This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Mental Health, is properly cited. The complete bibliographic information, a link to the original publication on http://mental.jmir.org/, as well as this copyright and license information must be included.
Evidence on technology-based psychological interventions (TBIs) for the treatment of depression is rapidly growing and covers a broad scope of research. Despite extensive research in this field, guideline recommendations are still limited to the general effectiveness of TBIs.
This study aims to structure evidence on TBIs by considering different application areas (eg, TBIs for acute treatment and their implementation in health care, such as stand-alone interventions) and treatment characteristics (eg, therapeutic rationale of TBIs) to provide a comprehensive evidence base and to identify research gaps in TBIs for diagnosed depression. Moreover, the reporting of negative events in the included studies is investigated in this review to enable subsequent safety assessment of the TBIs.
Randomized controlled trials on adults diagnosed with unipolar depression receiving any kind of psychotherapeutic treatment, which was at least partly delivered by a technical medium, were eligible for inclusion in our preregistered systematic review. We searched for trials in CENTRAL (Cochrane Central Register of Controlled Trials; until August 2020), MEDLINE, PsycINFO, PSYNDEX, CINAHL; until the end of January 2018), clinical trial registers, and sources of gray literature (until the end of January 2019). Study selection and data extraction were conducted by 2 review authors independently.
Database searches resulted in 15,546 records, of which 241 publications were included, representing 83 completed studies and 60 studies awaiting classification (ie, preregistered studies, study protocols). Almost all completed studies (78/83, 94%) addressed the acute treatment phase, being largely either implemented as stand-alone interventions (66/83, 80%) or blended treatment approaches (12/83, 14%). Studies on TBIs for aftercare (4/83, 5%) and for bridging waiting periods (1/83, 1%) were scarce. Most TBI study arms (n=107) were guided (59/107, 55.1%), delivered via the internet (80/107, 74.8%), and based on cognitive behavioral treatment approaches (88/107, 79.4%). Almost all studies (77/83, 93%) reported information on negative events, considering dropouts from treatment as a negative event. However, reports on negative events were heterogeneous and largely unsystematic.
Research has given little attention to studies evaluating TBIs for aftercare and for bridging waiting periods in people with depression, even though TBIs are seen as highly promising in these application areas; thus, high quality studies are urgently needed. In addition, the variety of therapeutic rationales on TBIs has barely been represented by identified studies hindering the consideration of patient preferences when planning treatment. Finally, future studies should use specific guidelines to systematically assess and report negative events.
International Prospective Register of Systematic Reviews (PROSPERO) CRD42016050413; https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42016050413.
RR2-10.1136/bmjopen-2018-028042
Depression is a common [
TBIs cover a heterogeneous group of treatments, differing in various aspects, as described by Ebert et al [
In the last decade, research on TBIs has grown rapidly [
First, there is no systematic review that structures available evidence on TBIs regarding different clinical phases of depression management (considering waiting periods, acute treatment, and aftercare) and their implementation in health care (stand-alone intervention, blended care, and stepped and/or collaborative care). Thus, little is known about the effectiveness and acceptance of TBIs concerning their specific application area (eg, as stand-alone interventions for acute depression treatment), as the majority of systematic reviews focus on the assessment of a specific TBI in general, such as computerized CBT (cCBT) for depression [
Second, to date, there is only one systematic review evaluating internet- and mobile-based interventions in people with formally diagnosed depression [
Third, there is little research considering different types of negative events with regard to TBIs [
We chose the methodology of a systematic review to structure a broad and rapidly growing research field. First, the systematic review should provide an overview considering published and unpublished evidence—including gray literature—in the field of TBIs for the treatment of depression. Second, by considering relevant aspects of TBIs as defined by Ebert et al [
In summary, our main aim is to structure available evidence on TBIs for the treatment of diagnosed depression, addressing the following research questions:
How much high-quality evidence (ie, RCTs) on TBIs in the treatment of diagnosed depression is available?
How is the evidence on TBIs distributed regarding general study characteristics (eg, year of publication)?
How is the evidence on TBIs distributed regarding treatment characteristics (investigated TBI programs, technologies for intervention delivery, degree and purpose of guidance, qualification of people providing guidance, intervention duration, and therapeutic rationale) and application areas (different clinical phases of depression management and their implementation in mental health care)?
