A Systematic Review and Meta-Analysis of Third-Wave Online Interventions for Depression

Full reference: Sierra, M. A., Ruiz, F. J., & Flórez, C. L. (2018). A systematic review and meta-analysis of third-wave online interventions for depression. Revista Latinoamericana de Psicología, 50(2), 126–135. https://doi.org/10.14349/rlp.2018.v50.n2.6

Study type: Systematic review; Meta-analysis; Evidence synthesis; Psychological intervention analysis

Background and objectives

According to the World Health Organization (WHO), at least one in ten people suffer from unipolar depression (approximately 7% of the global population), making it the most prevalent mental disorder and the third most disabling health condition globally. Psychological treatment is the first and most recommended approach for depression, with evidence showing greater efficacy and preventive significance compared to pharmacological treatment alone.

However, the low level of access to mental health services may result from relative difficulties in accessing them in specific contexts (epidemiological, geographic, economic, geopolitical). At the onset of the digital era, computer-based psychological interventions were proposed as a promising alternative to conventional interventions. Recent research on the efficacy of online psychological interventions shows these are a legitimate and promising therapeutic activity as an alternative to traditional psychotherapy.

Third-wave or "third-wave" therapies include Acceptance and Commitment Therapy (ACT), mindfulness-based cognitive therapy, Functional Analytic Psychotherapy (FAP), Behavioral Activation (BA), Metacognitive Therapy (MCT), Mindfulness-Based Cognitive Therapy (MBCT), and therapies focused on rumination in processes of repetitive negative thinking (RFT). These approaches focus emphasis on the functions of private experiences rather than their form, frequency, or "pathology." Third-wave interventions differ conceptually and methodologically, but share at least three common therapeutic objectives: (a) cognitive openness toward one's own experience; (b) awareness of one's own behavior; and (c) promotion of valued actions in the presence of private experience.

The development of online third-wave psychological interventions for depression has grown in recent years, though research on this topic remains limited. Existing reviews of online third-wave interventions for depression have found significant reductions in depression symptoms (d = 0.40 to 0.56). However, there is less evidence on the effect of adaptations of third-wave behavioral therapies for depression. Accordingly, the present study reviews evidence of online applications of third-wave therapies for depression. Nine randomized controlled trials were found with a total of 1910 participants. Most studies applied online versions of Behavioral Activation (BA) and Acceptance and Commitment Therapy (ACT). All studies reported positive effects of third-wave therapies. A meta-analysis was conducted with 6 studies (N = 1744) that compared an online intervention versus a wait-list or usual treatment.

Method

Study selection

Studies on depression treatment using third-wave therapies presented through virtual media were identified using the Scopus bibliographic database (search conducted at the end of November 2016). Keywords used were "Web-based" OR "online" AND ("depression") AND ("Acceptance and Commitment Therapy" OR "Behavioral Activation" OR "Mindfulness" OR "Metacognitive Therapy" OR "Rumination-Focused Cognitive-Behavioral Therapy").

Second, the first and third authors reviewed the search output and selected articles meeting inclusion criteria (see below). Third, reference lists of articles meeting inclusion criteria were reviewed and included additionally if meeting criteria. Finally, reference lists of existing reviews and meta-analyses by Spiekerman et al. (2016), Carlbring et al. (2014), and Cavanaugh et al. (2014) were reviewed, as well as websites related to third-wave therapies (e.g., www.textualscience.org).

Inclusion criteria

This review included studies presenting any type of third-wave psychological intervention designed for treating depression through virtual media (websites, email, mobile applications, etc.). Studies presenting depression as a secondary outcome (e.g., studies on anxiety disorders, wellbeing, health, etc.) were excluded. Studies presenting intervention development were also excluded. Due to language limitations of the research authors, only studies written in English or Spanish were considered.

Coding procedures

Variables of treatment, participant, methodological, and extrinsic characteristics were coded to analyze study characteristics that might contribute to intervention effect size. Treatment characteristics coded were: (a) type of third-wave intervention implemented; (b) intervention duration in weeks; (c) type of media used to deliver intervention; (d) number of sessions/modules; (e) total intervention duration in weeks; (f) type of media contact used for intervention delivery.

Participant characteristics coded for each study sample were: (a) mean age in years; (b) percentage female; (c) total sample size; (d) sample type (clinical/community); (e) attrition rate at posttreatment; (f) mean age of participants; and (g) percentage female among them.

Methodological characteristics coded were: (h) study design; (i) follow-up time in months. Finally, extrinsic characteristics were: (j) year of publication; and (k) country and continent where the study was conducted.

