15250-m-cuypers

164 Chapter 8 of patient preferences 35 . In our study we did not make use of automated systems to distribute the DA (e.g. automated sending though a link with patient record), although some studies suggested a system-based delivery method is potentially most successful 8 . Automated systems could help recognizing eligible patients with a lower risk of forgetting or misinterpreting patient preferences to use a DA. Furthermore, our study found mixed views on whether a DA should be introduced by a urologist or an (oncology) nurse and if DA distribution should be part of formal treatment guidelines. In this study, both urologist and oncology nurses felt themselves suited to distribute the DA. When this responsibility is not clearly defined in a local hospital’s care path, a diffusion of responsibility could emerge that hinders sustained DA implementation. Another potential barrier for DA implementation in Pca care is that health care providers in our control arm reported high levels of satisfaction with their existing (hospital specific) information materials and verbal information provision. Although health care providers in the current study acknowledged that patient information satisfactionmight be lower compared to their own satisfaction, this might even be an overestimation of actual patient satisfaction. Previous studies showed that considerable proportions of Pca patients are dissatisfied with information received and that large discrepancies exist between actual information preferences and their physician’s perceptions 36, 37 . This difference between actual and perceived information satisfaction could lead to reluctance among health care providers to use externally developed tools such as DAs. It is therefore important to inform health care providers about patients’ actual (information) needs, but also to provide the opportunity to adapt the DA to hospital specific materials that care providers already feel satisfied with. Some limitations of this study need to be mentioned. First, the current trial increased awareness for applying SDM in Pca treatment decision-making and for DAs in both study arms. Moreover, other Pca DAs initiatives were enrolled in a large number of Dutch hospitals as well 32, 33 . It can therefore not be assumed with certainty that no individual patient from the control group came into contact with any of the available DAs. However, we do not expect a significant effect on our results from this potential contamination, as no DAs were actively implemented in the hospitals from our study’s control group. A second limitation is the fact that DA use in this study was linked to the RCT, meaning that patients who were provided with the DA, were also asked to take part in the RCT and fill out three questionnaires. Discussing the DA with a patient therefore always