Are negative events reported in studies of TBIs for the treatment of diagnosed depression and what kind of negative events are addressed (eg, symptom deterioration, nonresponse)?
This study was preregistered (PROSPERO registration number CRD42016050413), and a study protocol was published a priori [
We included studies if (1) the whole sample (≥80%) consisted of people (aged ≥18 years) diagnosed with unipolar depression (assessed by a formal classification system or by conducting a diagnostic interview) with any comorbidities and in any clinical phase of depression management, (2) the intervention was at least partly delivered through technical devices (eg, smartphones, computers, telephones), (3) the intervention was based on an explicit psychotherapeutic theory, and (4) the study was conducted as a (cluster) RCT.
We excluded studies if (1) participants were solely diagnosed by applying cutoff scores on depression scales or when they had a depressive episode in the course of a bipolar disorder; (2) concurrent conditions (either somatic or mental) were the main focus of the intervention; and (3) the intervention provided solely psychoeducational content, patient decision aids, depression management tools, or focused only on drug adherence.
The study protocol by Köhnen et al [
We searched the following key databases: CENTRAL (Cochrane Central Register of Controlled Trials), MEDLINE, PsycINFO, PSYNDEX, and CINAHL; see the study protocol by Köhnen et al [
In total, 2 reviewers (MK and SL) independently screened the first 100 records for inclusion. As the interrater reliability for this sample was found to be high (98%), only one reviewer (MK) screened the remaining records in the course of the title or abstract screening. However, a second reviewer (SL) assessed publications labeled as
We developed and piloted a standardized data extraction sheet containing characteristics of interest (see study protocol by Köhnen et al [
Essential characteristics were either judged (risk of bias assessment [
The risk of bias assessment was evaluated using the criteria described in the Cochrane Handbook for Systematic Reviews of Interventions [
We applied the recommended definition of negative effects from the
According to this, negative events comprise the following categories:
Deterioration: worsening of target symptoms in the course of treatment measured by validated target symptom scales [
Adverse events: any types of adverse events occurring during or after treatment, including physical symptoms (eg, headache); psychological symptoms (eg, depressed mood); and psychosocial, legal, and economic consequences (eg, conflicts with the partner) [
Severe adverse events: any type of adverse events leading to serious consequences, such as death, mortal danger, hospitalization, or disability [
Novel symptoms: novel symptoms describe the emergence of new psychological symptoms (other than the symptoms addressed in treatment), independent of whether novel symptoms are associated with treatment [
Dropout from treatment.
Nonresponse.
Unwanted events: any type of events that are experienced as negative by patients in the course of the treatment, independently of whether unwanted events are associated with the treatment being used. In addition, unwanted events are not necessarily related to the treatment outcome, such as technical issues causing frustration or social stigma [
We structured the included studies according to application areas: clinical phases of depression management consisted of waiting periods, acute treatment, and aftercare. Within different phases of depression management, TBIs can be distinguished concerning their implementation in mental health care. They can be delivered as stand-alone interventions (TBIs replacing face-to-face [F2F] therapy), as blended treatments (combining TBIs and F2F therapies), or as part of stepped (TBIs are used as low threshold, initial treatment options for people with a mild-to-moderate depressive disorder) and/or collaborative care models (TBIs may be provided alongside different treatment components, such as a TBI is offered in addition to a monitoring care manager, general practitioners’ care, and the provision of an online discussion forum). In addition, treatment characteristics were used to structure the available evidence. We used descriptive statistics (eg, frequencies, measures of central tendency, measures of variability) for quantitative analysis using Microsoft Excel 2013 (Microsoft).
We actively involved patients and their relatives in the process of conducting our systematic review by means of 2 workshops (see study protocol by Köhnen et al [
Electronic searches yielded 20,613 records. After deduplication, 15,546 records were screened by title or abstract. In total, 901 full-text articles were assessed for eligibility. Not fulfilling the population criteria was the major reason (366/901, 40.6%) for exclusion, as many studies included their participants by applying cutoff scores on depression rating scales, rather than including participants on the basis of a formal diagnostic process (eg, clinical interview). Other reasons for exclusion were not fulfilling intervention (172/901, 19.1%) and study design (88/901, 9.8%) criteria. The remaining studies (34/901, 4.0%) were excluded for diverse reasons: publications were unavailable or relevant study information was missing (eg, distribution of diagnoses across the sample) and also not retrievable by contacting corresponding authors. Overall, we included 241 publications representing 143 trials (83 published trials and 60 trials awaiting classification) covering all clinical phases of depression management.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart.
