15250-m-cuypers

144 Chapter 7 relation between pre-diagnosis exposure to Pca treatment information and satisfaction and regret, indicating that the moment of presentation of information can have an effect on patient outcomes (27) Future research Relevant continuations of the current research are to investigate how the web- based aspect of the current DA is related to anxious and depressive symptoms, and incorporation of the DA into direct patient-doctor counseling. The current DA has been developed as online tool for its possibilities to tailor presentation of information (e.g. only displaying information that is relevant to the patient), include interactive VCEs, and because nearly all Dutch citizens have internet access (29, 54). As this study showed, anxious and depressive symptoms and the patient-doctor relation are associated with both regret and satisfaction. This raises the question how this relates to the fact that the DA was only available online. Use of the DA content was not integrated in clinical counseling, only discussion of the summary obtained from the DA was embedded in the follow-up consultation. Moreover, online sources could be perceived as less trustworthy, in particular for patients suffering from anxious and depressive symptoms (55-57). Possibly, these patients require offline materials that could be combined with the online DA. The current study focused on patient-reported outcomes in the first year after treatment decision-making. A longer follow-up period (up to 36 months) including clinical data about possible tumor recurrences (or further tumor progression in case of AS) could provide insight if the patients from the DA groups adjusted better due to more accurate risk perceptions (8). Strengths and limitations Strengths of the current study include the pragmatic approach that allowed effects from every day clinical routine to be included in the trail, and contributing to the external validity of our results. Furthermore, drop-out rates were low and equally distributed across trial arms. Patients who consented and completed the first questionnaire were also highly likely to complete the two follow-up questionnaires. The cluster randomizeddesignof this trial was chosen to reduce the riskof contamination from clinicians that counsel both patients included in the DA arm as in the control arm, as in a traditional individually RCT would occur. Such design is recommended when behavior is part of the intervention (58). By taking hospitals as unit of randomization instead of patients, we avoided that care providers, after they received DA training, had