This pragmatic randomized controlled superiority trial aims to assess the effectiveness of a telephone-supported depression SCI in cancer survivors with moderate depressive symptoms when given in addition to usual care, in comparison to usual care only. Because of the pragmatic nature of this trial, members of both groups will continue to receive "usual care" for their depression, which may vary widely depending on local resources and medical practice. At each follow-up, participants are asked about specific treatment they have received for depression since entering the study (antidepressant medication initiation, discontinuation, change of dose, or psychotherapy) and their participation in community support groups and related activities; investigators will explore whether this care mediates the effects of the intervention. The study includes safeguards for participants who become suicidal or severely depressed during the study.
Research questions
Among cancer survivors in the post-treatment phase who have moderate to moderately-severe depressive symptoms \[Patient Health Questionnaire (PHQ-9) score 8-19\]:
1. Primary research question: What is the effect of a telephone-supported depression self-care intervention (SCI) plus usual care compared to usual care only (control group CTL) on the primary outcome - severity of depression symptoms as measured by the Centre for Epidemiologic Studies Depression Scale (CES-D) 6 months after randomization? Hypothesis: Participants who receive the SCI will have better 6 month primary outcomes than those randomized to the Control group.
2. Secondary questions:
1. What are the effects of the SCI on: CES-D score at 3 months; depression diagnosis at 6 months; and mental health related quality of life, activation, severity of anxiety symptoms at 3 and 6 months, and use of health services over 6 months? Hypothesis: Participants who receive the SCI will have better health outcomes and lower use of health services than those randomized to the Control group.
2. Are the effects of the SCI on the primary outcome modified by 5 primary baseline patient characteristics \[major depression diagnosis; antidepressant medication; activation; recruitment source (clinical vs community); and level of education \] or by 9 secondary patient characteristics \[cancer type, ongoing adjuvant cancer treatment (yes, no), years from cancer diagnosis (quartiles), severity of anxiety (as measured by the Hospital Anxiety and Depression Scale - Anxiety subscale (HADS-A)), physical health-related quality of life (as measured by the Physical Component Summary (PCS) of the Short Form Health Survey (SF-12)) (quartiles), sex, age (quartiles), family support for self-management (quartiles), province of residence (Quebec, Ontario)\].
Hypothesis and support Effect moderators (or modifiers) represent variables e.g. patient characteristics, measured at baseline, that interact with treatment to change outcome for each sub-group. These specify for whom treatment is most effective and can improve power in subsequent trials by better selection of study groups for stratification. In selecting those potential modifiers/moderators that will be pre-specified for testing, investigators followed consensus criteria for assessment of these factors. Pinctus proposes 5 criteria for systematic reviews of randomized controlled trials confirming moderator effects: 1) was the analysis a priori; 2) was selection of factors for analysis clinically plausible and either theory-based or evidence-based; 3) were moderators measured prior to randomization; 4) adequate quality of measurement; 5) explicit test of the interaction between moderator and treatment. For meta-analyses exploring moderator effects, criteria 1 and 2 are not required.
Investigators will analyze 5 possible moderators using sub-group analyses:
Major depression They hypothesize that the intervention will be more effective among participants with a diagnosis of major depression (measured through the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders Disorders - Major Depressive Syndrome in the Mood Disorders Module (SCID)) than those without. A meta-analysis of psychosocial interventions in cancer populations found that studies in which patients were pre-selected for higher levels of distress found greater effectiveness of the intervention. A meta-analysis of depression self-care interventions also reported greater effectiveness in patients with major depression, although only in univariate analysis.
Antidepressant medication Investigators hypothesize that the intervention will be more effective among participants with baseline antidepressant medication than those without. A meta-analysis found that the combination of psychotherapy and pharmacotherapy is more effective in the treatment of depression than psychotherapy alone.
Activation Investigators hypothesize that the intervention will be more effective among participants with higher baseline activation than those with lower levels. As there is no accepted cut-point on the Patient Activation Measure (PAM) scale, higher activation will be defined as a PAM score in the highest quartile. In a randomized controlled trial of the depression self-care intervention among primary care patients with chronic physical conditions, investigators found that coaching was more effective among patients with higher baseline self-efficacy (one component of activation).
Recruitment source Investigators hypothesize that the intervention will be more effective among participants recruited from community settings than from clinical settings. A meta-analysis of depression self-care interventions also reported lower effectiveness in patients studied in clinical settings, although only in univariate analysis. Participants recruited from community settings in this study (e.g., from community groups and social media) will already be selected to be more motivated to participate actively in the intervention.
Level of education Investigators hypothesize that the intervention will be more effective among participants with higher education, defined as some education beyond high school (e.g, diploma or university), compared with those with only high school or less education. Investigators are not aware of evidence supporting this hypothesis, but the use of paper and pencil and electronic tools will be more familiar to those with a higher level of education.
In addition to these 5 hypothesized moderators, investigators propose to examine the following 9 secondary potential baseline moderators, each with a maximum of 4 categories:
cancer type, ongoing adjuvant cancer treatment (yes, no), years from cancer diagnosis (quartiles), severity of anxiety (HADS), physical health-related quality of life (PCS) (quartiles), sex, age (quartiles), family support for self-management (quartiles), province of residence (Quebec, Ontario).
3. In the Intervention group, is greater adherence to the SCI (e.g., use of self-care tools, completion of written exercises and mood monitoring, goal setting, greater number of coaching contacts) associated with better primary and secondary outcomes? Hypothesis: Participants who adhere to the SCI, particularly those who actively engage with the Cognitive behavioural therapy (CBT)-based tools will have greater reductions in depressive symptoms, diagnoses, and severity of anxiety symptoms, and greater increases in activation and quality of life.
3. Economic analysis:
An economic analysis of the SCI is beyond the scope of this study. In the current study, investigators collect patient-reported data on use of health services and request additional (optional) consent for linkage to health care administrative data (the Ontario Health Insurance Plan (OHIP) Claims database and the Discharge Abstract Database (DAD)). They plan to seek additional funding for analyses of these use of services and administrative data for a future study on impact of the SCI on use of services. They will only extract use of services data for patients who consent to this optional element, and only for a period 12 months before and 12 months after the date of enrolment in the present trial. Patients who consent to use of their administrative data are asked to provide their OHIP number to allow researchers to extract the required data.
