Parental Marital Discord and Treatment Response in Depressed Adolescents

A comparison of the effectiveness of cognitive behavioural interventions based on delivery features for elevated symptoms of depression in adolescents: A systematic review

Background

Depression is a public health problem and common amongst adolescents. Cognitive behavioural therapy (CBT) is widely used to treat adolescent depression but existing research does not provide clear conclusions regarding the relative effectiveness of different delivery modalities.

Objectives

The primary aim is to estimate the relative efficacy of different modes of CBT delivery compared with each other and control conditions for reducing depressive symptoms in adolescents. The secondary aim is to compare the different modes of delivery with regard to intervention completion/attrition (a proxy for intervention acceptability).

Search Methods

The Cochrane Depression, Anxiety and Neurosis Clinical Trials Register was searched in April 2020. MEDLINE, PsycInfo, EMBASE, four other electronic databases, the CENTRAL trial registry, Google Scholar and Google were searched in November 2020, together with reference checking, citation searching and hand-searching of two databases.

Selection Criteria

Randomised controlled trials (RCTs) of CBT interventions (irrespective of delivery mode) to reduce symptoms of depression in young people aged 10-19 years with clinically relevant symptoms or diagnosis of depression were included.

Data Collection and Analysis

Screening and data extraction were completed by two authors independently, with discrepancies addressed by a third author. CBT interventions were categorised as follows: group CBT, individual CBT, remote CBT, guided self-help, and unguided self-help. Effect on depressive symptom score was estimated across validated self-report measures using Hedges' g standardised mean difference. Acceptability was estimated based on loss to follow-up as an odds ratio. Treatment rankings were developed using the surface under the cumulative ranking curve (SUCRA). Pairwise meta-analyses were conducted using random effects models where there were two or more head-to-head trials. Network analyses were conducted using random effects models.

Main Results

Sixty-eight studies were included in the review. The mean age of participants ranged from 10 to 19.5 years, and on average 60% of participants were female. The majority of studies were conducted in schools (28) or universities (6); other settings included primary care, clinical settings and the home. The number of CBT sessions ranged from 1 to 16, the frequency of delivery from once every 2 weeks to twice a week and the duration of each session from 20 min to 2 h. The risk of bias was low across all domains for 23 studies, 24 studies had some concerns and the remaining 21 were assessed to be at high risk of bias. Sixty-two RCTs (representing 6435 participants) were included in the pairwise and network meta-analyses for post-intervention depressive symptom score at post-intervention. All pre-specified treatment and control categories were represented by at least one RCT. Although most CBT approaches, except remote CBT, demonstrated superiority over no intervention, no approaches performed clearly better than or equivalent to another. The highest and lowest ranking interventions were guided self-help (SUCRA 83%) and unguided self-help (SUCRA 51%), respectively (very low certainty in treatment ranking). Nineteen RCTs (3260 participants) were included in the pairwise and network meta-analyses for 6 to 12 month follow-up depressive symptom score. Neither guided self-help nor remote CBT were evaluated in the RCTs for this time point. Effects were generally attenuated for 6- to 12-month outcomes compared to posttest. No interventions demonstrated superiority to no intervention, although unguided self-help and group CBT both demonstrated superiority compared to TAU. No CBT approach demonstrated clear superiority over another. The highest and lowest ranking approaches were unguided self-help and individual CBT, respectively. Sixty-two RCTs (7347 participants) were included in the pairwise and network meta-analyses for intervention acceptability. All pre-specified treatment and control categories were represented by at least one RCT. Although point estimates tended to favour no intervention, no active treatments were clearly inferior. No CBT approach demonstrated clear superiority over another. The highest and lowest ranking active interventions were individual CBT and group CBT respectively. Pairwise meta-analytic findings were similar to those of the network meta-analysis for all analyses. There may be age-based subgroup effects on post-intervention depressive symptoms. Using the no intervention control group as the reference, the magnitudes of effects appear to be larger for the oldest age categories compared to the other subgroups for each given comparison. However, they were generally less precise and formal testing only indicated a significant difference for group CBT. Findings were robust to pre-specified sensitivity analyses separating out the type of placebo and excluding cluster-RCTs, as well as an additional analysis excluding studies where we had imputed standard deviations.

Authors' Conclusions

At posttreatment, all active treatments (group CBT, individual CBT, guided self-help, and unguided self-help) except for remote CBT were more effective than no treatment. Guided self-help was the most highly ranked intervention but only evaluated in trials with the oldest adolescents (16-19 years). Moreover, the studies of guided self-help vary in the type and amount of therapist support provided and longer-term results are needed to determine whether effects persist. The magnitude of effects was generally attenuated for 6- to 12-month outcomes. Although unguided self-help was the lowest-ranked active intervention at post-intervention, it was the highest ranked at follow-up. This suggests the need for further research into whether interventions with self-directed elements enable young people to maintain effects by continuing or revisiting the intervention independently, and whether therapist support would improve long-term outcomes. There was no clear evidence that any active treatments were more acceptable to participants than any others. The relative effectiveness of intervention delivery modes must be taken into account in the context of the needs and preferences of individual young people, particularly as the differences between effect sizes were relatively small. Further research into the type and amount of therapist support that is most acceptable to young people and most cost-effective would be particularly useful.

