RefCheck Maintenance Notice

On Monday, December 3, 2018, from 16:00-18:00 EST, RefCheck will be undergoing maintenance. RefCheck is the process where, during copyediting, all references are extracted from the manuscript file, parsed, matched against various databases (eg, PubMed and CrossRef), and automatically corrected. For more information on RefCheck, please visit our Knowledge Base.

Who will be affected?


Citing this Article

Right click to copy or hit: ctrl+c (cmd+c on mac)

Published on 23.06.15 in Vol 4, No 2 (2015): Apr-Jun

This paper is in the following e-collection/theme issue:


    Effectiveness, Mediators, and Effect Predictors of Internet Interventions for Chronic Cancer-Related Fatigue: The Design and an Analysis Plan of a 3-Armed Randomized Controlled Trial

    1Roessingh Research and Development, Telemedicine Group, Enschede, Netherlands

    2University of Twente, Faculty of Electrical Engineering, Mathematics and Computer Science, Telemedicine Group, Enschede, Netherlands

    3Helen Dowling Institute, Scientific Research Department, Bilthoven, Netherlands

    4Utrecht University, Department of Methods and Statistics, Utrecht, Netherlands

    5North-West University, Vanderbijlpark, South Africa

    Corresponding Author:

    Marije DJ Wolvers, MSc

    Roessingh Research and Development

    Telemedicine Group

    Roessinghsbleekweg 33b

    Enschede, 7522AH


    Phone: 31 534875777

    Fax:31 534340849



    Background: Internet interventions offer advantages that especially cancer survivors who suffer from fatigue could benefit from. Given the growing number of such patients, Internet interventions could supplement and strengthen currently available health care.

    Objective: This paper describes the design and analysis plan that will be used to study 2 Internet interventions aimed at reducing severe fatigue in cancer survivors: a mobile ambulant activity feedback therapy supported through a weekly email by a physiotherapist and a weekly Web- and mindfulness-based cognitive therapy supported online by a psychologist. The data resulting from this trial will be used to (1) investigate the effectiveness, (2) investigate potential mediators of these interventions, and (3) explore participant characteristics that can predict the effect of these interventions.

    Methods: A 3-armed randomized controlled trial is proposed that compares both Internet interventions with an active control condition that solely consists of receiving psycho-educational emails. The intervention period is 9 weeks for all 3 conditions. Six months after baseline, participants in the control condition can choose to follow 1 of the 2 experimental Internet interventions. Outcomes are measured in terms of fatigue severity, mental health, and self-perceived work ability. All are Web-assessed at baseline, 2 weeks after the intervention period, and at 6 and 12 months after baseline. Fatigue severity, mindfulness, physical activity, expectations and credibility of the intervention, therapeutic working alliance, sleep quality, and sense of control over fatigue are assessed 3 times during the intervention period for identifying mediators of the interventions. Recruitment is performed nationally throughout the Netherlands through patient organizations and their websites, newspapers, and by informing various types of health professionals. All participants register at an open-access website. We aim at including 330 cancer survivors who have finished curative-intent cancer treatment at least 3 months previously, and have been suffering from severe fatigue ever since. All cancer types are included. A detailed analysis plan is described to address the research questions, which allows for individual variation, and fully exploits the longitudinal design.

    Results: Recruitment started in April 2013 and will proceed until April 2015.

    Conclusions: This paper describes a systematic trial design for studying 2 different interventions for chronic cancer-related fatigue in order to gain insight into the effectiveness and mediators of the interventions. This design will also be used to identify predictors for the interventions’ effect on fatigue. By publishing our hypotheses and analysis plan before completion of data collection, this paper is a first step in reporting on this trial comprehensively.

    Trial Registration: The Netherlands National Trial Register (NTR3483). (Archived by WebCite at

    JMIR Res Protoc 2015;4(2):e77



    Crowdfunding campaign to support this specific research

    We help JMIR researchers to raise funds to pursue their research and development aimed at tackling important health and technology challenges. If you would like to show your support for this author, please donate using the button below. The funds raised will directly benefit the corresponding author of this article (minus 8% admin fees). Your donations will help this author to continue publishing open access papers in JMIR journals. Donations of over $100 may also be acknowledged in future publications.

    keyboard with crowdfunding key instead of enter key

    Suggested contribution levels: $20/$50/$100



    Behavioral interventions have shown to effectively relieve psychological and physical complaints in cancer survivors. However, the effect on the individual is less explicit, because patients differ greatly in the ways they experience and respond to such interventions. Therefore, when studying such an intervention, individual differences and temporal aspects need to be appreciated. This paper presents a detailed analysis plan for studying behavioral interventions that satisfies such needs.

    The protocol of a 3-armed randomized controlled trial is described to study the effectiveness, mediators, and effect predictors of 2 different Internet interventions that share the same aim: reducing fatigue for cancer survivors. Due to its longitudinal design and multiple assessments during the intervention, the temporal development of relevant factors rather than pre-post differences can be studied. Latent growth analysis can be performed and mixture models can be run, which allow for individual variance in growth trajectories. Furthermore, full longitudinal mediation analyses can be performed on the most important potential mediators of both interventions, and differentiating effect predictors can be identified in order to allocate individuals to the most suitable intervention.

    The goal of this paper is to present our trial design, hypotheses, and analysis plan. This paper will therefore be the basis for a number of papers that will present the results of the trial. We will first provide brief background information on the research population, the relevance of Internet interventions for this population, and introduce the 2 Internet interventions that are the subject of this trial. Next, the importance of identifying mediating and predicting factors for the intervention effect is discussed. In the remaining sections, we give a detailed description of the trial’s design, our hypotheses, and the analysis plan for handling the data that the trial will collect. The analysis plan is written in general terms, in order to facilitate the use of this strategy in other contexts, and to keep this paper focused. Consequently, the extended background of—and reasoning for—the specific hypotheses will be presented in future papers that will focus on the results of the proposed analyses.

    Chronic Fatigue and Cancer

    Cancer-related fatigue is defined as “a persistent, subjective sense of physical, emotional and/or cognitive tiredness or exhaustion related to cancer or cancer treatment that is not proportional to recent activity, and interferes with usual functioning” [1]. It is 1 of the most prevalent and distressing long-term consequences of cancer [2], interferes with the activities of daily living, work ability [3], and maintenance of social relations, and consequently impacts patients’ well-being [4]. As the number of cancer survivors in the Netherlands is expected to increase rapidly, with a growth of over 50% in the 10-year prevalence between 2009 and 2020 [5], there is a strong need for effective and accessible treatments.

    The etiology of cancer-related fatigue probably involves the deregulation of several interrelated physiological, biochemical, and psychological systems [6]. There is no definite somatic explanation for the persistence of fatigue after cancer [7-9], and estimates of the proportion of cancer survivors who suffer from persistent fatigue vary widely [8,10]. However, research has shown that if fatigue continues 3 months after treatment, it is unlikely to decrease of its own accord [8]. The term chronic cancer-related fatigue (CCRF) is used in this paper for severe fatigue that continues for 3 months or longer after cancer treatment completion.

    Management of Chronic Cancer-Related Fatigue

    Currently, both pharmacological treatments and nonpharmacological treatments are applied to the effective management of CCRF; see the overview articles published by Ahlberg et al [9] and Koornstra et al [11]. Guidelines state that if no primary association can be found for the persistence of fatigue with a somatic condition, behavioral interventions should also be considered [1]. The previously reported effects of nonpharmacological interventions on fatigue vary widely, as can be seen in the overview of recent meta-analyses in Table 1. Effect sizes tend to be higher when the intervention targets fatigue, and when increased fatigue was an inclusion criterion for the study. Not all studies that were included in the meta-analyses primarily targeted fatigue, therefore effect sizes might not be representative for nonpharmacological interventions that target fatigue.

    Table 1. Ten recent meta-analyses considering nonpharmacological interventions for cancer patients that included off-treatment fatigue.
    View this table

    Behavioral interventions are often based on energy balance models and/or stress coping models [12,19]. In energy balance models, CCRF is seen as a consequence of deconditioning and prolonged inactivity during cancer and its treatment. Secondary fatigue arises as a result of detraining and can lead to a downward spiral. In stress coping models, CCRF is conceptualized as a result of ineffective coping strategies and prolonged stress response [20]. Cognitive behavioral treatments that are based on these theories include physical activity interventions, exercise interventions [14,17,21,15,22], and mindfulness-based cognitive interventions [23-26] and have been shown to help reduce CCRF in previous studies [12,13]. However, all these interventions require the patient to travel to a health care facility, which can be a burden to the patient. Therefore, introducing effective interventions in a home-based setting could improve the health care options for this group.

