Iris Kanera

96 Chapter 4 Prior research confirms that the possibility to choose within multiple behavior interventions may prevent high attrition rates and could improve intervention outcomes (Brouwer et al., 2010; Kwak et al., 2010; Schulz et al., 2012). Possibly, offering wide ranging support in combination with personalized referral to relevant topics and the possibility to choose might prevent overload. Donkin et al. (2013) support this suggestion by reporting that a certain level of usage might be needed to obtain benefit from an online intervention for depression. However, after reaching a point of therapy saturation, little or no additional program gains might be expected. This is in line with a web-based study among cancer survivors and with another web-based obesity prevention study among the general population, which reported that more intervention use did not result in better intervention outcomes (Borosund et al., 2013; Walthouwer, Oenema, Lechner, & de Vries, 2015c). Using a higher number of modules may not be necessary for all users to benefit most from the KNW. To illustrate, the present results revealed that having no partner was related to the use of a higher number of modules, and participants who were in greater need of support (higher number of red/orange MRA) indeed used a higher number of modules. This is in line with the findings of Borosund et al. (2013) who reported that particularly cancer survivors with low levels of social support and a high illness burden used self-management components of a web-based illness management support system. Furthermore, higher perceived personal relevance was related to using a higher number of modules, which might be explained by receiving a higher amount of computer tailored content within the modules. The overall KNW was highly appreciated with an average grade of 7.5, indicating an appreciation from very satisfactory to good. The low variability ( SD = 1.2) indicates a considerably unanimous positive rating six months after getting access to the KNW. Results from the present study indicate that perceived personal relevance might be a key component to explain a higher appreciation. Computer tailoring was applied within the KNW in order to create personal relevant feedback. Since perceived personal relevance could not be predicted by demographic and cancer-related characteristics, we can conclude that the tailoring of information worked well. In comparison, the overall satisfaction of a generic fully automated web-based self-management intervention for breast cancer survivors was mean 7 ( SD = 1.2; van den Berg et al., 2013). In addition, the overall appreciation of a web-based weight management intervention for overweight adults was mean 6.6, and the overall appreciation of a web-based text and video tailored intervention for smoking cessation in the general population was mean 6.5 ( SD = 1.6; scales ranged from 1-10; Stanczyk et al., 2014; van Genugten et al., 2012). The overall appreciation ratings of KNW module users were more positive than the ratings of module non-users, although the module non-users were still quite positive in their ratings. In addition to the modules, the KNW comprises a user forum and participants received monthly emails inviting them to visit generic monthly news-items. Possibly, filling out the screening questionnaire and follow-up questionnaires,