Marjon Borgert

167 Implementation of the transfusion bundle level. This is consistent with the model of actionable feedback. 20 This model posits that A&F is the most e ective when it is timely given, individualized and in a non-punitive way. An important di erence is that we have used a shorter time span in which feedback was provided compared to Hysong et al.. 19 The reason was that nurses would still clearly remember their actions and it gives nurses the opportunity to change their behaviour before the next transfusion occurred. Furthermore, we provided individual feedback in a non-punitive way. Feedback was also given when the performance of an individual was optimal i.e. compliance was 100%. This respectful and non-punitive way may have improved bundle compliance. 20 Although this studyhas showna signi cant e ect oncomplianceduring implementation when using the combination of monthly A&F plus timely individual A&F, there are reasons for not reaching the optimal e ect. One of the reasons could have been that it was not always possible to meet the nurse within 72 hours. As per protocol, we then sent personal feedback by email. Even though the report was personalized and written in a non-putative manner, this might have had less impact than actual face- to-face feedback. As there is evidence that providing feedback face-to-face improves implementation. 18 Furthermore, there is evidence that when team members know each other’s individual performance levels, this will lead to an improved level of overall compliance. 19 This means that when we showed the compliance level of each individual nurse to the whole nursing team, higher bundle compliance levels would have been achieved for that nursing team. Moreover, by expending the implementation period higher compliance levels could have been achieved. In studies that achieved high levels of bundle compliance periods were used of at least one year. 26 Providing timely individual feedback is labour intensive especially when the teams are large and the implementation period is long. A cost-e ectiveness analysis would be recommended for future research. Cost-e ectiveness of A&F could be enhanced using information and computer technology. Zaydfudim et al. used an electronic monitoring and compliance system to sustain the implementation e ect of the ventilator bundle. 27 Achieving sustainability is a major challenge in implementation. 28 In our study, compliance gradually decreased in both teams during the post-implementation period. This so-called ‘washout phenomenon’ is a well-known factor in implementation. 28 Althoughcompliance levelsdidnot signi cantlydecreasedbetweenthe implementation and post-implementation period. Continuing the combined A&F strategy might have had a sustained e ect on bundle implementation. 29