    Potential Benefits of Internet Interventions

    Internet interventions offer advantages that cancer survivors who suffer from fatigue could especially benefit from. They have been found to be as effective as face-to-face therapies for a wide range of disorders, such as posttraumatic stress disorder, burnout or chronic stress, and depression [27-32]. Internet interventions have the ability to reach a wider range of patients compared to face-to-face interventions, especially severely fatigued patients, those with limited mobility, or patients in rural or even remote areas. Also, patients may benefit from the home-based setting of Internet interventions as these patients can practice more often, are less bound to the availability of care professionals, and can incorporate the intended behavioral change directly into their daily routine. Moreover, visiting a health care facility may no longer be desirable for some cancer survivors due to negative associations with the disease process or because they no longer want to be identified as a cancer patient and prefer the anonymity of their own environment.

    Internet Interventions for Fatigue


    In the Netherlands, to our best knowledge there are currently 3 Internet interventions that aim to reduce chronic fatigue: (1) an experimental mobile intervention aimed at changing physical activity behavior for participants with chronic fatigue syndrome [33]; (2) the Web-based mindfulness-based cognitive therapy “Minder Moe Bij Kanker” [34]; and (3) a Web-based cognitive behavior therapy for severely fatigued breast cancer survivors, which is the subject of the current CHANGE study (trial registration NTR4309) [35].

    This paper describes the design and analysis plan that studies the first 2 of these Internet interventions in a randomized controlled trial. Each of these 2 interventions is described below.

    Mobile Activity Management Intervention: Ambulant Activity Feedback Therapy

    The ambulant activity feedback therapy (AAF) is a mobile intervention that utilizes an ambulant activity coaching system, supported weekly by a physiotherapist through email [33]. The activity coaching system was developed by Roessingh Research and Development (Enschede, The Netherlands) and consists of a mobile phone and an accelerometer (Multimedia Appendix 1) that communicate through Bluetooth [33].

    In this intervention, the patient works to meet personal activity goals and subgoals that will be defined together with the therapist. The coaching system supports this process by showing real-time feedback about the accumulated activity of the patient relative to a personalized line of reference and tailored hourly feedback messages. Both the line of reference and the set of feedback messages of the activity coaching system can be adjusted by the therapist through a Web portal (see Figure 1 and Multimedia Appendix 2). Patients also have access to a Web portal where they can monitor their past personal activity records. Consequently, patients are expected to gain insight in their activity pattern and on how to increase or balance their daily activity in a way that improves their energy levels. More information about AAF is given in Wolvers and Vollenbroek-Hutten (in press) [36].

    Figure 1. Screenshot of the Web portal for the ambulant activity feedback therapy (Dutch).
    View this figure
    Web-Based Mindfulness-Based Cognitive Therapy

    Mindfulness-based cognitive therapy (MBCT) [37] adds elements of cognitive therapy to the mindfulness-based stress reduction program that was originally developed by John Kabat-Zinn [38]. The Helen Dowling Institute (Bilthoven, the Netherlands) developed a 9-week Web-based, therapist-guided, individual MBCT (eMBCT) specifically designed to reduce cancer-related fatigue [23]. On a personal Web page (see Figure 2 and Multimedia Appendix 3), each patient can download audio files of mindfulness exercises and read information about a specific mindfulness theme each week. Patients write down their experiences of following the mindfulness exercises in a log. On an agreed-upon day of the week, the therapist replies to this log, thereby guiding the patient through the program. It is hypothesized that by learning to raise awareness of their present experience nonjudgmentally and openly, the patient can become aware of potentially ineffective coping strategies that prolong stress and fatigue [39,40]. Patients learn to use a detached perspective as a skill to prevent the escalation of automatic negative thinking patterns. MBCT also teaches patients how to accept fatigue, physical limitations, or pain. The protocol of the eMBCT is discussed more extensively in the article by Bruggeman-Everts et al [34].

    Figure 2. Screenshot of the Web portal for eMBCT (Dutch).
    View this figure



    Our primary question is whether both interventions are effective in reducing fatigue. Therefore, the interventions will be compared to an active control group in a randomized controlled trial. The advantage of this design, as compared to a waiting-list control group, is that we can control for nonspecific influences of the trial, such as receiving attention. Also, we expect that in an active control group, fewer participants will drop out than in a waiting-list control group.

    Usually, results of interventions are presented in terms of an average improvement of the relevant outcome measure. However, practice shows that individuals benefit differently from interventions [41]. Therefore, the proposed trial will aim to identify individual fatigue trajectories, since that seems to be more informative and helpful in improving health care provisions for CCRF-patients than just presenting averages.


    To optimize interventions in terms of efficiency and effectiveness, treatment-specific and nonspecific working mechanisms should be identified that account for each intervention’s effect on fatigue. Knowledge about these mechanisms is an important prerequisite for improving the efficiency of interventions by shifting focus or shortening the intervention. Also, effectiveness can be increased by improving and tailoring the relevant items, subjects, or exercises, as well as improving the way these are embedded in the intervention. Therefore, the second objective of this study will be to identify the working mechanisms underpinning the interventions.

    By using a 3-armed randomized design, it is possible to study both treatment-specific (differentiating) and nonspecific working mechanisms. Also, by assessing important factors multiple times during the intervention, important time-specific information can be acquired.

    Effect Predictors

    Although we expect that, in general, both interventions are effective, personal factors, medical factors, and demographics may determine the effect that each intervention has on fatigue [41]. We do not expect all individuals to benefit similarly from the interventions. Therefore, studying potential predictors of each intervention’s effect will give us important input to inform both patients and caregivers and allow them to set reasonable expectations.

    CCRF has a multifactorial character (eg, physical, cognitive, motivation); therefore, studying the effect predictors of 2 theoretically differing interventions simultaneously might also reveal differentiating predictors for both therapies. By applying such knowledge carefully, the overall effectiveness of interventions that aim to reduce CCRF can be increased.



    A randomized controlled trial is performed including 3 parallel conditions: 2 experimental conditions (AAF and eMBCT) and a minimal intervention control condition. The intervention period is 9 weeks for all 3 conditions. Both experimental conditions are made as similar as possible in terms of time-investment and contact intensity with the therapist. Outcomes are self-reported and are Web-assessed at baseline (T0), 2 weeks post-intervention (T1), and at 6 months (T2) and 12 months (T3) after baseline. Figure 3 shows a schematic summary of the trial design.

    The baseline assessment consists of 3 time-points: (1) T0a, the assessment to check eligibility; (2) T0b, the main baseline assessment taken after the eligibility check and informed consent, but naive of condition; and (3) T0c, directly after randomization for assessing the participant’s credibility and expectancy about the condition. All participants are invited to fill out short questionnaires in weeks 1, 2, 3, 4, 6, and 9 (Mi) of the intervention period in order to study mediation of the interventions.

    After T2, patients in the control condition are offered 1 of the 2 experimental interventions, again in a research setting. Please note that the first 4 participants of this trial were randomized to 1 of the experimental conditions for the second semester, but to minimize dropout, all other patients will be allocated based on their own preference. During this second intervention period, these participants will again be assessed in weeks 1, 2, 3, 4, 6, and 9 (Mi’), the second week after the intervention (T1’), and 6 months after the second allocation (T2’). Participants in the control condition that are preferentially allocated to eMBCT after T2 do not wear the accelerometer during the second semester.

    The protocol allows delay within the intervention period of a maximum of 2 weeks in case of, for example, illness or holiday. For all participants, the duration of their participation is approximately 12 months. Additional qualitative feedback will be obtained through explorative interviews with a subset of participants in the experimental condition shortly after T1 or T1’.

    This trial was approved by the Twente Medical Ethical Committee (Enschede, the Netherlands) under number P12-26 and has been registered at The Netherlands National Trial Register under number NTR3483 [42].

    Figure 3. Flow chart of trial design. CS = cancer survivor; MD = medical doctor; PA = physical activity; T0(a-c)-T3 are the main assessments; Mi and Mi' represent assessments in week (i =) 1, 2, 3, 4, 6, and 9 of the intervention. For addressing the primary research questions of effectiveness, only data from the first semester will be used.
    View this figure

    Study Sample


    Participants are recruited in the Netherlands by advertisements in the newsletters of patient associations (both digital and print), on relevant websites, in regional newspapers, and through social media. Furthermore, participants also are recruited through oral presentations given to cancer patients and in other cancer-related seminars and symposia for patients, caregivers, or both.

    Social media and online advertising can be strong tools for reaching a large group of people [43] or even specific patients [44,45]. However, the sample might be younger, more highly educated, and might comprise more females compared to the Dutch CCRF population [46]. However, it will lead to a sample that represents the targeted population for such Internet interventions, and we are likely to include participants who would not have opted for therapy that includes traveling to a health care facility.


