Marjon Borgert

187 Implementation of a transfusion bundle products appear to be e ective. 26,27 Such systems are integrated in a blood ordering system in electronic patient les. Several studies showed signi cant reductions in blood products when a clinical decision support system was implemented. 26,28 Decision support systems complements might be used in conjunction with the transfusion bundle. According to (local) transfusion guidelines and the Joint Commission International (JCI) standards, informed consent should be obtained before transfusion. 16,29 Obtaining informed consent is one of the ve bundle interventions. Not obtaining informed consent was considered as one of the problems that can occur according to the Bow- Tie analysis. Additionally, obtaining informed consent is one of the JCI standards. Because of these reasons, the expert team have chosen to include this element in the transfusion bundle. Moreover, informed consent could have an indirect impact on reducing transfusions since patients are making well-informed decisions whether or not to receive blood products. Obtaining informed consent does not fully comply with the bundle requirements as set by the IHI, i.e. level one evidence. 6 However, according to the IHI, bundle interventions that are already recommended in (inter)national guidelines and by consensus of clinicians as being applicable to most patients might be considered for inclusion. 6 The strength of a care bundle is that a maximum of ve interventions can be included. 6 Not all recommendations can be put into care bundles. Other important problems or risk factors in the blood transfusion procedure could have been selected by the expert team as well, for instance, the mislabelling of blood samples, transfusion of two units of RBCs per transfusion without a re-check of the Hb after each transfused unit or wrong storage of RBCs. 15,16,20 We have chosen to include those intervention in the transfusion bundle that were marked as serious problems or risks in the Bow-Tie analysis and were based on evidence or international guidelines. The transfusion care bundle could result in diminishing costs by reducing the waste of scarce and expensive RBC products. These savings are apart from the indirect savings of transfusion-related adverse events when the restrictive policy is applied to the majority of the patients. 21 Limitations Our study has some limitations. The study was a single-center study, which limits the external validity of our results. Furthermore, a before and after design was used. Therefore secular trends may have in uenced our results. Baseline transfusions were retrospectively collected; however, each transfusion was thoroughly reviewed in the