Overall, the identified studies (N=83) included 14,080 participants, ranging from 14 to 1089 per study. The mean age of the participants was 44.9 (SD 12.1) years, and two-thirds were female (67%; see
Most included trials had a trial registration (58/83, 70%), and approximately one-fourth (22/83, 27%) of the included trials had an accompanying study protocol (
Studies on TBIs for depression were published from 1990 to the date of our search update in August 2020 (
General study characteristics (N=83).
Variables | Studies, n (%) | ||
|
|||
|
58 (70) | ||
With study protocol | 21 (25) | ||
Without study protocol | 37 (45) | ||
|
25 (29) | ||
With study protocol | 1 (1) | ||
Without study protocol | 24 (28) | ||
|
|||
2 | 64 (77) | ||
3 | 16 (19) | ||
≥4 | 3 (4) |
Publications on TBIs for depression per decade (N=83).
Decade | Studies, n (%) |
1990 to 1999 | 2 (2) |
2000 to 2009 | 4 (5) |
2010 to 2019 | 69 (83) |
2020 (end of August) | 8 (10) |
Trials by geographical region (N=83).
Geographical region | Studies, n (%) |
Europe | 44 (53) |
North America | 23 (28) |
Australia | 9 (11) |
Asia | 7 (8) |
Overall, 26 specific TBI programs were evaluated in the included studies; 18 of these programs were evaluated by 1 study, and 8 were evaluated by more than 1 study. However, approximately half of the studies (40/83, 48%) did not provide a name for the applied TBI program (
We identified 107 arms (from 189 arms) in the included studies that applied TBIs. Most TBIs (78%) were delivered by one technical medium (eg, internet or telephone), whereas 22% of TBIs applied more than one technical medium (eg, internet and telephone). Most TBIs were delivered via the internet (54%), followed by telephone (11%), offline computer programs (7%), videoconferencing tools (3%), and mobile phones delivering text messages (2%; see
The purpose of guidance in TBIs was heterogeneous (
To structure the guidance in TBIs, we summarized the reported purposes of guidance to categories and identified 5 functions of guidance: (1) informative function (eg, answering queries related to technical issues or treatment), (2) monitoring function (eg, symptom monitoring), (3) adherence-facilitating or motivational function (eg, encouragement to continue with intervention), (4) feedback function (eg, providing feedback for homework), and (5) therapeutic function (eg, goal setting).
Most guided TBIs fulfilled more than one function addressing different needs of participants.
We rated the degree of guidance in TBIs according to the framework of Newman et al [
Trials applying blended treatments were classified in an extra category because these trials provide F2F guidance (eg, by psychotherapists) The included trials applied TBI arms that were either unguided (20/107, 18.7%), guided (59/107, 55.1%; combination of predominantly self-help, 46/107, 43.0%, and minimal contact therapy, 13/107, 12.1%), therapist-delivered (14/107, 13.1%), or blended treatments (14/107, 13.1%).
The qualification of people who provided guidance on TBIs and who delivered treatment via TBIs ranged from lay supporters (technicians, research assistants, etc; 8/71, 11%) to clinicians with experience in the treatment of people with mental illness (trained psychotherapists, 6/71, 8% as well as psychiatrists, 1/71, 1%). Most people providing guidance and delivered treatment via TBIs had a background in psychology (36/71, 51%).
Interventions’ duration of identified TBIs ranged from 1 to 52 weeks, with most interventions lasting between 6 and 12 weeks (median treatment length of 8.5 weeks). Interventions of 8-week duration were the most frequent (26/107, 24.3%) in the included studies.