    The following criteria are used to check eligibility for participation in the trial:

    • Completion of a curative-intent treatment for cancer at least 3 months ago (checked by participant’s medical doctor). For this study, surgery, chemotherapy, radiotherapy, immunotherapy, and/or stem cell transplantation are considered treatment. However, hormonal therapy, the use of anti-inflammatories, and monitoring visits are not considered treatment for this study.
    • Patient has been suffering from severe fatigue for at least 3 months.
    • Patient scores 35 or higher on the fatigue severity subscale of the Checklist for Individual Strength (CIS).
    • Aged 19 years old or older.
    • At least 18 years old at disease onset.
    • Capable of reading and writing in the Dutch language and of using the Internet (implicit eligibility criterion accounted for during registration, but not checked explicitly).

    If patients meet 1 or more of the following criteria, they are excluded from participation:

    • Indication of current disease or tumor activity (checked by the participant’s medical doctor).
    • Current or former severe psychiatric morbidity, for example, major depression, psychosis, or schizophrenia (checked by the participant’s medical doctor).
    • Being dependent on a wheelchair for daily activity (self-report).
    • Recurrence of cancer during the course of the study (self-report).
    • Current substance abuse, except for smoking.
    • Previously attended the eMBCT of the Helen Dowling Institute.

    In addition to the mentioned exclusion criteria, please note that:

    • Mild depression is not an exclusion criterion. A score of 20 points or higher on the Hospital Anxiety and Depression Scale (HADS) during baseline is considered indicative of depression [47]. Therefore, if the patient scores 20 points or higher, he or she will be contacted by a psychologist from the Helen Dowling Institute to determine whether the participant has suicidal ideation or suffers from other severe psychiatric morbidity. A participant will only be excluded if, according to the involved psychologist, that is the case.
    • Comorbid somatic diseases—such as cardiovascular diseases, cerebrovascular diseases, diabetes, hypertension, and arthritis that are not treatable but are a possible cause of fatigue—are not exclusion criteria but will be registered during the study. Although this choice will probably lead to an underestimated effect size compared to studies that do exclude patients with comorbidities, we expect that such a sample will lead to a better representation of the CCRF population.
    • Participants are requested not to take part in any other therapy directed at overcoming fatigue during the study.
    • Data of participants who report pregnancy or recurrence of cancer during the course of the study will be excluded from analysis since the fatigue they experience cannot be considered to be of a chronic character according to our definitions. However, if requested, these patients will be allowed to finish the intervention.


    Participants apply for inclusion in the study at the project website [48,49].

    Informed Consent

    After online registration, participants receive the patient information and informed consent form by direct mail. They are requested to sign and return the informed consent in a prepaid envelope. Also, they receive a registration confirmation by email with login details for the participant’s Web portal on the project website. Participants are requested to complete assessment T0a as a check on eligibility. Also, the participant’s medical doctor is consulted to check 3 of the eligibility criteria: finished curative-intent treatment for cancer more than 3 months ago, no current signs of cancer activity, absence of current or former major psychiatric disease.


    If the eligibility-criteria are met, the researcher confirms the participant’s enrollment. Subsequently, the activity sensor is given to the participant and its setup is explained in a face-to-face meeting in the participant’s home or another mutually convenient location. The second baseline assessment starts (T0b), followed by randomization of the participant to 1 of the 3 conditions by a script embedded in the researchers’ Web portal and uses the random function of php (rand(1,3)) [50]. The researchers can neither influence nor predict the outcome of the randomization process. Subsequently, the researcher emails the participant about the outcome of randomization, requests the participant to complete the third baseline assessment (T0c), and assigns the participant to a therapist in case the participant has been randomized to an experimental condition. Participants who do not fill out T0c are considered as not being included. The allocation of a therapist is based on current availability of the therapists who are involved in the trial.

    Research Conditions

    Both experimental conditions are described in the Introduction and will be described more extensively in an article on eMBCT by Bruggeman-Everts et al [34], and a paper on the development of the AAF intervention by Wolvers and Vollenbroek-Hutten (in press) [36].

    Active Control Condition

    Patients who are assigned to the control condition receive weekly emails containing standard psycho-educational texts about CCRF in order to minimize the dropout rate, following the design of Postel et al [51]. An example of the information that is offered in this minimal intervention control condition is given in Multimedia Appendix 4 and overlaps completely with the information that is given during both experimental interventions. This condition controls for receiving information on CCRF and for being involved in eHealth research.

    Nonadherence and Withdrawal

    Participants who do not adhere to, or withdraw from, the study or the intervention are contacted by phone and asked for the reason for nonadherence or withdrawal. Participants who want to stop with the intervention are asked to complete a post-intervention assessment at T1 and follow-up assessments at T2 and T3. Participants who withdraw from the study are asked to answer the questions of the fatigue severity subscale of the CIS online or during a telephone conversation.


    All self-reported questionnaires are Web-assessed via a Web portal on the project website [48,49], developed by Roessingh Research and Development. Participants receive an email when an assessment becomes available and can log in to the Web portal to complete the questionnaires. During the intervention period, each assessment is available for 1 week, but can stay open longer if therapy is postponed due to, for example, illness or holiday. If a participant has not completed it within 6 days, he or she is reminded by email at least once to complete the questionnaire. Within each assessment, the questionnaires are grouped on the basis of importance and subject. Item sequences of the questionnaires for the mediating factors and outcome measures differ between the assessments. Personal data is stored separately from the research data. An overview of all the assessments is shown in Tables 2 and 3.

    Physical activity data is collected using the same device as that used for the ambulant activity feedback therapy: a 3D-accelerometer (ProMove 3D) combined with a mobile phone that collects the accelerometer data and sends it to a secured Web server at Roessingh Research and Development [52]. However, the mobile phone does not give feedback on activity, but does state whether the system is working properly and sends an error message if the connection to the sensor fails. Participants are reminded by email to wear the accelerometer on the day before the start of the week in which they will be using it.

    Outcome Measures


    Fatigue severity will be assessed with the CIS, which consists of 20 items that score on a 7-point Likert scale [53,54]. The CIS has 4 subscales (fatigue severity, motivation, concentration, and physical fatigue) of which the fatigue severity subscale will be used as the primary outcome (8 items, range: 8-56 points). The CIS has shown good discriminative validity in a working population [55], is sensitive to changes in the chronic fatigue syndrome population [56], and has previously been used with cancer survivors [7,57]. The CIS strongly resembles the Multidimensional Fatigue Inventory, which is often used in international studies [54]. Fatigue severity will be assessed at T0a, T0b, M3, M6, M9, T1, T2, and T3. However, the other 3 subscales will only be assessed at T0b, T1, T2, and T3.

    Mental Health

    Mental health will be assessed from the results of 2 questionnaires: the Positive and Negative Affect Scale (PANAS [52]) and the HADS [58], both of which are included in an item bank for cancer survivors [59]. The PANAS consists of 20 items that score on a 5-point Likert scale and has 2 subscales: positive and negative affect. The HADS consists of 14 items on a 4-point scale, has been validated for a Dutch-speaking population [60], and has previously been used to assess psychological distress in cancer patients [61]. Mental health will be assessed at T0a, T1, T2, and T3.

    Perceived Ability to Work

    The work ability score, which is assessed with the first question of the work ability index [62,63], will also be used as an outcome parameter. It asks: “Imagine that your working ability in the best period of your life is rated 10 points. How would you rate your working ability at the present moment?” It is assessed at T0b, T1, T2, and T3.

    Working hours and the level of absenteeism are assessed with questions from the Trimbos and iMTA questionnaire on costs associated with psychiatric illness (TiC-P) [64]). These will be assessed at T0b, T2, and T3.

    Table 2. Assessments of outcome measures and potentially mediating factors.
    View this table
    Table 3. Other assessments.
    View this table
    Mediating Factors

    Several categories of mediators will be considered: intervention-specific mediators for either eMBCT (eg, mindfulness, catastrophizing, and fear of cancer recurrence) or AAF (eg, physical activity, perceived physical activity, and self-efficacy on physical activity), and generic mediators (eg, sleep quality, sense of control over fatigue, credibility, expectancy, working alliance, and causal attributions). Furthermore, a distinction is made between primary and secondary mediating factors: primary factors are assessed at multiple occasions during the intervention in order to study the timely development of those factors; secondary factors are not assessed during the intervention. A complete overview of all assessments on mediating factors is given in Table 2.

    Demographics, Medical History, and Control Factors

    Several other factors are assessed, including demographics, medical history, and control factors. All are listed in Table 3.

    Analysis Plan


    SPSS software will be used for data management and Mplus [79], which is a latent variable modeling program, for the subsequent analyses. The exact versions of the software used will be reported in the future papers.

