Web-based lifestyle support for cancer survivors

The aim of this dissertation was to develop and evaluate the web-based Cancer AftercareGuide (Kanker Nazorg Wijzer, KNW), a fully automated and easy accessible self-managementintervention that includes lifestyle as well as psychosocial topics and targets survivors ofvarious types of cancer. A reason for developing this comprehensive eHealth intervention isthe steady increase of the number of individuals who survive cancer and the need for supportduring recovery from cancer. Additionally, cancer survivors are at higher risk of (long-term)physical, lifestyle, and psychosocial problems, as well as of developing comorbidities andnew cancers. Adopting and maintaining a healthy lifestyle, such as being physically active,consuming a healthy diet, and refraining from smoking, is highly beneficial in reducingthe risk of morbidity and mortality, and to improve quality of life. However, many cancersurvivors do not adhere to healthy lifestyle recommendations, and they report to be inneed of support to (re)gain a healthy lifestyle balance and to manage residual psychosocialproblems. Therefore, behavioral support is needed to achieve sustainable lifestyle changesand to cope with psychosocial problems.In the Netherlands, cancer aftercare needs to be improved to offer adequate support for thegrowing numbers of cancer survivors, as concluded by the Health Council of the Netherlandsin 2007. The subsequently developed guideline ‘Recovery from Cancer’ (Herstel na Kanker)pleads for a broad programmatic approach for oncology aftercare including attention forthe early recognition of survivors’ psychosocial and lifestyle risks and needs during earlycancer survivorship. Moreover, the guideline recommends stimulation of self-management,and applying a stepped care approach as an alternative care delivery system to providemore efficient and personalized aftercare. As the Internet is increasingly used as a sourceof health-related information, a stand-alone theory-grounded eHealth intervention wassuggested to fill this important gap in current cancer aftercare. Advantages of web-basedintervention are that they are accessible anytime and anywhere and that they can reachmany survivors at once without involvement of health professionals. In the Netherlands,there is a lack of comprehensive web-based cancer aftercare interventions, while thoseinterventions might provide “guided” self-help (i.e., self-management). However, existingresearch into the effect and usage of multi-behavior web-based interventions for cancersurvivors is very scarce. Previous studies into the effectiveness of eHealth interventionsincluded interventions which were mostly less comprehensive, included in-personcontact with a health professional, or included a relatively short follow-up period. Thus, itwas considered useful to test whether a fully automated multi-behavior cancer aftercareintervention might be effective in changing lifestyle behaviors, such as physical activity, dietbehavior, and smoking behavior in the short and long term. Moreover, it was considereduseful to study how cancer survivors would use and appreciate a stand-alone online canceraftercare intervention that addresses a broad range of topics.Chapter 1 provides the background and rationale of the current thesis and highlights theimportance of promoting the adoption and maintenance of healthy lifestyle behaviorsamong cancer survivors. Relevant determinants of lifestyle behaviors and behavior changemethods are considered, and the theoretical psychosocial framework used for the studiesconducted is explained. In addition, the pros and cons of applying web-based interventionstargeting cancer survivors are described, and an overview of previously conducted studiesevaluating web-based lifestyle interventions for cancer survivors is provided.Chapter 2 assesses the behavioral risks of the target population related to smoking, physicalactivity, alcohol, and fruit and vegetable consumption. Therefore, a cross-sectional studywas conducted. Relevant correlates of these different lifestyle behaviors among early cancersurvivors were identified. Results showed that only a small group of cancer survivors (11%)adhered to the recommendations for all five lifestyle behaviors, and that both adherence tothe recommendations for vegetable (27.4%) and fruit (54.8%) consumption was particularlylow. The majority of cancer survivors followed the recommendations on physical activity(87.4%), refrained from smoking (82%), and followed the alcohol recommendations (75.4%).Each separate lifestyle behavior was influenced by different patterns of correlates, however,self-efficacy, attitude, and intention contributed to the highest extent. The insights gainedfrom this study were valuable for the development of the intervention.Chapter 3 describes the systematic development of the web-based computer tailoredcancer aftercare intervention by using the Intervention Mapping protocol. The interventionaims to reduce cancer survivors’ experienced problems in seven areas, based on the needsassessment: (1) cancer-related fatigue, (2) difficulties concerning return to work, (3) anxietyand depression, (4) relationships and intimacy issues, (5) a lack of physical activity, (6) alack of healthy food consumption, and (7) difficulties in preparing or maintaining smokingcessation. To address these problem areas, seven self-management training modules weredeveloped. They were based on principles of problem-solving therapy, cognitive behavioraltherapy, social-cognitive, and self-regulation theories. It is expected that reducing theproblems that are experienced through behavioral change will ultimately result in a higherquality of life. Since the intervention comprises a broad range of topics and targets a variedgroup, the content is personalized by means of computer tailoring. A personalized ModuleReferral Advice (MRA), based on identified risks