Effectiveness of Transdiagnostic Cognitive-Behavioral Psychotherapy Compared With Management as Usual for Youth With Common Mental Health Problems

Key Points Question Can a transdiagnostic modular cognitive-behavioral therapy (CBT) program outperform management as usual for youth with emotional and behavioral problems? Findings In this randomized clinical trial of 396 youths aged 6 to 16 years, the parent-reported functional impairment was significantly reduced for youth allocated to transdiagnostic modular CBT compared with management as usual. Key secondary outcomes also indicated a broad range of benefits. Meaning This pragmatic study adds to the growing evidence that the wide-scale implementation of transdiagnostic modular CBT in nonspecialist care settings provides timely indicated prevention and quality care for help-seeking youth.

The steering committee also acted as the data and safety monitoring board. The steering committee met regularly throughout the study period and monitored the study's safety based on reports of local experiences without disclosing the allocated treatment group identity of cases.
Psykiatrifonden was responsible for the implementation of the MMM program, including training organization and therapist supervision, as well as video tool use and Web-based data collection. The trial was conducted in accordance with the principles of the Declaration of Helsinki. to select systematic reviews and meta-analyses fulfilling the following criteria: 1. Systematic reviews and/or meta-analyses of randomized controlled trials (RCTs) and cluster-randomized trials of a CBT intervention targeting anxiety, depressive symptoms/disorders, or behavioral problems/disorders in youth (younger than 18 years) that compare the experimental intervention with a control condition (e.g., waitlist control group, treatment as usual, attention control, or an active psychological intervention).
2. Reviews that assess and report the quality of the included studies to help readers assess the strength of the evidence in the review.
The systematic literature search and our reading of selected papers (see below) provided us with an overview of the evidence for the effects of CBT for anxiety, depressive symptoms and disorders, and behavioral problems and disorders in youth. Inspired by the distillation process described by Chorpita et al., 1  Here, we briefly discuss the current evidence base for problem-and disorder-specific CBT that targets anxiety, depressive symptoms and disorders, or behavioral problems and disorders in youth based on our literature search as of December 11 th , 2019. In general, the systematic reviews and meta-analyses assessed the quality of the evidence as low to moderate due to small samples, poor methodology, heterogeneity of results, and the risk of publication biases.

Evidence for the Effectiveness of CBT for Anxiety in Youth
The effects of CBT on children and adolescents with anxiety are documented in several systematic reviews and meta-analyses. [2][3][4][5][6][7][8] These studies proved the beneficial effects of CBT compared to a wait list control (WLC) on the rate of remission of anxiety diagnoses and on measures of psychopathology. The latest update of the Cochrane review (41 studies, 1806 participants) 7 examined an average of 13 weekly sessions of CBT compared to WLC and found significant beneficial effects on the remission rate (59% versus 15%, OR 7.85, number needed to treat [NNT]=3) and the reduction of anxiety symptoms. Apparently, there were no moderating effects of age, gender, or severity and comorbidity of the anxiety.
A recent network meta-analysis (101 studies, 6625 participants) 9 of the efficacy and acceptability of various types of psychotherapy for acute anxiety disorders in children and adolescents found that most formats of CBT and behavioral therapies (i.e., CBT without cognitive restructuring) were significantly more effective than WLC in reducing anxiety symptoms. The median number of sessions was 12; these were typically scheduled once per week. Group-based CBT was the only intervention that was significantly more effective than most other psychotherapies and control conditions in reducing posttreatment anxiety symptoms. It was likewise more effective than some psychotherapies and all control conditions after a short-term follow-up.
Therefore, in terms of posttreatment efficacy, the most effective treatments were group CBT and group behavioral therapy. There was a non-significant difference by age pointing toward group CBT for adolescents and group behavioral therapy for children. However, regarding the efficacy after a median follow-up period of 6 months (range 1-12 months posttreatment), the most effective treatments were parent-only CBT and individual behavioral therapy with parental involvement. In terms of health-related quality of life and functioning, almost all CBT, but not behavioral therapy, was significantly more beneficial compared to psychological placebo and WLC. The results 9 indicated that group CBT might be the initial choice of psychotherapy for anxiety disorders in children and adolescents. More research is needed to make clear recommendations regarding ageand disorder-specific treatment. Furthermore, there were indications of inflated estimates of the effects of CBT when compared with WLC 9 .