    Power Analyses

    The sample size for analyses for data relating to the primary objective has been calculated for a repeated measures analysis of variance: based on an alpha of .05, a minimal detectable effect size of f2=.15, and a power of .80, a total number of 55 participants [80] is required in each group to answer the primary research question of this study in a statistically valid manner.

    We expect to be able to include 330 eligible participants within a period of 2 years, based on a mean of 3.7 intakes per week for the eMBCT of the Helen Dowling Institute in 2011. An estimated attrition of 30% of the participants during both experimental interventions and 15% during the minimal intervention control condition [51] would leave us with 77 participants in each experimental group and 94 participants in the control group at T2. Again, we expect a dropout rate of 30% during the second semester. Such a dropout would leave a total of 110 participants completing each experimental intervention. Ten percent of the participants may have to be excluded from the analyses because of recurrence or diagnosis of metastasis. That would result in 198 participants that complete the full trial. We expect that this number will be enough for testing the 6 mediating factors or effect predictors: A classical, conservative power calculation (analysis of variance for testing 6 mediators or effect predictors with an intermediate effect size (f2=.08), corrected according to Bonferroni (alpha=.05/6), and at a power of .80 [81]) would result in approximately 254 participants being needed. We expect that the actual power when including 198 participants, and not the required 254 participants, will be great enough to detect up to 6 mediators or effect predictors with the use of Bayesian statistics [76]. Bayesian statistics allow analysis on small sample sizes [76,77], as more power can be generated with the use of prior information which is incorporated in the model that is being tested. Various papers describe comparisons between traditional null hypothesis testing and Bayesian estimation [82-85]. For this study, prior knowledge is available for many parameters, such as the effects of mindfulness in cancer survivors [23,25,66,86] and the role of working alliance in online interventions [87]. Examples of these methods can be found in both applied psychology and social science articles [88-92].

    Missing Data Handling

    Missing data will be analyzed considering their pattern and randomness following guidelines proposed by Schafer and Graham [93]. Bias due to systematic missing data will be managed according to guidelines proposed by Asendorpf et al [94].


    Quantitative analyses will be conducted on an intention-to-treat basis. A flow diagram following the CONSORT guidelines will be included. Descriptive statistics will be calculated and presented. Independent samples’ t-tests and χ2 tests will be performed to check for baseline differences between the respective experimental conditions and the control condition with respect to demographic variables (eg, family status, age, gender, and level of education), time since end of treatment, and baseline levels of the outcome variables. If we find statistically significant differences in the mean of fatigue severity across baseline descriptives, dummy variables will be added to the model as covariates to control for these differences.

    Core Analysis

    Five steps will be taken to evaluate the effectiveness of both interventions, which are explained here in a generic way. The specific hypotheses on the effectiveness of the interventions in our study are shown in Textbox 1.

    Textbox 1. Hypotheses on effectiveness.
    View this box
    Step 1

    Overall effectiveness will be tested in an intention-to-treat analysis by a multiple group latent growth model [79] using data from the first semester. This technique allows individuals to have an individual growth trajectory over time and compensates for missing data in an elegant way.

    Since different growth patterns are expected for the pre-intervention period, the intervention period, and the post-intervention period, we will apply piecewise growth modeling so that a slope factor will be estimated for each of the 3 periods (Figure 4). Initial intercepts will be configured to represent the T0b score. This intercept and the pre-intervention slope factor will be constrained to be equal between all 3 conditions (and this assumption will be checked), whereas the subsequent slope factors will be estimated separately for the 3 conditions.

    The fit of the piecewise model will be compared with a quadratic model. In the quadratic model, the entire first semester is modeled with 1 slope factor and 1 quadratic factor for each of the 3 conditions and an intercept that represents T0b and is constrained similarly to the piecewise model.

    Both models will be run both with and without using time-varying loadings in order to check whether corrections should be made for differences in timings between the questionnaires. Growth factor estimates and model fits for all 4 models will be reported (Table 5).

    Neither the participants nor the researchers (FBE and MW) are blinded to allocation. Therefore, an independent statistician (RvdS) who is blind to the allocation will test the primary hypothesis.

    The same procedure will be followed for the secondary outcomes, except that the initial intercept of mental health will represent T0a, rather than T0b, because T0b does not include an assessment of mental health.

    Results of frequentist analyses will be reported by P-values (significant in case <.05) and with 95% confidence intervals. Parameter estimates of models by means of Bayesian estimators will be reported with 95% central credibility intervals.

    Figure 4. Simplified representation of a piecewise linear latent growth model, with latent intercept factor (I), latent slope factors preintervention (S(pre)), during the intervention (S(int)), and postintervention (S(post)), and 7 indicators Y. Error terms, correlation coefficients, and covariances are left out.
    View this figure
    Table 5. Growth factor estimates of 4 different latent growth models.
    View this table
    Step 2

    The effect size of both experimental interventions will be calculated according to recommendations in Feingold (2009) [95] for both primary and secondary outcomes.

    Step 3

    The proportion of participants who make clinically relevant progress on the primary outcome will be calculated for all 3 conditions; again, in an intention-to-treat analysis. Percentages and standard deviations of the reliable change index will be presented.

    Step 4

    A latent growth model will be built of the primary outcome, in which the outcome measures that have been measured at T3 will also be included, as distal outcomes of changes in the primary outcome during the first semester.

    Step 5

    A growth mixture model (GMM) will be used to further explore differences between individuals, and more specifically to identify subpopulations (latent classes) with homogeneous growth trajectories of the primary outcome within the experimental groups. The Bayesian information criterion will be used for model selection [96].

    If convergence considerations allow, this model will be adjusted to allow covariance of the growth factors in order to acknowledge individual variation around the estimated growth trajectories. The trace plots will be inspected to check whether the models have converged to global solutions and a set of diverse starting values will be used. For more information on these analyses, we refer to an introduction to GMM and latent class growth analysis by Jung and Wickrama [97] and examples of similar analyses in the field of Internet interventions [98] and cancer patients [41].


    The analysis of the mediators of the experimental conditions can be roughly subdivided into two steps: first analyze the primary factors individually for their longitudinal correlations with the outcome (Step 6), then combine the relevant factors in a multivariate analysis (Step 7). The specific hypothesis on the mediating factors of the interventions in this study are shown in Textbox 2.

    Step 6

    For analyzing the mediators of the experimental conditions, first we want to see whether there is a correlation between the growth trajectories of our outcome parameter and the potential mediator over time. The hypotheses considering mediators are shown in Textbox 2. The combined data from the participants in the first semester and data from the preferentially assigned participants in the second semester will be used.

    The following subhypotheses will be tested for each primary mediator (these are also shown in Figure 5):

    1. Is the growth of the primary outcome (Sy) for the entire study population correlated with growth of the potential mediator (Sz)?
    2. Is such correlation independent of group?
    3. Does the potential mediator change over time in the specific group, so is the slope factor (Sz) substantially unequal to zero?
    4. Is the slope factor in the specific group substantially greater than the slope factors in the other groups?

    In these 4 subhypotheses, the first is congruent with testing the “conceptual theory” in classical mediation analysis, and subhypothesis 3 with testing the “action theory.” If subhypotheses 1-4 all are true, the factor will be considered a specific mediator for that intervention. If subhypotheses 1, 2, and 3—but not 4—are true, the factor will be considered a general mediator for fatigue severity. If either subhypothesis 1 (conceptual theory) or 3 (action theory) is false, the factor will not be considered a mediator.

    Textbox 2. Hypotheses on mediators.
    View this box
    Figure 5. Simplified representation of a correlated growth model in which Iy and Sy represent the intercept and slope factors of the latent growth model of the outcome parameter, and Iz and Sz represent the latent growth factors of the mediator. H1-4 represent the 4 subhypotheses of Step 6. All indicators have been left out for clarity.
    View this figure
    Step 7

    The next step in studying potential working mechanisms is a single-step, multiple-mediation analysis using structural equation modeling [104-106]. By estimating such a model, we expect to obtain a comprehensive model for all the working mechanisms of the intervention. It should be noted that this model assumes that an intervention works in the same way for all participants in a particular group [107]. Again, data from both semesters will be used.

    A separate model will be tested for each intervention. Each model will have the following paths (Figure 6), where X=independent variable (1/0 for specific intervention vs control group), Y=outcome variable (difference score T2-T0b of the primary outcome measure), and Z=mediator:

    • a: X regressed on Z.
    • b: Z regressed on Y;
    • c’: direct effect of X on Y.

    For each experimental intervention, the starting model will consist of all the significant primary mediators of Step 6 that have also been assessed in the control group. In other words, the factors that have shown to be mediators in the correlated growth model will be the starting point for this model. The models will then be complemented with the secondary mediating factors described in Textbox 2. Mediating factors for which the indirect effect (a × b) is insignificant will be removed stepwise, after which a final model will be created.

