Recruitment
A total of 400 subjects with of mild to moderate depressive symptoms are recruited via the “Selfapy” website, advertising on German television and in numerous information brochures from health insurance companies. The central recruiting tool is a study website (www.selfapy.de/studie/), where interested people can register to participate. If a person decides to participate by giving their consent, they will be immediately directed towards the determination of the deadline for the incoming survey (Mini International Neuropsychiatric Interview: MINI [22], Hamilton Depression Scale: HRSD-24 [23]). Electronic informed consent has been recognized by previous ethics committees as a safe and ethical means of giving consent [24] and is used in this study.
Inclusion and exclusion criteria
To clarify the inclusion and exclusion criteria, interviews will be conducted with all subjects via telephone (MINI, HRSD-24). To be eligible, participants must 1) be between 18 and 65 years old, 2) have sufficient German language skills, 3) have uninterrupted access to the Internet, 4) score 12 or higher on the BDI-II , 5) provide their electronic data with the “declaration of consent”, and 6) have a main diagnosis of a mild to moderate depressive episode or dysthymia (according to the MINI).
Subjects were excluded if they met one of the following criteria: 1) they have been previously diagnosed with bipolar disorder; 2) they are currently experiencing or have experienced psychotic symptoms or substance dependence in the past; or 3) they are currently experiencing or have experienced suicide ideations (operationalized via HRSD-24 – subjects are excluded if they score 3 or over on suicidality items) in the past.
A primary diagnosis of a condition other than depression is not an exclusion criterion, because we want to represent routine care. An active substance dependency is an exclusion criterion, as this conflicts with the conduction of the program.
Subjects who do not meet our inclusion criteria due to severity of illness are encouraged to seek professional help.
Adequate language skills are determined during the initial diagnosis by the MINI and HRSD-24. Our inclusion criteria were critically discussed in the study board.
Randomization and blinding
Subjects who meet the inclusion criteria will be randomly divided into one of three groups: (a) immediate access to the unaccompanied depression course of "Selfapy", (b) immediate access to the accompanied depression course of "Selfapy" or (c) access to " Selfapy ”after a delay of 24 weeks (control group).
Block randomization is performed by an independent researcher using a random number assignment plan for this list, which was created by a computer controlled random number generator. These numbers are assigned blindly, i.e. subjects and researchers have no knowledge of which group a patient belongs to. The subjects are assigned to one of the three groups in a ratio of 2: 1. The subjects will be informed by email about the result of the random assignment. The email contains a code with which the subjects can start the intervention directly. Diagnostic interviewers are blind to the assigned group of subjects. An illustration of the patient flow can be found in Figure 1.
Intervention
"Selfapy" is a browser-based commercial program for reducing depression symptoms (https://www.selfapy.de). To achieve this goal, the user receives instructions on evidence-based methods and exercises in the areas of CBT, system therapy, behavioural therapy and mindfulness training. The "Selfapy" program is available in an accompanied and unaccompanied version.
Subjects in the intervention groups (accompanied and unaccompanied) have free access to the 12-week Internet-based self-help treatment. The online course is divided into 12 different modules. Every week a new topic is discussed, such as "self-awareness", "challenging negative thoughts" or "sociability". The weekly modules consist of informative texts, videos, interactive exercises and worksheets. Table 1 contains the 12 module topics and a brief description of these and the exercises they contain. The program also includes a so-called "classroom" by allowing the psychologist to follow the program of the subjects. The "classroom" includes integrated chat system through which psychologists and subjects can communicate.
Accompanied group
In the accompanied version of the programme, the subjects is supervised by an experienced psychotherapists in training throughout the duration of the programme. In the beginning, the expert and the subject get to know each other and consequently hold 20- to 25-minute conversations by phone on a weekly basis. During these discussions with the psychologist, the programme content is explained, and questions are answered. In a one-hour training session, all clinical psychologists who were interested in accompanying subjects were informed. Contents included general information on the study, risks and their handling, discussion of the non-accident concept, the handling and documentation of drop-outs and the standardization of the discussion content.
Unaccompanied group
In the unaccompanied version, the subject works independently throughout the programme.
A chat function with a psychotherapist in training is available in the “classroom”, which can be used to clarify questions for improved understanding. There is no content-based discussion about the chat. The psychologist was also trained here.
