Background
Psychotherapy usually is a restricted resource, often associated with prolonged waiting periods for people seeking psychotherapeutic treatment. In Germany, on average, the waiting time to start psychotherapy is 4.5 months [
1], with rural areas being particularly undersupplied [
2]. These waiting periods are disadvantageous for people seeking help as well as the health care system. Individuals in need of mental health care experience an increased risk of chronification [
3] and they are more dissatisfied with the help they receive [
4]. They utilize more unspecific health care offers [
5], causing high direct and indirect costs for the health care system [
6]. Moreover, waiting periods contribute to people not starting psychotherapeutic treatment [
7,
8] or not considering psychotherapy in the first place [
9]. In order to address this issue, appropriate interventions should be offered to those unable to receive immediate psychotherapy.
One possibility to produce relief for people waiting for psychotherapy is the implementation of web-based self-help interventions. Web-based interventions have the potential to bridge treatment gaps [
10,
11], as they can be applied flexibly, with comparably little time, space, and personnel resources [
12,
13]. Since the late 1990s, a large body of research has emerged, confirming that web-based interventions are effective in reducing a range of psychological symptoms [
14‐
18]. Particularly guided interventions offering feedback to participants have put forth promising effects [
19]. Most recent research suggests that web-based interventions can be comparably effective to face-to-face psychotherapy [
20]. While the efficacy of web-based interventions has been shown in various trials [
21‐
24], the implementation of web-based interventions into health care systems worldwide is still in its infancy. One possibility to integrate web-based interventions into the health care system is their implementation during waiting periods. Offering web-based interventions to those waiting for psychotherapy may prove to be superior to mere waiting for several reasons: Individuals unable to enter psychotherapy could receive immediate access to an evidence-based intervention, therefore be provided with instant help; they could access the intervention 24 h a day, and use it in a familiar environment without time or travel costs [
25]. For those in need of further treatment, interventions during waiting periods may facilitate progress in the subsequent face-to-face psychotherapy [
26]. Hence, the implementation of a web-based intervention during waiting periods may prove to be particularly beneficial for both, help-seeking people and the adequate allocation of resources in our health care systems.
In order to assess the potential benefits of web-based interventions, they must be implemented in populations in need of help. One common complaint during waiting periods is depressive symptomatology [
5]. With a point prevalence of about 8%, depressive symptoms are widespread in Germany [
27], substantially reducing the quality of life of those affected [
28,
29]. As almost half the patients in German outpatient clinics present with a depressive disorder [
30,
31] and depressive symptoms often occur comorbid with other mental disorders [
32], an intervention targeting the reduction of depressive symptoms is likely to be beneficial for many individuals waiting for psychotherapy.
The efficacy of web-based interventions for the reduction of depressive symptoms has been shown in several trials [
22,
33]. However, there is limited evidence for their effectiveness and acceptance when implemented during waiting periods. To the best of our knowledge, three studies have examined web-based depression interventions during waiting periods [
34‐
36]. Two studies reported intervention take-up rates between 26 and 53% [
34,
37], high satisfaction among users [
34] and large pre-post effect sizes of
d = .75 [
36]. The only randomized controlled trial testing a web-based depression intervention against mere waiting (treatment as usual) implemented an unguided intervention and had a number of methodological limitations, such as baseline assessment after randomization, substantial study dropout combined with missing intention-to-treat analyses and including participants without depressive symptoms [
35]. At this point, no conclusions concerning the effectiveness of guided web-based depression interventions compared to treatment as usual during waiting periods can be drawn. Also, considering the results on intervention take-up and user satisfaction, findings concerning the acceptance of web-based interventions during waiting periods are quite heterogeneous and vary depending on the measure of acceptance (e.g., uptake vs. satisfaction rates) [
34‐
36].
Taken together, the former studies highlight some of the challenges, such as low take-up rates [
37], as well as the potentials of web-based interventions in this setting, such as high satisfaction among users [
34] and possible reductions of depressive symptoms [
36]. More research is needed to determine to what extent web-based interventions during waiting periods are superior to treatment as usual and which participant and intervention characteristics might impact on effectiveness and acceptance.
Statistical analyses
Effectiveness analyses
Analyses will be based on an intention-to-treat principle by including all randomized participants into the analyses. Primary and secondary outcomes will be analysed using a linear mixed model, assuming data are missing at random. The mixed model for the primary outcome (depression) will include group, time (all three points of assessment) and the interaction of group and time as fixed effects and recruiting outpatient clinic as random effect. Secondary outcomes will be analysed accordingly. We will calculate between-group effect sizes for the primary outcome using the post-treatment depression means and their pooled observed standard deviation. Additional per protocol analyses will include only those participants who have not started psychotherapy until the end of all study procedures, completing all three assessments, and, regarding the intervention group, at least five intervention modules.
We will evaluate the clinical relevance of any given development in a generalized linear model by estimating numbers of treatment response and deterioration (based on the reliable change index; [
71]) and symptom remission (ADS-L score < 22; [
41]), and by calculating the number needed to treat (NNT) for one more remitted participant.