    Model fit, standardized path coefficients—including indirect effects—and the total effect of at least the first and final models will be reported with 95% confidence intervals.

    Figure 6. Multiple mediation model with independent variable (X), dependent variable (Y), and 2 mediators (Z). Direct effect (c’) and indirect effects (through a x b) are shown.
    View this figure
    Effect Predictors

    Two complementing approaches for analyzing the effect predictors are addressed in steps 8 and 9 of this analysis plan. The specific hypotheses on the effect predictors for both interventions in this particular study are shown in Textbox 3.

    Textbox 3. Hypotheses on effect predictors.
    View this box
    Step 8

    To find out which participants benefit most from each intervention, the final model of fatigue severity of Step 1 will be extended with potential effect predictors (Textbox 3) that load on the latent growth factors “linear slope” (the “post randomization” linear slope in case of the piecewise model) and, if applicable, “quadratic slope.” As the regression coefficients of all potential effect predictors on the development of fatigue severity will be freely estimated across the 3 intervention groups, this is also called a moderation effect of intervention.

    Factor loadings of all hypothesized effect predictors will be reported. Those with the highest loadings will be compared between the conditions in order to find differential effect predictors.

    Step 9

    To identify common effect predictors of homogeneous subpopulations within the heterogeneous population, rather than identifying effect predictors for individual growth patterns, the final step will consist of regressing predictors on latent classes. Therefore, the final model of Step 5, the GMM, will be extended. Again, several potential effect predictors will be regressed onto this model, but this time on the latent class factor, instead of on the latent growth factors. The 3-step procedure proposed by Vermunt [109] will be used for model selection. This step will be carried out separately for each experimental condition.

    Factor loadings of all hypothesized effect predictors will be reported. Those with the highest loadings will be compared between the conditions in order to find differential effect predictors.


    Recruitment for the trial started in March 2013 and is expected to continue until April 2015. No major changes have been made to the protocol. However, due to an error in the randomization algorithm between January 14, 2014, and July 15, 2014, allocation was dependent on the number of participants who were allocated at once. This in turn was completely random. Consequently, 10 participants were allocated directly to the AAF group; 4 other participants were divided equally between the 2 experimental interventions; and 15 accounts (of which, 1 was a dummy account) were equally divided between the 3 groups. None of the researchers were aware of this error, as this allocation could very well have simply been the result of the “roll the dice” scenario that should have been applied. How many participants were allocated at once was not the subject of the researchers’ decision-making. Therefore, we argue that allocation has still been random and, accordingly, data for all considered participants will be processed as originally planned.

    At the time of this writing in January 2015, 269 patients have registered at the project website. Of these, 111 have been officially included in the study, 50 were excluded from participation, and 35 withdrew before their eligibility was checked. The remaining patients are still in the enrollment phase. The main reason for exclusion so far has been a score lower than 35 on the CIS fatigue severity subscale (60%). Furthermore 11% did not meet the psychiatric stability requirement, 8% were younger than 18 at the time of cancer diagnosis, and 8% were still receiving cancer treatment.

    Current group sizes as of January 2015 for participants in the first semester are 36 (AAF), 24 (eMBCT), and 32 (control). However, 19 participants have not yet been randomized. Initial responses to the primary research question are expected to be available by the end of 2015.


    Principal Findings

    This paper has described the design, hypotheses and analysis plan of a randomized controlled trial in order to study the effectiveness, mediation, and effect predictors of 2 Internet-based interventions for CCRF. Although recruitment and inclusion have already started, publishing the analysis plan is of great value because it will help to prevent outcome reporting bias [110] and adds validity information to the final studies [111].

    By using multiple assessments during the intervention, the proposed trial design is suitable for studying the chronological development of both potential mediators and fatigue. That has 2 main advantages. Firstly, the data will be suitable for analyses that allow for variation in the individual fatigue trajectories. We do not expect that either of the interventions that have been included in the trial will be beneficial for all participants: our study sample will be highly heterogeneous considering for example tumor and treatment types. Therefore, the analyses on individual growth trajectories can acknowledge that expectation and test that hypothesis. This will substantiate the interpretation of the results on effectiveness and will be an important first step in identifying what works for whom. Secondly, this study design enables us to use a fully longitudinal mediation analysis, at least for the most important factors, rather than using indirect effects analysis in cross-sectional mediation analysis.

    Another important feature of the proposed design is that by comparing 2 different interventions with an active control group, therapy-specific elements of the interventions can be distillated from the data acquired during this trial. This advantage counts for both the effect predictors and the mediators. Knowledge about such differentiating factors can and should be used to better inform patients with CCRF and to improve allocation of patients with CCRF to suitable interventions. As a result, an increase in the overall effectiveness of relevant interventions can be established.

    In this paper, we have presented the trial design, our hypotheses, and a detailed analysis plan. In accordance with good clinical practice, and to avoid outcome reporting bias, this paper was submitted before any of the data was analyzed. All methods are now openly predetermined, therefore any future publication describing this trial can be valued reliably on its quality.


    A limitation of the current paper is that for most instruments, this paper does not include information on its properties or a thorough rationale for its choice. More extensive information on the actual instruments will be reported in subsequent papers on the results of the various research questions posed in this trial.


    Given the growing number of patients suffering from CCRF, the availability of effective Internet interventions potentially strengthens current health care for this population substantially. We have proposed a design to study 2 Internet interventions in order to gain insight into their effectiveness, mediators, and effect predictors, which fully acknowledges differences between individual patients and differences in the way they respond to each intervention. Results on the effectiveness and mediators will give useful information for improving both the quality and availability of such interventions. Also, identifying effect predictors for positive intervention effects will improve the referral of patients to relevant interventions. By presenting our hypotheses and analytic strategy before completion of data collection, this paper is a first step in carefully reporting on this comprehensive trial.


    The project “Fitter na kanker” in which this trial will be carried out is financed by the “Alpe d’HuZes/KWF-fonds” (project number 2011-5264) and is a collaboration between the Helen Dowling Institute and Roessingh Research and Development. We thank Richard Evering for his contributions to the grant proposal.

    Authors' Contributions

    Study conception and design: all authors; acquisition of data: FE and MW; analysis and interpretation of data: not applicable; drafting of manuscript: MW; critical revision: all authors.

    Conflicts of Interest

    None declared.

    Multimedia Appendix 1

    Picture of the ProMove accelerometer and mobile phone app with feedback.

    PNG File, 1MB

    Multimedia Appendix 2

    Additional screenshot of the therapist Web portal for ambulant activity feedback therapy (Dutch).

    PNG File, 82KB

    Multimedia Appendix 3

    Additional screenshot of the patient Web portal for eMBCT (Dutch).

    JPG File, 130KB

    Multimedia Appendix 4

    Example of an information letter for the minimal intervention control condition (Dutch).

    PNG File, 110KB

    Multimedia Appendix 5

    CONSORT E-HEALTH checklist V1.6.1 [112].

