Iris Kanera

168 Chapter 7 2018). Since significant increases on QoL domains emotional and social functioning, and a reduction in depression and fatigue were observed within the intervention group after six months that remained fairly stable in long term (Willems et al., 2017a; Willems, Mesters, Lechner, Kanera, & Bolman, 2017b), it would be relevant to study how the two main themes influence each other in the main results. A recent review showed that moderate intensity exercise in particular is beneficial in reducing fatigue among cancer survivors, which in turn has a positive effect on QoL (Dennett et al., 2016). Hence, we expect that our revealed increase in moderate PA might boost QoL for early cancer survivors . Moreover, it is considered relevant to gain further insights into the quantitative and qualitative aspects related to the process of engagement in the KNW. For instance, log data could be used to analyze possible usage patterns of individuals and subgroups. Moreover, qualitative data could be collected on whether non-adherence with the MRA was a result of a conscious choice of the participants or was caused by other reasons. In addition, reasons for not accessing the KNWwebsite at all could be an issue for further qualitative exploration. In addition, quantitative exploration of eligible cancer survivors that never logged in to the KNW is possible as well. Characteristics such as gender, age, type of cancer, type of treatment, and the termination date of primary cancer treatment were registered for all approached cancer survivors during recruitment. Based on those data, characteristics may be identified to predict the initial use of the KNW that might reveal valuable insights about the target group of the KNW. This information might be beneficial for decision-making concerning the approaching of future cancer survivors to participate in the intervention. As discussed in Chapter 4, the associations between the MRA and module use must be interpreted with caution, since there was no intervention group using modules while not receiving the MRA. Future experimental research might explore the specific added effect of an automated referral system (such as the MRA) on subsequent choices and possible behavioral effects. In addition, since no advanced statistical assessment of smoking was possible due to a low number of smokers and a related power problem, research on a possible KNW effect on smoking should be conducted in a sample with smokers exclusively. PROPOSALS FOR IMPROVEMENT OF THE INTERVENTION First, if the KNW would be implemented into practice, the intervention must be kept up- to-date, regarding the content and technical features. Recently, the Dutch diet and PA recommendations were revised, which means that KNW elements that are based on these recommendations need to be modified; e.g., the recommended amount of daily vegetable intake was increased from 200 g to 250 g; to the recommended weekly amount of 150 min PA, strength exercises (two times per week) were added. Moreover, hyperlinks that