen, Yigu; Uhl, Lynne
• Is Part Of: Vox sanguinis. Volume 116:Issue 6(2021)
• Page Start: 702
• Page End: 712
• Publication Date: 2020-12-05
• Subjects: haemovigilance -- patient blood management -- platelet transfusion -- transfusion medicine; Blood -- Periodicals; Blood -- Transfusion -- Periodicals; Immunohematology -- Periodicals; Immunopathology -- Periodicals; 615.39
• DOI: 10.1111/vox.13050
• Languages: English
• ISSNs: 0042-9007

Abstract

Abstract : Background and Objectives: Judicious utilization of platelet products protects a limited resource and mitigates risks of transfusion. At many institutions, computer physician order entry systems provide prompts to guide transfusion decisions; many capture the indication for transfusion, and generate metadata when orders are dissonant with guidelines. We conducted a retrospective review to examine adherence to and overrides of hospital guidelines for platelet transfusion to identify opportunities for improved transfusion practice. Materials and methods: Physician override reports (1/1/2018–3/31/2019) were examined and physician‐entered justification comments accompanying override orders were extracted, in addition to patient‐specific data (clinical service, age, sex, and pretransfusion platelet count). Two transfusion medicine physicians independently assessed comments in context of patient data and institutional guidelines and categorized as: indicated, protocol driven, or not indicated. Following adjudication, consensus was reached between the two reviewers. Override keyword frequencies were also determined. Results: Over 15‐months, 1373 override orders were placed for 558 unique patients (25% of all adult inpatient platelet transfusions). haematology/oncology providers placed 573 (42%) override orders (261 unique patients), 46% of which were categorized as "not indicated", based on consensus review. Overall, 470 (34%) override orders were categorized as "not indicated". Examples of recurring key words included "bleeding/risk of bleeding", "falling platelet count", "platelet goal of XX". Conclusions: A large percentage of override orders for platelet transfusions were determined to be "not indicated" and out of compliance with institutional guidelines. The metadata captured identified concerns regarding clinical transfusion practice and opportunities for revised indications (e.g. threshold for retinal haemorrhage).