5 Design study I: How to create an ambulatory app for patients with BPP 161 4.1 Strategy & scope (I) To be able to meet the needs of target users, it is crucial to really get to know them [88]. For this purpose, semi-structured interviews were used. Two different sets of questions were created, of which one was bound to the therapists and the other to the patients (interviews were held in Dutch, schemes available upon request). Interviews were recorded with video and audio equipment. The interviews were transcribed verbatim and analyzed according to the method described by Baarda, De Goede [89]. The analysis was performed with the ATLAS.ti software package. Transcribed text fragments were provided with codes or labels (i.e., characteristic names or descriptions) that were relevant for answering the research questions. All labels were sorted with respect to content and grouped to form (core) themes. This led to hierarchical structures of the labels, or taxonomies. Two researchers (YD and TdV) separately grouped the labels into themes. Cohen’s Kappa was calculated for both the therapist labels and the patient labels. Kappa values were 0,74 and 0,65, which represent a substantial agreement [90]. Taxonomies were summarized and user goals were derived. User goals were then used to formulate user needs, which in return were reformulated in terms of design requirements. Requirements were ranked according to the number of labels that corresponded with them in order to determine their priority for the future system. Together with overall statistics on age, diagnosis and level of education for each user group, these results were used to form Personas. The formation of personas was done according to the guidelines provided by the ‘PACT framework’ [91]. Personas provides a manageable yet insightful overview of the acquired data on user needs and provide a context for easily relating gathered information to when determining design requirements [92, 93]. 4.2 Structure (II) To determine the structure of the future e-coaching intervention it is essential to have an image of the mental map of the target user groups. Mental maps help to understand the flow of the task users will perform with the system, how users talk about the content of the system and how they will interact with the system [83]. They can also serve as a means to validate user requirements [94] and can bring forward new design requirements that take into account the way users interact with the system. Card sorting was selected as main method to examine and bring into view the mental maps of the target users. Card sorting is a commonly used method in UCD in which participants have to group a list of labeled cards into coherent sets [95]. Evidence
RkJQdWJsaXNoZXIy MjY0ODMw