Marjon Borgert

80 Chapter 4 DISCUSSION The results of our systematic review show that besides the IHI approach various additional methods exist to design care bundles. Most included studies reported only one part of the design process (67%, 12/18), while in 33% (6/18) the whole process was described. Given the diversity in the methods used for designing care bundles, it might be suggested that the original IHI methods may not always be applicable to all ICUs and in every situation. For example, Romero et al. selected a set of elements by using the results of their analysis on medication errors. The potential elements were based on the types and causes of medication errors that were reported during their baseline period. 30 To prevent these errors, a care bundle was created based on these types and causes of medication errors. In this case, the IHI method for identifying risks might not have given the best results for this ICU. Moreover, Khalid et al. used a RCA for identifying risks. 11 They show that this is an e ective tool to clearly identify the local risks and discover the potential weak links in the process. They show that the results of a RCA could form a perfect basis to design new care bundles. Furthermore, we identi ed studies in which di erent types of bundles were developed. Besides the well-known central line bundle and ventilator bundle, other care bundles were described in the literature such as the bundle for prescribing antibiotics 22 , ventriculostomy placement 26 or for the bundle in palliative care. 27 The rst step in the bundle design process is to identify (potential) problems or risks. 2 The IHI used the adverse event trigger tool for this step. 34 Besides this tool, we identi ed additional risks assessment tools, such as a RCA 11 or FMEA. 26 These can be highly e ective in the bundle design process due to their focus on local problems or risks. 11,26 This is important for designing care bundles because the included bundle elements should be a generally accepted practice in order to deliver reliable care. 2-4 Rello et al. used the Multi-Criteria Decision Analysis (MCDA) to design the ventilator bundle. They showed that this method is highly structured and e cient to use in the bundle design process. 10,29 Another example is the use of a systematic review. The IHI used this method for designing the ventilator bundle.[4] Systematic reviews were also reported in the literature to underpin evidence for the bundle elements in step 3 of the development process. It is important that care bundles meet the IHI criteria. One of the criteria is that bundle elements must be supported by level 1 evidence. 2-4 However, robust evidence of care processes in relation to patient outcomes is often not available. 35,36 Therefore, evidence could also consist of clinical practice guidelines or other peer-reviewed synthesis of the evidence or studies published in a peer-reviewed journal. 35,36 Even though care bundles