Chapter 6 148 differences, the evaluators determined a consensus score. Because groups had more than one IMR trainer, ratings were based on how the trainers functioned together (7). The principal investigator was trained by two US experts at assessing IMR fidelity (19). The two educated co-auditors were a psychologist and an advanced nurse practitioner, both of whom had experience in providing IMR themselves. Besides semi-structured interviews with IMR participants and trainers of the particular IMR group, assessments consisted of one observational session and monitoring of forms, such as anonymous IMR Goal Tracking Sheets and progress notes regarding the five latest IMR sessions. To get a complete picture, IMR fidelity assessments were conducted during one of the last sessions of the curriculum. This was on average about a year after the IMR trainers' initial training. During the interviews, all IMR elements were reviewed with all respondents. While the applied assessment procedure was based on the IMR Fidelity Scale protocol (24), the IT-IS was designed to rate the fidelity of clinicians to the IMR program based on observations of treatment sessions (either live, audio recorded, or video recorded) (7, 36). However, the logistics of organizing multiple recordings per IMR group and subsequently assessing them were not feasible in the context of this RCT. Therefore, in this study, the IT-IS rating was adapted based on the IMR Fidelity Scale. Interviews for the IMR Fidelity Scale were extended to allow for the rating of the additional items in the IT-IS. Moreover, using the indicators of excellence, the rating of the IT-IS items required detailed questioning on the application of various competencies; this included questions on competencies that were not or hardly used during the observed session. These interviews aimed to capture IMR practitioners’ knowledge and actual skill level in the concrete application of the various aspects of the IMR competencies specified in the IT-IS protocol for each item. This also included requesting examples. For both scales per IMR group, various participants’ responses to the same question were continuously checked for consistency. Furthermore, the observations and chart review results were always the starting point for the interviews. For all elements, information from direct observations, interviews, and chart reviews was integrated by the assessors to score the fidelity scales. Thus, a good overall picture of the fidelity within each group was obtained, and assessing all items of the IMR Fidelity Scale and the IT-IS could be achieved. With this fidelity data collection method, we could minimize the potential impact of which particular IMR session was observed. Moreover, previous research on the IT-IS showed the limited impact of the module covered in the rated session (36). The clinicians leading the groups and the team leaders received a report on the scores in their group. Subsequently, researchers provided fidelity feedback on the
RkJQdWJsaXNoZXIy MjY0ODMw