Comorbidity of anxiety with depression and conduct disorder is frequent and usually predicts an overall worse course during CBT treatment for anxiety compared to no comorbidity, 10,11 and even though both anxiety and depressive symptoms are reduced during CBT for anxiety, 11,12 a better treatment of comorbid conditions is warranted.
Although there is no clear evidence of additive beneficial effects of parental involvement in youth CBT, there is a clear rationale for the involvement of parents as support for the child (according to the developmental needs of the child) and as co-agents of change. Because most parents naturally try to protect their children from unpleasant situations and feelings, they may accidentally contribute to maladaptive patterns of avoidance behavior, thereby reinforcing the child's anxiety. 13 An important goal of parental involvement in CBT is therefore to make the parents aware of the maladaptive responses and teach them better ways to support their child in the sessions and during homework in which the child learns to apply more adaptive and realistic ways of thinking and acting. A study of various types of parental involvement 14 found that the active training of parents in contingency management and the transfer of control to parents were associated with better long-term effects. 14 A recent meta-analytic study 15 of CBT for internalizing disorders (anxiety, depression, posttraumatic stress disorder, and obsessive-compulsive disorder) in children and adolescents included 76 RCTs testing 106 CBT programs. The results indicated that parental involvement was significantly associated with larger pretreatment to posttreatment and pretreatment to follow-up effect sizes. The results indicate that parental involvement may help maximize the long-term effectiveness of youth CBT. This is also in line with the findings of the network meta-analysis 9 in which parent-only CBT and individual behavioral therapy with parental involvement produced the highest efficacy at long-term follow-up.
A comparison of CBT programs with and without booster sessions 16 found that CBT programs with booster sessions were associated with significantly better effects. In contrast, a recent meta-analytic study 15 of CBT for internalizing disorders found no significant associations between the use of booster sessions, goal setting, and maintenance/relapse prevention in CBT on one hand and the effect sizes at post-treatment or follow-up on the other hand.
In conclusion, CBT is established as an effective psychological treatment for anxiety disorders in children and adolescents across a range of ages, co-morbidities and delivery formats.
No clear evidence indicates that one way of providing CBT is more effective than another, but parental involvement may help maximize the long-term effectiveness of CBT for internalizing conditions in children and adolescents. There is no evidence for the long-term efficacy of youth CBT beyond 12 months after the end of treatment.
The most well-documented CBT programs for the treatment of anxiety disorders in children and adolescents are Coping Cat 17 , which was developed into a brief version 18 , and Cool Kids. 19 The effect of Cool Kids was tested in Denmark and demonstrated significant positive effects when delivered as group CBT with parental involvement and compared to WLC. 20

Symptoms and Disorders in Youth
The effectiveness of CBT for the prevention and treatment of subclinical and milder clinical levels of depressive disorders in youth was studied in several systematic reviews and meta-analyses of RCTs and cluster-randomized trials. 3,6,[21][22][23][24][25][26][27][28][29][30] The Cochrane review and meta-analysis of preventive psychological interventions 25 included 83 trials, of which 67 were carried out in school settings, eight in colleges or universities, and four in clinical settings. The preventive psychological interventions included CBT, interpersonal therapy (IPT) and third-wave CBT, which were compared with no intervention or with an attention placebo when available. Most interventions were delivered in a group format. The primary outcome was depression diagnosis at follow-up, which assessed up to 12 months after the intervention. The comparison of intervention versus no intervention (32 RTCs, 5965 participants) showed a small statistically significant effect (a depression rate of 17% versus 19%, NNT=33).
There were also small positive benefits associated with the psychological depression prevention programs measured as symptom reduction. However, when the interventions were compared with an attention placebo control, there were no effects. This was supported by a recent review 31 that found little evidence to support school-based interventions for universal and targeted prevention of depression or anxiety.
