Finding My Way: protocol of a randomised controlled trial evaluating an internet self-help program for cancer-related distress

Patients are eligible to participate if they have been diagnosed with any cancer (e.g., breast, prostate, colorectal, testicular, lymphomas, haematologic or gynaecologic malignancies); their cancer is being treated with curative intent; they are currently receiving active cancer treatment (surgery, chemotherapy, radiotherapy), or, if they received surgery alone, have been diagnosed within the last six months; they are aged 18 years or over; and they have sufficient English language literacy to understand the information sheet and consent form and navigate the website, access to the internet (home or work), and an active email address and phone number.

Patients are ineligible to participate if they have a diagnosis of advanced cancer.

Patients who opt to withdraw from the study are asked if they would consent to continue completing follow-up measures, and for any existing data to be included in analyses. If consent is not given, their data will be erased from the database along with any documentation related to their involvement.

An overview of recruitment and enrolment processes is provided in Figure 1. Participants are recruited from Australian hospitals and cancer centres according to any of the following methods:

Participants follow the email link or type in the web address in order to access the website and register for the program. Registration generates an automatic email to the participant containing a link to activate their account. Following account activation, participants are directed to read an online information page and consent form. Eligibility criteria are clearly highlighted on the information page, and consent requires participants to check a box stating they do not have advanced cancer. Consent is obtained by the participant clicking on an “I agree” button on the online consent form.

Following the consent process, participants complete the baseline questionnaire. Participants are then randomised to either the intervention or attention-control condition and are directed to a welcome tutorial and their user home screen, from which they are able to access their first module immediately. Participants work through 6 modules: one new module is accessible each week, along with previously accessed modules. The order of module release is self-determined during a goal-setting exercise conducted when first logging into the program, which allows participants to identify which topics are most immediately relevant to them, and can be subsequently modified if certain topics become more relevant as participants progress through the program. Participants receive automated email reminders to use the program each week. A booster module, comprised of a summary of the key information points from the program and links back to topics and worksheets, is accessible one month later. The opportunity to provide qualitative and quantitative feedback is provided at the end of each module.

This study is a parallel RCT, with equal numbers of participants randomised to the intervention and control arms. Randomisation is automated within the website; the random allocation sequence is generated by a researcher blinded to the identity of participants using computer-generated random numbers, and allocation is then automated by the website in a 1:1 ratio. Randomisation is stratified by gender.

Participants are blinded to their intervention assignment; while they are aware there are two different versions of the website (explained as ‘information’ and ‘activities’ versions), they are not informed that one is a ‘control’ condition, are not aware which is the primary intervention being evaluated, and are blinded as to their own treatment group allocation. A list of blinded participant details is accessible to the research assistant for the purposes of conducting reminders about follow-up surveys.

The 6 modules address: (i) starting treatment – working with your medical team, covering assertive communication and decision making, (ii) coping with physical symptoms and side effects – including fatigue, pain, insomnia, and provides activity pacing worksheets, and relaxation audiotracks; (iii) coping with emotional distress – which covers depression, anxiety, anger and stress, and provides cognitive restructuring diaries, and mindfulness audio-tracks; (iv) body image, identity and sexuality – with psychosexual worksheets, and therapeutic writing activities; (v) your family and friends – comprising further assertive communication and needs assessment worksheets; and (vi) completing treatment, which includes self-management strategies to facilitate healthy lifestyles.

A private online note-taking feature, mood monitoring/management, and a resource section with links to reputable cancer-related organisations and other health websites are also available throughout the duration of the program. The program is outlined in Table 1.

An information-only version of Finding My Way, developed to provide an appropriate attention control for the study, covers the same 6 module topics without any of the worksheets, activities, relaxation exercises, or note-taking feature. Our previous research indicates that an attention control condition does not significantly reduce distress [29], and provides a more rigorous test of the intervention.

The primary outcome for this study is distress, measured as general distress and as cancer-specific distress. General distress is measured using the total scale score of the 21-item Depression Anxiety Stress Scale – short form (DASS: [33]), which assesses levels of anxiety, depression and stress over the previous week on a four-point scale ranging from (0) Did not apply to me at all to (3) Applied to me very much, or most of the time. Cancer-specific distress is measured using the 17-item Posttraumatic Stress Scale-Self Report [34], which measures on a 4-point scale (0 = Not at all or only one time, 3 = 5 or more times per week / almost always), the severity of each DSM-IV post-traumatic stress disorder symptom criterion in terms of how often respondents experienced each symptom in the previous week. For this study, the measure was adapted in order to reflect cancer diagnosis as the stressor.

Secondary outcomes for this study include Health-Related Quality of Life (HRQOL), coping, and the economic impact of the intervention in terms of Health Service Utilisation. HRQOL is measured using the European Organisation for Research and Treatment of Cancer Quality of Life Core Questionnaire [35], a 30-item comprehensive health related quality of life assessment for cancer patients which yields a global QOL score, and five functional subscales (physical, emotional, social, role, cognitive).

Coping is measured with the mini-Mental Adjustment to Cancer Scale (mini-MAC: [36]), a 29-item scale yielding 5 factors: Fighting Spirit, Helplessness/Hopelessness, Anxious Preoccupation, Fatalism, and Cognitive Avoidance. Items are responded to on a 4-point scale ranging from ‘definitely does not apply’ to ‘definitely does apply’.

