Dunja Dreesens

128 of the review and Delphi were compared with other studies. Although glossaries existed of healthcare terms and some tools, and reviews of one specific tool type, similar exercises as carried out in the Netherlands were not found. The approach of the review and the Delphi could be repeated in other countries and/or in an international setting to see if it renders comparable results. Though this might be more a criterion for dependability. The data of the scoping review and Delphi-study were transferred to an international setting when developing the framework of patient-directed knowledge tools (chapter four). Again, the international experts did not add many new tool types when developing the framework. They, however, constricted themselves to patient-directed tool types. Possibly the framework could be transferred to other populations, such as healthcare providers or policy advisors, and the set up with purposes and core elements could be maintained. It is doubtful if the purposes and core elements would be (exactly) the same content wise, as the target group(s) of the tool types differ. The findings of the study concerning different kinds of knowledge and different kinds of reasoning (chapter five) might be transferable to other domains. Not just health care tries to be as much evidence-based as possible to decide what is ‘the best’ way to go forward and to change practice and policy. Evidence-based working is becoming more eminent in other domains such as policy decision-making and the practice of law as well (296), even though they might have different knowledge sources and have not progressed as far with the evidence- based approach yet, they have to deal with some of the same issues and pitfalls as health care has. The findings of the interviews regarding integrating knowledge translation tools with shared decision making (chapter six) could be transferred to guidelines on other subjects and guidelines for other healthcare professionals than paediatricians. Guideline developments groups are increasingly looking into how they can incorporate patient perspective and preferences in the clinical practice guidelines. Dependability Qualitative research is characterised by fluid structures, which change because of incoming and available data. Therefore, for each study the approach and/or method chosen and the changes therein, if any, were exhaustively described. Furthermore, proven methods such as the RAND-modified Delphi were used. Another limitation of qualitative data (but also of quantitative research) is that non-reported data can disappear. By being transparent about the found and used data, making them available with the publication of each article or on request, this limitation is hopefully addressed. That does not mean that findings will be similar if the studies were to be repeated. Reasons why findings might be different could be due to firstly different participants in the Delphi-study, interviews and invitational meeting, or original participants might have gained more knowledge on the subject and changed their mind. Secondly there might have been developments since the studies were executed which could affect settings and the findings; for example, the uptake of a core set of tool types and the definitions of these tool types and a revised clinical practice guideline Palliative care for children. Thirdly, different researchers with another mind-set might infer different conclusions, as qualitative research is more subjective in nature and is influenced by the researcher. That is why, the research was executed under supervision and within a team, so that not one perspective could get the upper hand. Field notes were taken and discussed to Chapter 7