Chapter 2 32 - Shared-decision-making, evaluated using the Shared Decision Making Questionaire-9 for physicians (SDM-Q9-DOC). 9 items are scored on a 6-point Likert scale ranging from 0 (totally disagree) to 5 (totally agree). Items are summed and multiplied by 20/9 to provide a score with 0 indicating the lowest and 100 the highest possible level of SDM (24–27). - Doctor preparedness to discuss treatment wishes, evaluated through 8 questions ranging from very generic to care decision specific, and a mock question about medication to mask the focus of this study. - Patient appreciation of the conversation aid (intervention group only), evaluated through 10 questions on aspects of the conversation aid, an overall score, and a free text space for additional suggestions. In summary: for each patient seen by a physician, the physician needed to complete the SDM questionnaire and physician preparedness assessment, combined in one questionnaire. All patients completed the satisfaction items, and the patient-doctor relationship questionnaire. Patients in the patient intervention group additionally completed the questions on their appreciation of the conversation aid. 2.5 Statistical analysis We performed an intention to treat analysis. Patient characteristics are shown stratified by intervention group. Physicians’ characteristics were described narratively. The primary outcome of mean patient satisfaction score was shown stratified by intervention using medians and interquartile ranges. The primary outcome in the intervention group (both patient and physician trained) was first compared to the reference group (neither patient nor physician trained) with a Mann-Whitney U test. Following a gatekeeping procedure to reduce the risk of a type I error, further statistical comparisons between the patient intervention-group and physician intervention-group versus the reference group would have been performed only if the primary outcome differed between the intervention group and the reference group (fixed sequence hierarchical testing). We used the same strategy for the patient-doctor relationship and shared-decision-making outcomes. To adjust for confounders while taking into account dependence between scores of patients within physicians, primary and secondary outcomes were analysed using a multilevel mixed
RkJQdWJsaXNoZXIy MTk4NDMw