Health Service Utilisation is examined using the 17-item Health Service Utilisation Questionnaire [37], which assesses health service use in the previous 12 months in terms of whether patients have been hospitalised and the number of visits to doctors and other health professionals. Hospitalisation is indicated as no/yes day only/yes spent at least one night (with number of days then recorded). Visits to GPs/hospital doctors/specialist doctors are indicated on a 7-point scale ranging from none to 25 or more times, and utilisation of other health professionals (e.g. physiotherapist) is indicated as yes or no.

Participant and medical characteristics, availability of social support, information-seeking preferences, emotion regulation, vulnerability to cancer-distress, adherence to the program, and participant satisfaction are assessed as postulated moderators of program efficacy.

Participant and medical characteristics are assessed at baseline only, and consist of age, marital status, occupational status, annual gross income, level of educational attainment, area of residence (rural/urban, state), ethnicity, cancer type, time since diagnosis, treatment received (surgery, chemotherapy, radiotherapy, hormonal therapy, other), any other chronic health conditions, and other support-services accessed.

Availability of social support is assessed using the 20-item Medical Outcome Study Social Support Survey [38], which consists of four subscales: emotional/informational, tangible, affectionate, and positive social interactions, with availability of the support type listed in each item rated on a 5-point scale (1 = none of the time, 5 = all of the time).

Information seeking preferences in terms of the type and amount of cancer information sought by patients is measured by the Miller Behavioral Style Scale scale [39]. This scale identifies information seekers who look for and amplify threat-related cues (monitors) and distracters who avoid and minimize such cues (blunters), and has been used extensively in cancer populations. The scale asks the respondent to imagine two stress-evoking scenarios, each of which is followed by eight statements that describe different ways of coping with the stressor, with four statements being of a monitoring or information-seeking variety and four of a blunting or information-avoiding variety. Respondents are asked to check all statements that apply to them.

Vulnerability to distress is assessed using the Difficulties in Emotion Regulation Scale [40], a 36-item measure of difficulties in emotion regulation across six dimensions, namely a) lack of awareness of emotional responses, b) lack of clarity of emotional responses, c) non-acceptance of emotional responses, d) limited access to emotion regulation strategies perceived as effective, e) difficulties controlling impulses when experiencing negative emotions, and f) difficulties engaging in goal-directed behaviours when experiencing negative emotions. Participants respond to each item on a 5-point scale (1 = almost never, 5 = almost always).

Adherence to the study is measured in two ways. First, the self-help compliance scale [41] measures participant self-reported engagement with program’s (i) information, (ii) worksheets, and (iii) weekly time usage. Second, website use indicators are tracked on the website itself, including: number of visits, length of time logged in, and number of modules and worksheets completed.

Finally, levels of satisfaction with the website will be assessed as a measure of whether this intervention improved psychosocial support for patients. Qualitative feedback will be gathered within the website, and at follow-up assessments.

Longitudinal sample size calculation for repeated measures was conducted using a program developed by Hedeker [42]. Previous online interventions targeting distress in community samples have effect sizes ranging from .42 to .65 for depression, and from .29 to 1.74 for anxiety [19]. Therefore, a conservative small-to-moderate effect size was selected for the current study. Based on the pilot RCT findings, an attrition rate of 21% at each of the assessment points can be expected [29]. Thus for a study with 2 groups and 4 assessment points, power set at 0.80, statistical significance set at α = .05 (two tailed), total attrition set at 21%, an effect size set at 0.35 SD, and an expected primary endpoint standard deviation of 4.0 at each time point [43], 94 participants per group are required, summing to a total sample of 188 participants.

Quantitative data will be analysed using the Statistical Package for the Social Sciences (SPSS) Version 17.0. Baseline differences between groups will be investigated using t-tests for continuous measures and χ² tests of independence for categorical measures. Linear Mixed Effects Models (LMEM) will be employed to examine the efficacy of the intervention. The LMEM calculates a regression line for each individual while controlling for baseline observation (fixed main effects). Baseline observations will be entered as covariates to eliminate the influence of baseline variability, resulting in a 2 (group: intervention; control) X 3 (time: post-program; 3-month follow-up; 6 month follow-up) fixed effects model for each outcome variable, with random effects accounting for individual variation. This approach effectively equalises conditions at baseline and consequently allows for direct comparison between conditions at each follow-up point. In this context, (a) interactions between condition and time, (b) main effects of group and (c) post-hoc pairwise comparisons at each follow-up point are all indicators of intervention effects. This analysis offers the benefit of estimation maximisation; it provides joint modelling for each participant of observed and missing data based on maximising likelihood for population parameters as a function of the observed data. LMEM analyses are therefore robust with respect to handling missing data and unbalanced designs in longitudinal research as all participants with at least one observed post-treatment data point are included. That is, as baseline data is entered as a covariate, participants who do not complete any of the post-program assessments are not retained. To overcome this limitation and ensure this trial adopts a fully intention-to-treat design, multiple imputation will be performed for participants with only baseline-data. Therefore all participants, regardless of missing data at one or more assessment points, will be included in analyses and accurate parameter estimates obtained. Within-group effect sizes and reliable change indices will be calculated to provide a measure of the magnitude and clinical significance of changes respectively. To test for moderation, separate LMEM analyses will be conducted for each combination of moderator and outcome variable, where the moderator and group are entered in model 1 to control for main effects. The product term is then entered in model 2, to represent the interaction between group and moderating variable, and will indicate the moderating effect.