Effect size calculation

Hedges' g effect sizes corrected for between-group control bias (Hedges, 1981) were computed on all outcome and process measures at pretreatment. Effect sizes were interpreted using Cohen's rule: small (0.2), medium (0.5), and large (0.8).

Results

Description of randomized controlled trials

The database search yielded a list of 172 studies. Eighteen studies met inclusion criteria for full-text review. Nine studies met inclusion criteria for the systematic review. Six studies met inclusion criteria for meta-analysis.

Table 1 presents the main characteristics of the described studies: first author, publication year, country of application, type of third-wave intervention used, delivery mode (website, smartphone application), primary outcome, general results, number of sessions or modules, intervention duration in weeks, sample size before randomization, attrition rates at posttreatment, mean age of participants, and percentage female among them.

Lappalainen et al. (2014) conducted a two-arm randomized clinical trial comparing two experimental conditions applying ACT: online therapy versus face-to-face therapy (N = 38). The online intervention consisted of six weekly modules of a website based on ACT content. The selection process consisted of a semi-structured interview delivered online where inclusion criteria were re-assessed, and explicit informed consent was required to be sent by email. Participants never had face-to-face contact with the research team. The BDI-II was the primary outcome measure, and follow-ups were conducted. Additionally, 97.3% of participants completed the intervention (N = 37) and 92.1% provided 18-month follow-up data (N = 35). The ACT protocol showed clinically significant decreases in depressive symptoms in both online and face-to-face conditions at posttreatment with few differences between treatments (d = 0.15). Gains were maintained at 6-month (d = 0.17) and 12-month follow-ups (d = 0.32), though differences between conditions were smaller.

Pots et al. (2016) conducted a randomized clinical trial comparing online BA versus a wait-list control condition (WLC) (N = 226). The intervention was based on the protocol developed by Kelders, Pots, Osbain, Bothernier, and Van Gemert-Pijnen (2013). Participants received reminders and feedback at the end of each module. The selection process consisted of (a) a battery of self-administered psychometric instruments presented to participants within the website; (b) the Mini International Neuropsychiatric Interview (MINI; Sheehan et al., 1998) delivered by phone. Participants never had face-to-face contact with the research team. The CES-D was the outcome measure. The BA protocol showed 51.2% of participants responded to CES-D at the end of intervention (N = 121), 53.3% provided 6-month follow-up data (N = 121), and 50% provided 12-month follow-up data (N = 126).

O'Mahen et al. (2013) conducted a two-arm randomized clinical trial with treatment as usual (TAU) versus control (N = 90). The intervention was designed for women presenting depressive symptoms after childbirth. Women in the experimental group and usual care control group were included. Those receiving only usual mother care received eleven weekly modules on a virtual platform based on BA. The intervention focused on depression symptoms with training in behaviors helping infants born alive. Women in the intervention group never had face-to-face contact with the research team. The EPDS was the outcome measure. Attrition rates: 62.3% of participants completed the intervention (N = 71) and 71% provided follow-up data (N = 98). The BA protocol showed significant decreases in depressive symptomatology at posttreatment (d = 0.63).

O'Mahen et al. (2014) conducted another two-arm randomized clinical trial with women presenting depressive symptoms after childbirth (N = 81). Both BA and ACT interventions were implemented. Participants were randomly assigned to an experimental group and usual care control group (N = 81). Both interventions lasted 8 weeks and consisted of six weekly modules in a mobile phone application. Participants could freely advance through the program. Limited support consisting of motivational voice calls and email reminders was provided. The selection process consisted of (a) a battery of psychometric instruments presented to participants within the website after providing virtual informed consent; and (b) the EPDS was the outcome measure. Participants never had face-to-face contact with the research team. The percentage of participants completing the intervention showed reductions in depressive symptoms at posttreatment according to BDI-II (d = 0.86).

Carlbring et al. (2013) conducted a two-arm randomized clinical trial with BA+ACT versus wait-list (WLC) (N = 80). The intervention included BA and ACT components and was implemented in 7 weekly modules. Participants could freely advance through the program. Limited support consisting of motivational voice calls and email reminders was provided. The selection process consisted of (a) a battery of psychometric instruments presented to participants within the website after providing virtual informed consent; and (b) the BDI-II was the outcome measure. Participants never had face-to-face contact with the research team. The percentage of participants completing the intervention (N = 71) and providing follow-up data (N = 78). Results at posttreatment favored the BA condition, but the difference was not statistically significant. Symptom level at pretreatment was modeled because BA showed greater symptom decreases in participants with higher pretreatment symptoms (within-group effect size at posttreatment: d = 1.8; within-group effect size at 6-month follow-up: d = 1.07).