Control group
At the end of the follow-up period, subjects in the control group will receive free access to the online intervention programme. They are free to choose the type of access to the programme of their choice (accompanied or unaccompanied). The control group received isolated, nonspecific mood-stabilizing activities such as body scan instructions, guided abdominal breathing, and guided imagery. Subjects in the control group will not receive treatment or support from the researchers. However, they are free to seek any other help they desire, including pharmacological and psychological treatments. All concurrently used treatments will be measured repeatedly through self-report. Subjects in the control group will receive access to the intervention after the last follow-up assessment, 12 weeks after the baseline assessment. The control group could not participate in “Selfapy” within 24 weeks. Duplicate registrations are to be prevented by comparing email addresses. Subjects in the interviews are also specifically asked for double registrations.
Responses to crises and suicidality
A physical overuse can be excluded. The online survey at the time of measurement T1, T3 and T4 of the survey will take about 60 minutes. The processing time for T2 is about 30 minutes. A general side effect of the interventions could possibly be a lack of treatment success. In the case of severe depressive symptoms, the subjects are shifted to inpatient or outpatient psychotherapy. Suicidality is explored as part of the evaluation and rating and each rater has an emergency plan. Subjects allowed to attend ongoing psychotherapy. This depicts everyday treatment and ensures high external validity. However, all potentially relevant factors for treatment outcome are taken into account for the statistical evaluation. The study design was approved by the ethics committee of the medical faculty of the Charité University Medicine Berlin.
Table 1. Overview of course content
Module
|
Content
|
Exercises
|
Your beginning
|
A questionnaire is used to collect general data about this person and his/her problems.
|
Lifeline; problem cakes; miracle question
|
Target selection
|
First insights
|
In this module, the user is brought closer to the connection between thinking, feeling and acting. Based on this triangular connection, our course concept is explained to him/her.
|
Your personal triangles
|
Resources
|
Explanation and gathering of power sources. Prepare the behavioural activation. Different methods to assist users in collecting power sources.
|
Evaluation daily protocol; power source images; treasure chest; resource walk
|
Behavioural activation
|
Actively plan power sources during the day.
|
Activity plan
|
Automatic thoughts
|
These automatic thoughts and thinking patterns are what we want to address this week.
|
Automatic thoughts; core beliefs
|
Negative thoughts
|
Thinking about reality, thinking realistically
|
Think realistically
|
Self-esteem
|
Also refer to the resources.
|
-
|
Self-efficacy
|
Therefore, this week will make you aware of your recent successes. This will enable you to better allocate your power reserves and increase confidence in your abilities.
|
"Your personal achievements - be proud of yourself"
|
Social environment + stigma + social support
|
"This week's focus is on your environment and your social environment, and when we talk about sharing with our environment, another factor is very important: our social support can help us deal appropriately and successfully with our stress."
|
"Difficult communication -your backing"
|
Mindfulness + excursus: pleasure training
|
"Mindfulness is the focus of this week, and we want to work with you to cultivate a mindful approach to you and your environment, because a mindful attitude to life will help you to improve your stress management. Therefore, a well-being workout can help you to feel more joie de vivre and increase your well-being.
|
Mental note; mindfulness in everyday life; sitting meditation; body scan; your enjoyment moments
|
Solve problem
|
This week, we would like to introduce you to a training that will enable you to perceive a specific problem, capture potential responses in a problematic situation, and develop the competence to implement a particular course of action to solve this problem optimally.
|
Problem-solving training
|
Relapse cases
|
Finally, we want to tie up your anti-anxiety package with you. It should help you to recognize when the stress threatens to become too much again. We want to show you once more how you can best handle the stress when it cannot be avoided.
|
Risk situations; early warning symptoms; first aid plan; relapse protocol
|
Assessments
Assessments will be made at four distinct points in time. Baseline (T1, study entrance), six weeks after baseline (T2), three months after baseline (T3, end of the program), and twelve weeks after the completion of the program (T4, follow-up). All other questionnaires are completed by the subjects (self-rating). All MINI and HRSD-24 interviews will be conducted by the same two trained interviewers, namely, a psychologist and a medical student. The training of the interviewers was conducted at the Charité Department of Psychiatry and Psychotherapy with depressive patients and was followed by a discussion of the ratings guided by a psychiatric consultant.