Overall, we identified 15 different therapeutic rationales for TBIs, ranging from mindfulness to psychodynamic therapy. Most TBIs were based on cognitive behavioral treatment approaches (79%;
Concerning the clinical phase of depression management, almost all trials were related to the acute treatment of people with acute depression (94%), followed by trials addressing the aftercare of people with depression (5%), and one trial that applied a TBI as a tool for bridging waiting periods (1%). Regarding the implementation of mental health care, most TBIs were delivered as a (enhanced) stand-alone intervention (80%), followed by blended treatment approaches (14%), and 5 studies (6%) delivered TBIs as part of a collaborative (4%) or stepped (2%) care interventions (
Distribution of studies (N=83) on application areas. Color highlighting of cells indicates format of implementation of TBIs: grey = (enhanced) stand-alone intervention; blue = blended treatment approach; yellow = TBI as part of collaborative/stepped care interventions. TAU: treatment as usual; TBI: technology-based psychological intervention; F2F: face-to-face.
We applied a rather broad definition for blended treatments, since we included all studies that provided any type of F2F treatment tailored to depression (eg, psychotherapy, medication, depression specific GP care) in addition to TBIs irrespective of the study’s definition/label. In contrast, trials concurrently providing treatment as usual in addition to TBIs were considered as enhanced stand-alone interventions, if treatment as usual consisted of a systematically offered generic treatments (eg, general GP care for all participants) that was not specifically tailored to depression.
Most studies (70/83, 84%) reported dropout rates. However, reporting on dropouts is heterogeneous, as studies differed in their definitions on dropouts: there were studies reporting dropouts from treatment (or treatment completion rates; 38/83, 46%), whereas other studies applied other definitions of dropouts, for instance, treatment completers as defined by authors or withdrawals from the study (32/83, 39%), and approximately 16% (13/83) of included studies did not report any extractable dropouts (
Report of negative events in included studies (N=83).
Report of negative events | Studies, n (%) |
Dropout | 70 (84) |
Deterioration | 21 (25) |
Adverse events | 18 (22) |
Nonresponse | 16 (19) |
Severe adverse events | 14 (17) |
Novel Symptoms | 0 (0) |
Unwanted Events | 0 (0) |
The aim of our study was to structure available evidence on TBIs for the treatment of diagnosed depression to build a comprehensive evidence base and to identify research gaps.
As shown in
In addition, most studies implemented TBIs either as (enhanced) stand-alone interventions (80%) or as blended treatment approaches (14%) in the acute treatment phase of depression treatment, indicating a comprehensive evidence base useful for further analyses on the differential indication for this clinical phase of depression management. There is little evidence on TBIs as part of collaborative (4%) or stepped (2%) care interventions in our review (
Stepped care approaches incorporating TBIs are recommended by the German [
In summary, there is evidence on the acute treatment phase of depression, and there are promising approaches to improve mental health care for people with depression by using TBIs. However, there are only a few studies investigating TBIs outside of acute treatment and applying innovative treatment approaches, which is why we call for (1) more research in previously less-considered clinical phases of depression management (aftercare, waiting periods) and (2) more studies investigating stepped care approaches with different TBI components.