and needs, offers guidance on whichmodules are most essential to use. Also, the information and advice provided within themodules is tailored to demographic, cancer-related, psychological, and motivational factors,and to the current risk behaviors and needs. Moreover, interactive features, video material,animations, and hyperlinks are included in order to substitute an element of personalcontact and to support the recall of the provided information. This chapter also includes adetailed description of the study design for evaluation of this cancer aftercare intervention.Chapter 4 describes the process evaluation of the use and appreciation of the KNWintervention. It is investigated whether the participants allocated to the interventionfollowed the advice of the MRA. Moreover, the use of the modules and its predictors, theappreciation of the KNW and its predictors, and the predictors of personal relevance of themodules were identified. Almost all (98.3%) participants were referred by the MRA to at leastone self-management module (M 2.9, SD 1.5), and the majority (85.7%) visited on average2.1 (SD 1.6) modules. The results indicate that the MRA might be an important interventionelement to guide the users to a preferred selection of modules, considering that participantswere more likely to use relevant modules after a referral by the MRA. All modules wereused to varying degrees, and a higher number of modules used were predicted by a highernumber of risks and needs and having no partner. The overall KNW and its modules werehighly appreciated, which was related to a higher perceived personal relevance. Notably,the intervention was perceived just as personally relevant by participants with differentdemographic and cancer-related characteristics.Chapter 5 studied possible effects of the KNW on lifestyle behaviors (i.e., respectivelyvegetable, fruit, whole grain bread, and fish consumption, physical activity, and smokingbehavior) six months after baseline. Therefore, a randomized controlled trial was conducted.Indications were found that participating in the intervention increased moderate physicalactivity and vegetable intake. A meaningful increase in moderate physical activity of 151minutes per week was observed, which was 75 minutes per week higher compared to thecontrol group. Moreover, using the behavior-specific modules indicated possible effectsfor moderate physical activity, fruit, and fish consumption. However, the results did notremain significant after correction for multiple testing and should, therefore, be interpretedwith caution. No significant intervention effect was found on smoking behavior due to lownumbers of smokers, although smokers in the intervention condition were almost threetimes more likely to quit than smokers allocated to the control condition.Chapter 6 examined whether the positive changes in moderate physical activity andvegetable consumption determined after six months were maintained at 12 months.Additionally, possible moderator effects of using behavior-specific modules, gender, age,and education level were investigated. The results showed that the KNW was effectivein increasing and maintaining moderate physical activity in the long term among earlycancer survivors younger than 57 years with a moderate effect size. This effect was clinicallyrelevant. The 6-month increases in vegetable consumption were not sustained in the longterm.Chapter 7 gives a summary and discussion of the main findings of the studies included inthis thesis and discusses the methodological considerations. It offers suggestions for futureresearch, proposals for intervention improvement, and implications for implementationof the intervention into clinical practice. Important strengths of the presented KWNintervention studies were the large study population, the strong study design (randomized controlled trial), the follow-up period of 12 months after baseline, the low dropout rates,the advanced statistical analyses including intention-to-treat analyses, and correctionsfor multiple testing. There are also some limitations such as the composition of the studysample. The recommendations for future research include the investigation of the workingmechanisms of the intervention, i.e., whether the effects occurred by the hypothesizedchanges in social cognitive factors and an expanded quantitative and qualitative evaluationrelated to the process of engagement in the intervention. Furthermore, future studiesshould focus on the determinants of behavior maintenance. Additionally, the effects of theautomated referral system (MRA) could be tested more fundamentally. Importantly, afterimplementation of the KNW into clinical practice, the use and effects of the interventionshould be followed up. In addition, a number of suggestions for improvement of the KNWare provided, including adaptations in the content and layout of the website to keep upwith updates and (technical) developments. The KNW intervention can be implementedby linking the intervention to trusted websites. Moreover, the KNW can be combined within-person guidance to address a broader group of cancer survivors.In conclusion, indications were found that having access to the web-based KNW mayincrease vegetable consumption and moderate physical activity after six months, while theKNW was effective in increasing moderate physical activity among cancer survivors youngerthan 57 years of age at 12 months. Despite the broad scope, relevant KNW modules wereused and the intervention was highly appreciated and perceived as personally relevant. Thestudies in this thesis indicate the usefulness and value of the KNW to respond to the growingdemand of comprehensive and personalized cancer aftercare. Overall, the fully automatedKNW is applicable to a large audience and provides personalized self-management supportfor a broad range of problem areas that cancer survivors face after completing primarycancer treatment.