Potential moderators influencing treatment effects will be analysed in the mixed model analysis. As there is at this point little research concerning moderating variables in the field [
72], these analyses follow an exploratory approach. Potential influencing variables include socio-demographics, internet affinity or attitudes towards psychotherapy.
Analyses will be performed using an alpha level of .05 and two-sided tests. All analyses will be conducted using IBM SPSS.
Acceptance analyses
Intervention adherence will be calculated by assessing the response rate of returned screeners with interest in study participation and the number of modules completed by intervention group participants. Recruitment and dropout rates will be examined using absolute and percentage frequencies. Participants’ satisfaction with the intervention (T2) will be reported descriptively. Potential predictors influencing intervention adherence and intervention satisfaction (e.g., age, depressive symptoms, internet affinity, former psychotherapy) will be assessed in an exploratory multiple regression analysis.
Discussion
This study investigates the effectiveness and acceptance of a guided web-based intervention for individuals with depressive symptoms seeking psychotherapy. Participants are individuals with elevated depressive symptoms on a waitlist for psychotherapy at several psychotherapeutic outpatient clinics, who consent to applying an intervention for mood improvement. A randomized controlled trial is conducted, comparing an intervention group receiving immediate access to a web-based intervention to a waiting control group. We expect the intervention group to benefit from the web-based intervention with regard to depressive symptoms, psychological symptoms and quality of life at post intervention and 3-months follow up. In an exploratory approach, the acceptance of the web-based intervention during waiting periods will be assessed, taking various sources of information into account (e.g., take-up and dropout rates).
This study features a number of strengths. While a series of trials have highlighted the efficacy of web-based depression interventions [
22], more research on their effectiveness in practical settings is needed [
73]. This study focuses on the applicability of a web-based intervention in a setting where low-intensity interventions are scarce and urgently needed. At the same time the waiting period differs from other investigated settings, as participants are distinctly seeking face-to-face psychotherapy. Thus, participants might perceive the web-based intervention as less credible, thinking that the “real treatment” is yet to come, potentially reducing the effectiveness of the intervention in this setting [
74,
75]. The study is based on a solid methodology, applying a randomized controlled trial with three times of assessment. The pragmatic study design allows high external validity. As inclusion and exclusion criteria are reduced to a minimum, the investigated sample consists of individuals with depressive symptoms seeking outpatient psychotherapy. In line with current standards, analyses will be based on an intention-to-treat principle and performed using linear mixed-model analyses.
This study also has a few limitations which deserve note. As the study will be conducted in the context of routine mental health care, the waiting periods of participants will vary depending on the capacities of the cooperating outpatient clinics. Some participants will presumably start psychotherapy between post- and follow-up assessments. However, due to the limited time span between post and follow-up assessments, the take-up of a subsequent psychotherapy cannot be reliably assessed. Per protocol analyses will be conducted including only those participants who have not started psychotherapy until the completion of follow-up assessments. Due to the routine care setting and in favor of external validity, we include participants who may suffer from other primary psychological symptoms (e.g. anxiety). These participants might benefit more from other disorder-specific interventions. Nonetheless, trials have implemented web-based depression interventions in routine care settings with high external validity and put forth large effect sizes (e.g. [
36,
37]). Similar effects are expected in this study as all participants suffer from depressive symptoms and take an informed choice to partake in an intervention for mood improvement.
When it comes to web-based interventions, considerable dropout rates have been reported [
76]. The participants dropping out of the intervention will be asked for their reasons to do so, thus dropouts will be used for a deeper understanding of intervention acceptance and applicability. In an attempt to minimize intervention dropouts, we implement a web-based intervention with guidance [
19]. Additionally, per protocol analysis will be performed for a high quality data analysis. Another limitation is that the main outcome data is based on self-reports. This is a frequently conducted approach with a favorable study cost–validity balance, still clinician-rated outcome measures would be beneficial. We have selected measures with high internal consistency and validity. In line with other research in the field (e.g., [
63,
77]), retest-reliabilities are not reported as studies vary greatly in terms of evaluation times and sample characteristics, which makes retest-reliabilities difficult to interpret in the context of this study design. Also, due to related effort, no analyses of cost-effectiveness are performed in this study. Last, the recruitment of participants has started in February 2017; however, the study has been registered before the beginning of recruitment and no changes have been made to the registry.
The results of this study will be of great relevance for daily clinical practice, as they reflect the applicability of an evidence-based self-help treatment option for individuals seeking psychotherapeutic treatment. Addressing the problem of prolonged waiting periods, it is essential to investigate which treatment options are effective and accepted by those in need. Evidence-based interventions with minimal effort for outpatient clinics are scarce and urgently needed. Thus, implementing a web-based intervention in this setting may be beneficial for those on a waitlist and health care providers. Also, this study investigates a model to viably integrate web-based interventions into the health care system. Considering the growing interest and realization of new treatment approaches, such as stepped-care models [
78,
79], web-based interventions have the potential to play an important role as low-intensity interventions in the treatment of depression [
80]. As Kazdin and Blase [
81] emphasize, innovative treatment approaches are needed to decrease the burden of mental diseases on a large scale.