We found that most TBIs were based on cognitive behavioral treatment approaches (79%), especially CBT (65%), and that other guideline-recommended treatment rationales for F2F depression treatment, such as psychodynamic treatments, are barely researched in TBIs of diagnosed depression. This may be because of the fact that psychodynamic-oriented therapists have a more negative attitude toward internet interventions compared with other therapists [
Owing to the lack of studies investigating other approaches recommended by guidelines, we call for more studies on TBIs considering a broader variety of treatment approaches. This would allow for more differentiated guideline recommendations, as they are currently limited to the effectiveness of cCBT [
Considering the report of negative events in included studies, we found that apart from dropouts, other negative events such as deterioration, nonresponse, and (severe) adverse events were reported in a few studies (range 17%-25%) or not at all (unwanted events and novel symptoms). Although dropouts were reported by most studies (84%), there were reported quite heterogeneously, as only 46% of all studies reported dropouts from treatment (or completion rates), which is an important indicator for treatment adherence in TBIs. For example, it is well known that unguided TBIs produce significantly more dropouts than guided TBIs [
Although adverse events have been reported in 22% of studies, many studies have reported adverse events unsystematically. For instance, by stating that no adverse events have been noted for any study participants, which did not specify the method of capturing adverse events as well as the definition of adverse events in included trials. It was not clear if participants were asked about the occurrence of adverse events during or after treatment. On the other hand, there were trials systematically assessing adverse events, for instance, by mapping them to different symptom domains (eg, somatically and psychologically) and specifying time points for the assessment. In total, the method on how and when adverse events were captured remained unclear in most included studies, which may contribute to an underestimation of the occurrence of adverse events because it is more likely to report such events when specifically asked for it in comparison to spontaneous reports [
Our findings on the report of negative events are in line with a previous systematic review, which also noted that adverse events were heterogeneous and insufficiently reported in RCTs on people with a persistent depressive disorder, especially in psychotherapeutic studies, where the report of adverse events was largely neglected [
Given the great heterogeneity in the reporting of negative events in included studies, we suggest the use of specific guidelines or guideline extensions to future trialists, such as the CONSORT extensions for harms [
Our review was conducted in line with the Cochrane guidelines, reported following PRISMA guidelines [
Despite strict eligibility criteria (eg, diagnosed depression), the focus of our review is still broad because all TBIs were considered irrespective of application areas and certain treatment characteristics. Following a broad focus resulted in large heterogeneity of the included studies, which is probably challenging for subsequent meta-analyses. It may be possible that certain questions regarding the differential indication are unsuitable for evidence synthesis because of large heterogeneity (eg, differences in intervention duration) or because there are not enough studies available (eg, evidence for TBIs bridging waiting periods [n=1]).
We conducted a highly sensitive literature search considering key databases, databases of gray literature, and clinical trial registries, without limiting the literature search to language. Nonetheless, we may have missed trials published in languages other than English because databases containing primarily English records may fail to find trials published in other languages even when language restrictions were avoided [
Overall, the results indicated that there is a proper evidence base for TBIs in the acute treatment phase being either implemented as stand-alone or blended treatments. However, the evidence base of TBIs in aftercare or for bridging waiting periods was found to be hardly convincing. Moreover, most TBIs were theoretically based on cognitive behavioral treatment rationales. Thus, a (broader) evidence base including TBIs based on other therapeutic rationales is still missing.
Concerning the report of negative events in studies evaluating TBIs, it was found that some information on negative events was reported in almost all studies, but the report was quite inconsistent between studies.
Despite the unequal distribution of evidence concerning differing clinical phases of depression management and treatment characteristics, we compiled a comprehensive evidence base to subsequently assess the effectiveness, safety, and acceptance of TBIs.
Study characteristics.
Risk of bias assessment.
Treatment characteristics of technology-based psychological interventions.
Technologies for intervention delivery.
Therapeutic rationales of technology-based psychological interventions.
Report of negative events.
cognitive behavioral therapy
computerized cognitive behavioral therapy
Cochrane Central Register of Controlled Trials
Consolidated Standards of Reporting Trials
face-to-face
internet-based CBT
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
randomized controlled trial
technology-based psychological intervention
The authors thank Eileen Wehmann for her support in the data extraction process. The Federal Ministry of Education and Research (in German: Bundesministerium für Bildung und Forschung) funded the systematic review on the comparative effectiveness of technology-based interventions in different steps of depression care (TIDECA; funding code: 01KG1705, funding period: August 2017 to October 2019). The funding institution had no role in the design of this study, its execution, analyses, interpretation of the data, or the decision to submit results.
HB received consultancy fees, reimbursement of congress attendance, and travel costs as well as payments for lectures from psychotherapy and psychiatry associations as well as psychotherapy training institutes in the context of e–mental health topics. He has been the beneficiary of study support (third-party funding) from several public funding organizations. HB and MH participated in the current revision of the German S3 national clinical practice guideline on the treatment of adults with unipolar depression. MH and LK participated in the 2015 revision of the German S3 national clinical practice guideline on the treatment of adults with unipolar depression. MH and SL are licensed psychotherapists. SL is additionally employed at the institute for psychotherapy at the University Medical Center Hamburg-Eppendorf, which provides psychotherapist training in CBT. MK and MD are psychotherapists in training (CBT). TS is a psychotherapist in training (psychodynamic therapy).