Table 2. Assessment tools utilized throughout the programme
Assessment timeframe
|
Assessment instrument
|
Pre (T1)
|
Self-assessment – Questionnaires:
BDI-II; QIDS-SR16; BAI; WHOQOL-BREF; SWOP-K9; WAI-SR; APOI; SASS; Bado
|
Assessment by a therapist – Questionnaires:
MINI; HRSD-24
|
During (T2)
|
Self-assessment – Questionnaires:
BDI-II; QIDS-SR16; BAI; WAI-SR; APOI
|
Post (T3)
|
Self-assessment – Questionnaires:
BDI-II; QIDS-SR16; BAI; WHOQOL-BREF; SWOP-K9; WAI-SR; APOI; SASS; Bado
Assessment by a therapist – Questionnaires:
HRSD-24
|
Follow-up (T4)
|
Self-assessment – Questionnaires:
BDI-II; QIDS-SR16; WHOQOL-BREF; Bado
Assessment by a therapist – Questionnaires:
HRSD-24
|
Abbreviations. MINI = Mini International Neuropsychiatric Interview; BDI-II = Beck Depression Inventory-II; QIDS-SR16 = Quick Inventory of Depressive Symptomatology; BAI = Beck Anxiety Inventory; WHOQOL-BREF = WHO Quality of Life-BREF; SWOP-K9 = Self-Efficacy, Optimism, and Pessimism Scale; WAI-SR = Working Alliance Inventory - Short Revised; APOI = Attitudes towards Psychological Online Interventions Questionnaire; Bado = basic documentation; HRSD-24 = Hamilton Depression Scale; SASS = Social Activity Self-Assessment Scale
Outcome measures
Primary outcome measure
The primary outcome measure is the change in the self-assessment of depressive symptoms after the completion of the programme using the BDI-II throughout all assessment points.
Secondary outcome measures
Changes in depressive symptoms will be additionally assessed at baseline (T1) and post (T3) after 12 weeks as well as at follow-up (T4) using observer-rated inventories, namely, the HRSD-24 and the QIDS [25].
Furthermore, changes in the self-assessment of anxiety-related symptoms will be analysed using the BAI [26]. Changes in self-esteem will be evaluated using the SWOP-K9 [27]. Similarly, the SASS [28] questionnaire will be utilized to record the levels of social activation. The WHOQOL-BREF [29] will be used to assess the impact of the programme on physical changes and life satisfaction. The APOI captures attitudes towards online interventions [30].
The WAI [31] is an employment inventory for the collection of therapeutic alliances and is used to evaluate work alliances. The ACSA is a self-anchoring rating scale for subjective well-being that was originally developed as a simple method to measure quality of life consecutively in the patient-physician relationship [32]. The Bado sheet is used to assess the impact of the programme on subjects’ use of the healthcare system, including contact with health care providers, the number of therapy sessions, and the amount of contact with psychotherapists and psychiatrists.
Engagement and usage measures for iCBT
The log files on the online “Selfapy” platform will be utilized to keep track of certain data. These data include but are not limited to the number of times a subject visit the platform, the amount of time elapsed between logins, and the amount of correspondence shared between a subject and his/her respective therapist. As an addendum, the platform’s track and change functionalities shall also enable data to be collected on subject exact amount of usage on the platform. This includes the frequency of visits to the online modules, how much time a subject spends on each module, and the modules that have been completed by the subject and in which order. All the aforementioned information will provide an overall picture of the efficiency and success of the individual elements of the programme.
Sample size
The between-group effect size estimate is based on meta-analytic evidence for the effect size
observed in unaccompanied psychological interventions (d =0.28) [16]. Treatment effect sizes in a sample with mild to moderately depressed subjects may be quite low. We conservatively reduced the estimated effect size to d = 0.23 [33].
Based on this effect size, a power of 0.80 and an alpha level of 0.05, we need a total of 200 subjects to address our hypotheses. The sample size was further estimated based on an expected drop-out rate of 50%. Drop-out rates in previous studies that used similar interventions ranged between 9% [9, 10] and 50% [34]. Based on this assumption, we would need a total of 400 subjects. The sample size was calculated using G-Power [35]. In addition, a comparison will be drawn between the unaccompanied and the accompanied version of the online programme. The previously expanded experimental groups are compared to a control group, resulting in a total of three study groups to which the subjects are randomly assigned in a ratio of 2:1 (accompanied group: n=160, unaccompanied group: n=160, control group: n=80).
Planned statistical analysis
The recorded data from different points in time (T1, T2, T3, and T4 for both treatment groups and T1, T2, and T3 for the treatment groups in comparison to the control group) will be analysed. Treatment outcomes for 200 subjects (pre-post difference) will be analysed using intention-to-treat analysis, effect strength calculation (in between) and a linear mixed model (based on primary outcome measures). Effect size calculations (before and after therapy success) are performed. Independent t-tests and χ2 tests will be used to estimate between-group differences in demographics and pre-treatment measures at baseline.
To assess the magnitude of treatment effects, Cohen’s [36] effect sizes for each time point will be calculated by dividing mixed model parameter estimates of fixed effects at each post-treatment assessment by the pooled standard deviation of baseline scores. Effect sizes under 0.2 are deemed small, 0.5 moderate and 0.8 large [36]. Differences in response rates, remission rates and adherence rates will be examined using t-tests and χ2 tests.