    PDF File (Adobe PDF File), 1021KB


    1. Mock V, Atkinson A, Barsevick A, Cella D, Cimprich B, Cleeland C, National Comprehensive Cancer Network. NCCN Practice Guidelines for Cancer-Related Fatigue. Oncology (Williston Park) 2000 Nov;14(11A):151-161. [Medline]
    2. Ploos van Amstel Floortje K, van den Berg Sanne W, van Laarhoven Hanneke W M, Gielissen Marieke F M, Prins JB, Ottevanger PB. Distress screening remains important during follow-up after primary breast cancer treatment. Support Care Cancer 2013 Aug;21(8):2107-2115. [CrossRef] [Medline]
    3. Duijts Saskia F A, van Egmond Martine P, Spelten E, van MP, Anema JR, van der Beek Allard J. Physical and psychosocial problems in cancer survivors beyond return to work: a systematic review. Psychooncology 2014 May;23(5):481-492. [CrossRef] [Medline]
    4. Curt GA, Breitbart W, Cella D, Groopman JE, Horning SJ, Itri LM, et al. Impact of cancer-related fatigue on the lives of patients: new findings from the Fatigue Coalition. Oncologist 2000;5(5):353-360 [FREE Full text] [Medline]
    5. Meulepas JM, Kiemeney LALM, Benraadt J. Trends en prognoses van incidentie, sterfte, overleving en prevalentie. In: Van Driel F, Knoop L, editors. Kanker in Nederland tot 2020: trends en prognoses (Dutch). Oisterwijk: VDB Almedeon; 2011:65-65.
    6. Ryan JL, Carroll JK, Ryan EP, Mustian KM, Fiscella K, Morrow GR. Mechanisms of cancer-related fatigue. Oncologist 2007 Jan;12 Suppl 1:22-34 [FREE Full text] [CrossRef] [Medline]
    7. Servaes P, Gielissen M F M, Verhagen S, Bleijenberg G. The course of severe fatigue in disease-free breast cancer patients: a longitudinal study. Psychooncology 2007 Sep;16(9):787-795. [CrossRef] [Medline]
    8. Goedendorp MM, Gielissen Marieke F M, Verhagen Constans A H H V M, Bleijenberg G. Development of fatigue in cancer survivors: a prospective follow-up study from diagnosis into the year after treatment. J Pain Symptom Manage 2013 Feb;45(2):213-222. [CrossRef] [Medline]
    9. Ahlberg K, Ekman T, Gaston-Johansson F, Mock V. Assessment and management of cancer-related fatigue in adults. Lancet 2003 Aug 23;362(9384):640-650. [CrossRef] [Medline]
    10. Cella D, Davis K, Breitbart W, Curt G, Fatigue C. Cancer-related fatigue: prevalence of proposed diagnostic criteria in a United States sample of cancer survivors. J Clin Oncol 2001 Jul 15;19(14):3385-3391. [Medline]
    11. Koornstra Rutger H T, Peters M, Donofrio S, van den Borne Ben, de Jong Floris A. Management of fatigue in patients with cancer -- a practical overview. Cancer Treat Rev 2014 Jul;40(6):791-799. [CrossRef] [Medline]
    12. Jacobsen PB, Donovan KA, Vadaparampil ST, Small BJ. Systematic review and meta-analysis of psychological and activity-based interventions for cancer-related fatigue. Health Psychol 2007 Nov;26(6):660-667 [FREE Full text] [CrossRef] [Medline]
    13. Kangas M, Bovbjerg DH, Montgomery GH. Cancer-related fatigue: a systematic and meta-analytic review of non-pharmacological therapies for cancer patients. Psychol Bull 2008 Sep;134(5):700-741. [CrossRef] [Medline]
    14. Speck RM, Courneya KS, Mâsse LC, Duval S, Schmitz KH. An update of controlled physical activity trials in cancer survivors: a systematic review and meta-analysis. J Cancer Surviv 2010 Jun;4(2):87-100. [CrossRef] [Medline]
    15. Brown JC, Huedo-Medina TB, Pescatello LS, Pescatello SM, Ferrer RA, Johnson BT. Efficacy of exercise interventions in modulating cancer-related fatigue among adult cancer survivors: a meta-analysis. Cancer Epidemiol Biomarkers Prev 2011 Jan;20(1):123-133 [FREE Full text] [CrossRef] [Medline]
    16. Duijts Saskia F A, Faber MM, Oldenburg Hester S A, van BM, Aaronson NK. Effectiveness of behavioral techniques and physical exercise on psychosocial functioning and health-related quality of life in breast cancer patients and survivors--a meta-analysis. Psychooncology 2011 Feb;20(2):115-126. [CrossRef] [Medline]
    17. Cramp F, Byron-Daniel J. Exercise for the management of cancer-related fatigue in adults. Cochrane Database Syst Rev 2012 Jan;11:CD006145. [CrossRef] [Medline]
    18. Tomlinson D, Diorio C, Beyene J, Sung L. Effect of exercise on cancer-related fatigue: a meta-analysis. Am J Phys Med Rehabil 2014 Aug;93(8):675-686. [CrossRef] [Medline]
    19. Mock V. Cancer-related fatigue. In: Given CW, Given BA, Champion V, Kozachik S, Devoss DN, editors. Evidence-based cancer care and prevention: behavioral interventions. New York: Springer Publishing Company; 2003.
    20. Gélinas C, Fillion L. Factors related to persistent fatigue following completion of breast cancer treatment. Oncol Nurs Forum 2004;31(2):269-278. [CrossRef] [Medline]
    21. Barsevick AM, Dudley W, Beck S, Sweeney C, Whitmer K, Nail L. A randomized clinical trial of energy conservation for patients with cancer-related fatigue. Cancer 2004 Mar 15;100(6):1302-1310 [FREE Full text] [CrossRef] [Medline]
    22. Bourke L, Homer KE, Thaha MA, Steed L, Rosario DJ, Robb KA, Taylor Stephanie J C. Interventions for promoting habitual exercise in people living with and beyond cancer. Cochrane Database Syst Rev 2013 Jan;9. [CrossRef] [Medline]
    23. Van der Lee ML, Garssen B. Mindfulness-based cognitive therapy reduces chronic cancer-related fatigue: a treatment study. Psychooncology 2012 Mar;21(3):264-272. [CrossRef] [Medline]
    24. Lengacher CA, Johnson-Mallard V, Post-White J, Moscoso MS, Jacobsen PB, Klein TW, et al. Randomized controlled trial of mindfulness-based stress reduction (MBSR) for survivors of breast cancer. Psychooncology 2009 Dec;18(12):1261-1272. [CrossRef] [Medline]
    25. Carlson LE, Garland SN. Impact of mindfulness-based stress reduction (MBSR) on sleep, mood, stress and fatigue symptoms in cancer outpatients. Int J Behav Med 2005 Jan;12(4):278-285. [CrossRef] [Medline]
    26. Shapiro SL, Bootzin RR, Figueredo AJ, Lopez AM, Schwartz GE. The efficacy of mindfulness-based stress reduction in the treatment of sleep disturbance in women with breast cancer: an exploratory study. J Psychosom Res 2003 Jan;54(1):85-91. [Medline]
    27. Ruwaard J, Schrieken B, Schrijver M, Broeksteeg J, Dekker J, Vermeulen H, et al. Standardized web-based cognitive behavioural therapy of mild to moderate depression: a randomized controlled trial with a long-term follow-up. Cogn Behav Ther 2009 Dec;38(4):206-221. [CrossRef] [Medline]
    28. Ruwaard J, Lange A, Bouwman M, Broeksteeg J, Schrieken B. E-mailed standardized cognitive behavioural treatment of work-related stress: a randomized controlled trial. Cogn Behav Ther 2007;36(3):179-192. [CrossRef] [Medline]
    29. Lange A, Rietdijk D, Hudcovicova M, van de Ven Jean-Pierre, Schrieken B, Emmelkamp Paul M G. Interapy: a controlled randomized trial of the standardized treatment of posttraumatic stress through the internet. J Consult Clin Psychol 2003 Oct;71(5):901-909. [CrossRef] [Medline]
    30. Knaevelsrud C, Maercker A. Long-term effects of an internet-based treatment for posttraumatic stress. Cogn Behav Ther 2010 Jan;39(1):72-77. [CrossRef] [Medline]
    31. Knaevelsrud C, Maercker A. Internet-based treatment for PTSD reduces distress and facilitates the development of a strong therapeutic alliance: a randomized controlled clinical trial. BMC Psychiatry 2007;7:13 [FREE Full text] [CrossRef] [Medline]
    32. Andersson G, Cuijpers P. Internet-based and other computerized psychological treatments for adult depression: a meta-analysis. Cogn Behav Ther 2009;38(4):196-205. [CrossRef] [Medline]
    33. Evering R, Drossaert C, Vollenbroek-Hutten MMR. Ambulatory feedback at daily physical activities in treatment of chronic fatigue syndrome: a randomized controlled trial. In: Ambulatory feedback at daily physical activity patterns - a treatment for the chronic fatigue syndrome in the home environment?. Enschede: Gildeprint Drukkerijen; 2013:123-148.
    34. Bruggeman-Everts FZ, Van der Lee ML, De Jager Meezenbroek E. Web-based individual Mindfulness-Based Cognitive Therapy for cancer-related fatigue - A pilot study. Internet Interventions 2015 May;2(2):200-213. [CrossRef]
    35. Nederlands trial register. 2013. Web-based cognitive behaviour therapy for severely fatigued breast cancer survivors: CHANGE study   URL: [accessed 2014-10-22] [WebCite Cache]