Meyer et al. (2009) conducted a two-arm randomized clinical trial with wait-list control (WLC) (N = 396). The intervention lasted 9 to 10 weeks and consisted of 10 modules presented on a virtual platform based on various approaches (BA, MBCT, and CBT). Participants never had face-to-face contact with the research team. Participants did not receive reminders or feedback. Attrition was very high: 45.4% of participants did not provide posttreatment data. The BDI-II was the outcome measure. Results at posttreatment showed significant depression reductions (d = 0.63).

Meta-analysis

For meta-analysis, weighted effect sizes of online third-wave interventions versus WLC and TAU conditions were computed. Table 2 presents statistical data for studies included in the meta-analysis. Medium to large effect sizes were found with very significant heterogeneity of effect sizes favoring online third-wave interventions (d = 0.61, 95% CI [0.47, 0.75], p < .001). Figure 2 shows a graphic representation of effect sizes. Heterogeneity was null (Q = 1.61, p > .05), and there was no significant heterogeneity of effect sizes around the mean effect (I² = 0%). The six studies comparing the effect of an online intervention versus WLC or TAU conditions were included in the meta-analysis. Two studies tested the efficacy of ACT interventions (Lappalainen et al., 2015; Pots et al., 2016) and another two studied BA (O'Mahen et al., 2013, 2014). Remaining studies reported a combination of BA (Carlbring et al., 2013) and MBCT (Meyer et al., 2009). Participant sample size ranged from 38 (Lappalainen et al., 2014) to 910 (O'Mahen et al., 2013).

Discussion and conclusions

Nine randomized controlled trials exploring the efficacy of online third-wave interventions for depression were found. All studies were conducted in northern Europe. Most studies tested ACT or BA interventions and used the BDI-II as the primary outcome measure. The average duration of interventions was 8.8 weeks and included a mean of 7.4 modules. Sample size varied widely: from 38 (Lappalainen et al., 2014) to 910 (O'Mahen et al., 2013). In all included studies, participants remained anonymous through the virtual platform and their personal information was protected by a username and password system.

All studies reported positive results in terms of efficacy of third-wave therapies for depression. Meta-analysis was conducted with six studies comparing the effect of an online intervention versus WLC or TAU conditions. Effect sizes ranged from d = 0.55 to 0.86, with a weighted effect size of d = 0.61. There was very little variability around the mean effect (I² = 0%).

Some limitations of the current empirical evidence merit mention. First, previous CBT research has shown that the type of support moderated effect sizes for online interventions. Second, due to limitations in the number of studies included in the meta-analysis, moderation analyses could not be conducted in this study. Further research might explore this issue and establish whether a certain amount of support is more adequate for online third-wave interventions for depression. Third, online third-wave studies that have been implemented through the internet, with most studies focused on BA and ACT. Future research could explore the feasibility and efficacy of other third-wave and hybrid approaches and empirical comparisons among them. Fourth, all studies were conducted in northern Europe, which prevents extrapolation of these results to other contexts with different cultures such as Latin America or Asia.

In conclusion, the current empirical evidence is supportive of the efficacy of third-wave interventions implemented through the internet (mainly BA and ACT). The internet can be a very useful medium for delivering psychological interventions due to its easy access and good cost-effectiveness. Thus, the design and implementation of online psychological interventions could contribute to increasing mental health and quality of life globally. Future research should explore the efficacy of online third-wave interventions in other geographic and cultural contexts, and investigate mediator variables of depression administered through virtual media.

Significance and contribution

This study contributes significantly to the state of knowledge on online psychological interventions for depression, specifically those based on third-wave approaches. By synthesizing evidence from 9 randomized controlled trials (N = 1910 participants) and conducting a meta-analysis of 6 studies (N = 1744 participants), this work provides a comprehensive evaluation of the effect of online ACT, BA, MBCT, and other transdiagnostic approaches on reducing depressive symptoms. The finding of a weighted effect size of d = 0.61 (95% CI [0.47, 0.75]), representing a medium to large effect, validates online third-wave interventions as legitimate treatment options for depression, particularly in contexts where face-to-face therapy access is limited. Implications include potential expansion of evidence-based depression treatments through digital platforms, improved global accessibility to psychological interventions, and validation of innovative psychotherapeutic approaches based on acceptance, mindfulness, and commitment to values. The work also points to directions for future research, including exploration of moderator and mediator variables, extension of this approach to other cultural and geographic contexts, and evaluation of cost-effectiveness of online interventions.