Marjon Borgert

66 Chapter 4 available that could also support the development of new bundles for the ICU besides the IHI approach. Therefore, we conducted a systematic review. The primary objective was to identify what di erent methodologies were used in the literature to design new evidence-based ICU care bundles. Based on the results, we built a comprehensive owchart to provide an overview of the methods used so that others could choose their own desired method and to guide through the necessary steps of the development of new evidence-based care bundles for the ICU. MATERIALS & METHODS Design A systematic reviewwas conducted to identify methods for designing new care bundles for adult ICUs. The protocol for this study was not registered. Selection criteria We included studies that described the di erent methods within the whole care bundle design process in adult ICUs or the methods described in just certain parts of the design process. Studies were also included in case one or more IHI methods were used. Studies of any design were included and published in the English language. Search strategy A systematic search was performed in the electronic databases PubMed, EMBASE and CINAHL from the year care bundles were designed in January 2001 to August 2014. Furthermore, the reference lists of the full-text articles were screened. The search was designed for maximal retrieval, with no limitation of language or types of study design to be identi ed. The complete list of search terms and strategy of PubMed can be found in Supplementary File 1. Study selection The screening of the titles and abstract was conducted in two parts. At rst, one author (M.B.) roughly screened all titles and abstracts. Studies were excluded when: (i) the language was not in English; (ii) the bundle was designed for pediatric departments or non-ICU departments or (iii) care bundles were not the subject of the study. Secondly, the titles and abstracts of the remaining articles were again screened. Two authors independently screened the titles and abstracts (M.B.,D.D.). In case of discrepancies, we reached consensus through discussion. A third reviewer was involved in case of disagreement. Full-text studies were reviewed and selected by two authors