    36. Wolvers MDJ, Vollenbroek-Hutten MMR. An mHealth Intervention Strategy for Physical Activity Coaching in Cancer Survivors. : CEUR Workshop Proceedings; 2015 Presented at: International Workshop on Personalisation and Adaptation in Technology for Health; 2015 June 30; Dublin   URL:
    37. Segal Z, Williams M, Teasdale J. Mindfulness-Based Cognitive Therapy for Depression: A New Approach to Preventing Relapse. New York: The Guilford Press; 2001.
    38. Kabat-Zinn J, Lipworth L, Burney R, Sellers W. Four-year follow-up of a meditation-based program for the self-regulation of chronic pain: treatment outcomes and compliance. Clin J Pain 1987;3(3):60.
    39. Bishop SR, Lau M, Shapiro S, Carlson L, Anderson ND, Carmody J, et al. Mindfulness: a proposed operational definition. Clinical Psychology: Science and Practice 2004 Aug 01;11(3):230-241. [CrossRef]
    40. Kabat-Zinn J, Lipworth L, Burney R. The clinical use of mindfulness meditation for the self-regulation of chronic pain. J Behav Med 1985 Jun;8(2):163-190. [Medline]
    41. Zhu L, Schroevers MJ, Van der Lee ML, Garssen B, Stewart RE, Sanderman R, et al. Trajectories of personal control in cancer patients receiving psychological care. Psychooncology 2015 May;24(5):556-563. [CrossRef] [Medline]
    42. Everts FZ, Wolvers MDJ. Nederlands trial register. 2012. Investigating two home-based interventions for people suffering from chronic fatigue after cancer   URL: [accessed 2014-02-20] [WebCite Cache]
    43. Quach S, Pereira JA, Russell ML, Wormsbecker AE, Ramsay H, Crowe L, et al. The good, bad, and ugly of online recruitment of parents for health-related focus groups: lessons learned. J Med Internet Res 2013 Jan;15(11):e250 [FREE Full text] [CrossRef] [Medline]
    44. Heffner JL, Wyszynski CM, Comstock B, Mercer LD, Bricker J. Overcoming recruitment challenges of web-based interventions for tobacco use: the case of web-based acceptance and commitment therapy for smoking cessation. Addict Behav 2013 Oct;38(10):2473-2476 [FREE Full text] [CrossRef] [Medline]
    45. Morgan AJ, Jorm AF, Mackinnon AJ. Internet-based recruitment to a depression prevention intervention: lessons from the Mood Memos study. J Med Internet Res 2013;15(2):e31 [FREE Full text] [CrossRef] [Medline]
    46. Stopponi MA, Alexander GL, McClure JB, Carroll NM, Divine GW, Calvi JH, et al. Recruitment to a randomized web-based nutritional intervention trial: characteristics of participants compared to non-participants. J Med Internet Res 2009 Jan;11(3):e38 [FREE Full text] [CrossRef] [Medline]
    47. Le FP, Devereux J, Smith S, Lawrie SM, Cornbleet M. Screening for psychiatric illness in the palliative care inpatient setting: a comparison between the Hospital Anxiety and Depression Scale and the General Health Questionnaire-12. Palliat Med 1999 Sep;13(5):399-407. [Medline]
    48. Fitter na kanker. 2013. First project Web site   URL: [accessed 2014-02-26] [WebCite Cache]
    49. Fitter na kanker. 2014. Revised project Web site   URL: [accessed 2014-06-05] [WebCite Cache]
    50. PHP Manual. Function reference 'rand'   URL: [accessed 2015-01-28] [WebCite Cache]
    51. Postel MG, de Haan Hein A, ter Huurne Elke D, Becker ES, de Jong Cor A J. Effectiveness of a web-based intervention for problem drinkers and reasons for dropout: randomized controlled trial. J Med Internet Res 2010 Jan;12(4):e68 [FREE Full text] [CrossRef] [Medline]
    52. Op den Akker H, Tabak M, Marin-Perianu M, Huis in ’t Veld R, Jones VM, Hofs D, et al. Development and evaluation of a sensor-based system for remote monitoring and treatment of chronic diseases - the continuous care & coaching platform. In: Proc 6th Int Symp eHealth Serv Technol.: SciTePress; 2012 Jul Presented at: Sixth international symposium on e-health services and technologies; 3-4 July 2012; Geneva p. 19-27   URL:
    53. Vercoulen JH, Swanink CM, Fennis JF, Galama JM, van der Meer J W, Bleijenberg G. Dimensional assessment of chronic fatigue syndrome. J Psychosom Res 1994 Jul;38(5):383-392. [Medline]
    54. Dittner AJ, Wessely SC, Brown RG. The assessment of fatigue: a practical guide for clinicians and researchers. J Psychosom Res 2004 Feb;56(2):157-170. [CrossRef] [Medline]
    55. Beurskens AJ, Bültmann U, Kant I, Vercoulen JH, Bleijenberg G, Swaen GM. Fatigue among working people: validity of a questionnaire measure. Occup Environ Med 2000 May;57(5):353-357 [FREE Full text] [Medline]
    56. Prins JB, Bleijenberg G, Bazelmans E, Elving LD, de Boo T M, Severens JL, van der Wilt G J, van der Meer J W. Cognitive behaviour therapy for chronic fatigue syndrome: a multicentre randomised controlled trial. Lancet 2001 Mar 17;357(9259):841-847. [CrossRef] [Medline]
    57. Servaes P, Verhagen S, Bleijenberg G. Determinants of chronic fatigue in disease-free breast cancer patients: a cross-sectional study. Ann Oncol 2002 Apr;13(4):589-598 [FREE Full text] [Medline]
    58. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983 Jun;67(6):361-370. [Medline]
    59. Smith AB, Armes J, Richardson A, Stark DP. Psychological distress in cancer survivors: the further development of an item bank. Psychooncology 2013 Feb;22(2):308-314. [CrossRef] [Medline]
    60. Spinhoven P, Ormel J, Sloekers PP, Kempen GI, Speckens AE, Van Hemert AM. A validation study of the Hospital Anxiety and Depression Scale (HADS) in different groups of Dutch subjects. Psychol Med 1997 Mar;27(2):363-370. [Medline]
    61. Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res 2002 Feb;52(2):69-77. [Medline]
    62. Van den Berg TIJ, Elders LAM, De Zwart BCH, Burdorf A. The effects of work-related and individual factors on the Work Ability Index: a systematic review. Occup Environ Med 2009 Apr;66(4):211-220. [CrossRef] [Medline]
    63. Ilmarinen J. The Work Ability Index (WAI). Occupational Medicine 2006 Oct 17;57(2):160-160. [CrossRef]
    64. Hakkaart-Van Roijen L. Manual Trimbos/iMTA Questionnaire for Costs Associated with Psychiatric Illness. 2002.   URL: http:/​/www.​​fileadmin/​ASSETS/​bmg/​english/​iMTA/​Publications/​Manuals___Questionnaires/​Manual_TIC_P_Versie_Engels_september_2010.​pdf [accessed 2015-06-16] [WebCite Cache]
    65. Walach H, Buchheld N, Buttenmüller V, Kleinknecht N, Schmidt S. Measuring mindfulness—the Freiburg Mindfulness Inventory (FMI). Personality and Individual Differences 2006 Jun;40(8):1543-1555. [CrossRef]
    66. Foley E, Baillie A, Huxter M, Price M, Sinclair E. Mindfulness-based cognitive therapy for individuals whose lives have been affected by cancer: a randomized controlled trial. J Consult Clin Psychol 2010 Feb;78(1):72-79. [CrossRef] [Medline]
    67. Cox CMM. Quality of sleep in hospital settings. Maastricht: Universitaire Pers; 1992. The Subjective Sleep Quality Scale   URL: http:/​/digitalarchive.​​fedora/​get/​guid:d6583a11-c0b9-4da6-aca4-3ddacc2ce799/​ASSET1 [accessed 2015-06-16] [WebCite Cache]
    68. Devilly GJ, Borkovec TD. Psychometric properties of the credibility/expectancy questionnaire. J Behav Ther Exp Psychiatry 2000 Jun;31(2):73-86. [Medline]
    69. Vervaeke G, Vertommen H. De werkalliantievragenlijst. Gedragstherapie 1996;29:139-144 Dutch.
    70. Busseri MA, Tyler JD. Interchangeability of the Working Alliance Inventory and Working Alliance Inventory, Short Form. Psychol Assess 2003 Jun;15(2):193-197. [Medline]
    71. Bandura A. Guide for constructing self-efficacy scales. In: Kirshner B, editor. Self-efficacy beliefs adolesc. Charlotte: Information Age Publishing; 2006:307-337.
    72. Rodgers WM, Sullivan MJL. Task, Coping, and Scheduling Self-Efficacy in Relation to Frequency of Physical Activity. J Appl Social Pyschol 2001 Apr;31(4):741-753. [CrossRef]
    73. Rodgers WM, Wilson PM, Hall CR, Fraser SN, Murray TC. Evidence for a multidimensional self-efficacy for exercise scale. Res Q Exerc Sport 2008 Jun;79(2):222-234. [CrossRef] [Medline]
    74. Donovan KA, Small BJ, Andrykowski MA, Munster P, Jacobsen PB. Utility of a cognitive-behavioral model to predict fatigue following breast cancer treatment. Health Psychol 2007 Jul;26(4):464-472 [FREE Full text] [CrossRef] [Medline]