Predictors, Moderators, and Mediators Associated With Treatment Outcome in Randomized Clinical Trials Among Adolescents With Depression: A Scoping Review

Importance

The application of precision medicine principles for the treatment of depressive disorders in adolescents requires an examination of the variables associated with depression outcomes in randomized clinical trials (RCTs).

Objective

To describe predictors, moderators, and mediators associated with outcomes in RCTs for the treatment of depressive disorders in adolescents.

Evidence Review

A scoping review of RCTs for the treatment of depression in adolescents was conducted. Databases searched included MEDLINE, Embase, APA PsycInfo, and CINAHL. Included publications tested predictors, moderators, and/or mediators associated with depression symptom outcomes (eg, symptom reduction, response, remission) in RCTs pertaining to the treatment of adolescents, ages 13 to 17 years. Predictors were defined as variables that were associated with depression outcomes, independent of treatment group. Moderators were defined as baseline variables that were associated with differential outcomes between treatment groups. Mediators were defined by a formal mediation analysis. In duplicate, variables were extracted and coded with respect to analysis type (univariable or multivariable), statistical significance, direction of effect size, reporting of a priori hypotheses, and adjustment for multiple comparisons. Aggregated results were summarized by variable domain and RCT sample.

Findings

Eighty-one articles reporting on variables associated with outcomes across 33 RCTs were identified, including studies of biological (10 RCTs), psychosocial (18 RCTs), and combined (4 RCTs) treatments as well as a service delivery model (1 RCT). Fifty-three variable domains were tested as baseline predictors of depression outcome, 41 as moderators, 19 as postbaseline predictors, and 5 as mediators. Variable domains that were reported as significant in at least 3 RCTs included age, sex/gender, baseline depression severity, early response to treatment, sleep changes, parent-child conflict, overall psychopathology, suicidal ideation, hopelessness, functional impairment, attendance at therapy sessions, and history of trauma. Two publications reported a priori hypotheses and adjustment for multiple comparisons, both finding that baseline depression severity and family conflict were associated with poorer outcomes.

Conclusions and Relevance

This review identified commonly researched variables requiring more scrutiny as well as underresearched variables to inform future study designs. Further efforts to discover predictors, moderators, and mediators associated with treatment response have great potential to optimize care for adolescents with depression.

Marital and Parent-Child Relationships during Treatment for Adolescent Depression: Child-Driven and Bidirectional Effects

Adolescent depression can be a stressor for parents and families. This study evaluated how treating adolescent depression affects marital and parent-child relationships. We examined whether marital adjustment and parent-child conflict improved over the course of active treatment of depressed adolescents (36-week visit) and long-term follow-up (one year after discontinuation of treatment) in a sample of 322 clinically depressed youth participating in the Treatment for Adolescents with Depression Study (TADS). We also explored the bidirectional influences of adolescent depression and family relationships. Results indicated that marital adjustment was stable during active treatment but declined during long-term follow up. A structural equation model (SEM) examining the bidirectional relation between adolescent depression and marital adjustment indicated that higher adolescent depression at the conclusion of maintenance treatment (24-week visit) predicted a deterioration of marital adjustment at the end of active treatment (36-week visit). Parent-child conflict was unchanged during treatment and follow up. SEM analyses examining the bidirectional relationship between youth depression and parent-child conflict revealed that reduced depressive symptoms at the end of the active treatment period predicted improvement in parent-child conflict at subsequent time points. These findings suggest that youth depression and its treatment may influence long-term family functioning.

Evidence Base Update of Psychosocial Treatments for Child and Adolescent Depression

Depression in youth is prevalent and disabling and tends to presage a chronic and recurrent course of illness and impairment in adulthood. Clinical trial research in youth depression has a 30-year history, and evidence-based treatment reviews appeared in 1998 and 2008. The current review of 42 randomized controlled trials (RCTs) updates these reviews to include RCTs published between 2008 and 2014 (N = 14) and reevaluates previously reviewed literature. Given the growing maturity of the field, this review utilized a stringent set of methodological criteria for trial inclusion, most notable for excluding trials based in subclinical samples of youth that had been included in previous reviews (N = 12) and including well-designed RCTs with null and negative findings (N = 8). Findings from the current review suggest that evidence for child treatments is notably weaker than for adolescent interventions, with no child treatments achieving well-established status and the evidentiary basis of treatments downgraded from previous reports. Cognitive behavioral therapy (CBT) for clinically depressed children appears to be possibly efficacious, with mixed findings across trials. For depressed adolescents, both CBT and interpersonal psychotherapy are well-established interventions, with evidence of efficacy in multiple trials by independent investigative teams. This positive conclusion is tempered by the small size of the interpersonal psychotherapy literature (N = 6) and concern that CBT effects may be attenuated in clinically complicated samples and when compared against active control conditions. Data on predictors, moderators, and mediators are examined and priorities for future research discussed.