Dorien Brouwer

135 trend was no longer present after six months. Initially these results surprised me. The strong support in our control group, the timing and duration of the intervention, and cognitive problems may partially explain the lack of effect of motivational interviewing. An earlier meta-analysis of studies comparing motivational interviewing with a weak control group showed significantly higher effect sizes. [27] On the other hand, a high percentage of patients changed their health-related behavior (more than in our earlier study) in both groups. In my opinion, this is due to the nursing care that the patients received. Empathy, open questions, reflective listening and emphasis on patients’ autonomy are the basic communication skills of motivational interviewing. [28] These skills can be particularly expected of nurses and may therefore have contributed to the health-related behavior change of the patients in both groups in our study. A recent systematic review and meta-analysis showed a significant effect on reducing blood pressure, improving physical activity, diet, medication adherence and knowledge of risk factors on secondary prevention interventions after TIA or ischemic stroke in which nurses had a primary role. [29] Nurses should therefore play an important role in supporting patients in behavior change after TIA or ischemic stroke. They are also trained in supporting self-management and can approach problems such as anxiety from both a medical and psychological point of view. Nurses are able to integrate education, self-management support, motivational interviewing and treatment of risk factors in outpatient consults or daily care to support patients in health-related behavior change after TIA or ischemic stroke. Methodological considerations To the best of our knowledge, our prospective cohort study was the first to focus on determinants of health-related behavior change in patients with TIA or ischemic stroke. It provided insight in the behavior change process and potential point of applications for interventions. A limitation was that our study was not designed to change health- related behavior, and as a result only a few patients changed their health-related behavior. Therefore, we were not able to assess determinants of actual health-related change. This might partly explain why we only found a trend towards increased health-related behavior in patients with higher intention to change. Previous studies in health-related behavior has shown this intention-behavior gap as well [21] . The use of questionnaires in this prospective cohort study enabled us to collect a lot of information in a short time. We also thoroughly assessed the patients by conducting cognitive assessments and screening for depression. A limitation was that we were unable to examine patients’ subjective perspective of health behavior. Therefore, we decided to examine the patients’ personal experience and view on health-related behavior change after TIA or ischemic stroke in a subsequent qualitative study. Patients were able to talk freely, and factors and determinants emerged that were still unclear in our previous study. A limitation is that we did not collect any data other