    75. Jacobsen PB, Andrykowski MA, Thors CL. Relationship of catastrophizing to fatigue among women receiving treatment for breast cancer. J Consult Clin Psychol 2004 Apr;72(2):355-361 [FREE Full text] [CrossRef] [Medline]
    76. Laubmeier KK, Zakowski SG, Bair JP. The role of spirituality in the psychological adjustment to cancer: a test of the transactional model of stress and coping. Int J Behav Med 2004 Jan;11(1):48-55. [CrossRef] [Medline]
    77. Zimet GD, Powell SS, Farley GK, Werkman S, Berkoff KA. Psychometric characteristics of the Multidimensional Scale of Perceived Social Support. J Pers Assess 1990;55(3-4):610-617. [CrossRef] [Medline]
    78. Paulhus D. Measurement and control of response bias. In: Robinson J, Shaver P, Wrightsman L, editors. Meas Personal Soc Psychol Attitudes. San Diego: Academic Press; 1991:17-59.
    79. Muthen LK, Muthen BO. Mplus users guide. Sixth edition. 2010.   URL: [accessed 2015-06-16] [WebCite Cache]
    80. Faul F, Erdfelder E, Lang A, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007 May;39(2):175-191. [Medline]
    81. Faul F, Erdfelder E, Buchner A, Lang A. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 2009 Nov;41(4):1149-1160. [CrossRef] [Medline]
    82. Kuiper RM, Hoijtink H. Comparisons of means using exploratory and confirmatory approaches. Psychol Methods 2010 Mar;15(1):69-86. [CrossRef] [Medline]
    83. Hoijtink H, Huntjes R, Reijntjes A, Kuiper R, Boelen PA. An evaluation of Bayesian inequality constrained analysis of variance. In: Hoijtink H, Klugkist I, Boelen PA, editors. Bayesian Evaluation of Informative Hypotheses. New York: Springer; 2008:85-108.
    84. Hoijtink H, Klugkist I. Comparison of hypothesis testing and Bayesian model selection. Qual Quant 2007 Feb;41(1):73-91. [CrossRef]
    85. Van de Schoot Rens, Hoijtink H, Mulder J, Van Aken MAG, De Castro BO, Meeus W, et al. Evaluating expectations about negative emotional states of aggressive boys using Bayesian model selection. Dev Psychol 2011 Jan;47(1):203-212. [CrossRef] [Medline]
    86. Ledesma D, Kumano H. Mindfulness-based stress reduction and cancer: a meta-analysis. Psychooncology 2009 Jun;18(6):571-579. [CrossRef] [Medline]
    87. Knaevelsrud C, Maercker A. Does the quality of the working alliance predict treatment outcome in online psychotherapy for traumatized patients? J Med Internet Res 2006;8(4):e31 [FREE Full text] [CrossRef] [Medline]
    88. Kammers MPM, Mulder J, De Vignemont F, Dijkerman HC. The weight of representing the body: addressing the potentially indefinite number of body representations in healthy individuals. Exp Brain Res 2010 Jul;204(3):333-342 [FREE Full text] [CrossRef] [Medline]
    89. Meeus W, Van de Schoot R, Keijsers L, Schwartz SJ, Branje S. On the progression and stability of adolescent identity formation: a five-wave longitudinal study in early-to-middle and middle-to-late adolescence. Child Dev 2010;81(5):1565-1581. [CrossRef] [Medline]
    90. Van de Schoot R, Wong TML. Do delinquent young adults have a high or a low level of self-concept? Self and Identity 2012 Apr;11(2):148-169. [CrossRef]
    91. Van Well S, Kolk AM, Klugkist IG. Effects of sex, gender role identification, and gender relevance of two types of stressors on cardiovascular and subjective responses: sex and gender match and mismatch effects. Behav Modif 2008 Jul;32(4):427-449. [CrossRef] [Medline]
    92. Kroese FM, Adriaanse MA, Vinkers CDW, Van de Schoot R, De Ridder DTD. The effectiveness of a proactive coping intervention targeting self-management in diabetes patients. Psychol Health 2013 Jan;29(1):110-125. [CrossRef] [Medline]
    93. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Methods 2002 Jun;7(2):147-177. [Medline]
    94. Asendorpf JB, Ran de Schoot R, Denissen JJA, Hutteman R. Reducing bias due to systematic attrition in longitudinal studies: The benefits of multiple imputation. International Journal of Behavioral Development 2014 Jul 03;38(5):453-460. [CrossRef]
    95. Feingold A. Effect sizes for growth-modeling analysis for controlled clinical trials in the same metric as for classical analysis. Psychol Methods 2009 Mar;14(1):43-53 [FREE Full text] [CrossRef] [Medline]
    96. Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal 2007 Oct 23;14(4):535-569. [CrossRef]
    97. Jung T, Wickrama KAS. An introduction to latent class growth analysis and growth mixture modeling. Social Pers Psych Compass 2008 Jan;2(1):302-317. [CrossRef]
    98. Mouthaan J, Sijbrandij M, De Vries GJ, Reitsma JB, Van de Schoot R, Goslings JC, Luitse JSK, Gersons BPR, et al. Internet-based early intervention to prevent posttraumatic stress disorder in injury patients: randomized controlled trial. J Med Internet Res 2013;15(8):e165 [FREE Full text] [CrossRef] [Medline]
    99. Gielissen MFM, Verhagen S, Witjes F, Bleijenberg G. Effects of cognitive behavior therapy in severely fatigued disease-free cancer patients compared with patients waiting for cognitive behavior therapy: a randomized controlled trial. J Clin Oncol 2006 Oct 20;24(30):4882-4887 [FREE Full text] [CrossRef] [Medline]
    100. Simard S, Thewes B, Humphris G, Dixon M, Hayden C, Mireskandari S, et al. Fear of cancer recurrence in adult cancer survivors: a systematic review of quantitative studies. J Cancer Surviv 2013 Sep;7(3):300-322. [CrossRef] [Medline]
    101. Servaes P, Verhagen S, Schreuder HWB, Veth RPH, Bleijenberg G. Fatigue after treatment for malignant and benign bone and soft tissue tumors. J Pain Symptom Manage 2003 Dec;26(6):1113-1122. [Medline]
    102. Servaes P, Verhagen CAHHVM, Bleijenberg G. Relations between fatigue, neuropsychological functioning, and physical activity after treatment for breast carcinoma: daily self-report and objective behavior. Cancer 2002 Nov 1;95(9):2017-2026. [CrossRef] [Medline]
    103. Buffart LM, Ros WJG, Chinapaw MJM, Brug J, Knol DL, Korstjens I, van den Borne B, Hoekstra-Weebers JEHM, et al. Mediators of physical exercise for improvement in cancer survivors' quality of life. Psychooncology 2014 Mar;23(3):330-338. [CrossRef] [Medline]
    104. Bryan A, Schmiege SJ, Broaddus MR. Mediational analysis in HIV/AIDS research: estimating multivariate path analytic models in a structural equation modeling framework. AIDS Behav 2007 May;11(3):365-383. [CrossRef] [Medline]
    105. Hayes AF. Beyond Baron and Kenny: statistical mediation analysis in the new millennium. Communication Monographs 2009 Dec;76(4):408-420. [CrossRef]
    106. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 2008 Aug;40(3):879-891. [Medline]
    107. MacKinnon DP, Taborga MP, Morgan-Lopez AA. Mediation designs for tobacco prevention research. Drug Alcohol Depend 2002 Nov;68 Suppl 1:S69-S83 [FREE Full text] [Medline]
    108. Gielissen MFM, Schattenberg AVM, Verhagen CAHHVM, Rinkes MJ, Bremmers MEJ, Bleijenberg G. Experience of severe fatigue in long-term survivors of stem cell transplantation. Bone Marrow Transplant 2007 May;39(10):595-603. [CrossRef] [Medline]
    109. Vermunt JK. Latent class modeling with covariates: two improved three-step approaches. Political Analysis 2010 Sep 23;18(4):450-469. [CrossRef]
    110. Dwan K, Gamble C, Williamson PR, Altman DG. Reporting of clinical trials: a review of research funders' guidelines. Trials 2008 Jan;9:66 [FREE Full text] [CrossRef] [Medline]
    111. Finfer S, Bellomo R. Why publish statistical analysis plans? Crit Care Resusc 2009 Mar;11(1):5-6. [Medline]
    112. Eysenbach G, Consort- E. CONSORT-EHEALTH: improving and standardizing evaluation reports of Web-based and mobile health interventions. J Med Internet Res 2011;13(4):e126 [FREE Full text] [CrossRef] [Medline]

    Edited by G Eysenbach; submitted 23.02.15; peer-reviewed by L Zhu; comments to author 08.04.15; revised version received 22.04.15; accepted 23.04.15; published 23.06.15

    ©Marije DJ Wolvers, Fieke Z Bruggeman-Everts, Marije L Van der Lee, Rens Van de Schoot, Miriam MR Vollenbroek-Hutten. Originally published in JMIR Research Protocols (, 23.06.2015.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on, as well as this copyright and license information must be included.