In terms of psychotherapy for the treatment of child and adolescent depression, the number of trials is relatively small, and the evidence is limited. A network meta-analysis 24 investigated the comparative efficacy and acceptability of psychotherapies for depression in children and adolescents (52 RCTs, 3805 participants) of nine psychotherapies and four control conditions. At post-treatment, only IPT and CBT were significantly more effective than most control conditions, and only IPT and CBT were more effective than play therapy. At follow-up, IPT and CBT were significantly more effective than most control conditions, and IPT and CBT were also superior to problem-solving therapy. Regarding acceptability, IPT and problem-solving therapy had significantly fewer all-cause discontinuations than CBT. The authors concluded 24 that IPT and CBT are the best available psychotherapies for depression in children and adolescents. Furthermore, the results indicated that use of WLC may have inflated the effect of the psychotherapies that were investigated in the network meta-analysis.
A 2006 meta-analysis 28 of the effects of psychotherapy for depression in children and adolescents found modest benefits associated with psychotherapy to treat depression. The mean effect size (0.34) was inferior compared to the mean effect size for CBT treatment of anxiety.
Furthermore, CBT fared no better than other approaches, and effects were not maintained after long-term follow-up. A recent update 26 included 55 RCTs of psychotherapy versus a control condition for youth depression. This meta-analysis found beneficial effects of active therapy (mean effect size 0.36) at posttreatment, and at follow-up (mean effect size 0.21) at an average of 42 weeks posttreatment. The effect sizes were significantly larger for interpersonal therapy compared to CBT. The effects of psychotherapy showed some specificity for depressive symptoms that were reduced more than measures of anxiety and externalizing behavior. The authors concluded 26 that the evidence base for psychotherapy, including CBT, for youth depression is limited.
An exploratory meta-analysis 29 found insufficient evidence to determine the active components of CBT for depression, but programs with consistent beneficial effects seemed to focus on cognitive restructuring skills and problem solving. 29 In contrast, Weisz et al. 28 found evidence pointing toward greater effects for CBT with a strong focus on behavioral methods (e.g., behavioral activation) relative to the focus on cognitive restructuring. Behavioral activation is aimed at increasing the engagement in nurturing activities and thereby creating opportunities for the individual to experience positive affect. 32 Behavioral activation is an evidence-based treatment for depression in adults with extensive research supporting its effectiveness. So far, few studies have focused on behavioral activation as a stand-alone treatment in youth. 33 These studies do, however, show promising reductions in depressive symptoms following behavioral activation, 33,34 so behavioral activation may thus be effective when delivered as a stand-alone treatment or integrated into CBT for depression in.
Regarding suicidality, a Cochrane review and meta-analysis 30 of psychological treatment versus antidepressant medication demonstrated a significant reduction in the incidence and severity of suicidal ideation with psychological treatment (mainly CBT) compared to drug treatment.
Very few studies have investigated the effects of CBT on preadolescent children with depression. A meta-analysis 21 focused on CBT for depressive symptoms in children with a mean age below 13 years (10 RCTs, 523 participants) and found significant effects of CBT compared to control conditions (WLC or active placebo), with larger effects in older studies, among older children, and with higher numbers of sessions. A more recent meta-analysis 22 of psychological treatments for depression in the same age group found inconclusive evidence for the effectiveness of all psychological treatments of depression in pre-adolescent children. The study included, however, only seven studies, of which only three were CBT studies of good quality.
In conclusion, there is evidence to support CBT as an effective treatment for depressive symptoms and disorders in mature children and adolescents, but the evidence is less clear for preadolescent children. The mean effect sizes are modest and significantly smaller than the effect sizes reported for CBT for youth anxiety disorders. CBT may be superior to psychopharmacological treatments of depression in terms of improved safety and reduced suicidal ideation. There is insufficient evidence to support school-based interventions for universal and targeted prevention focusing on anxiety and depressive symptoms.
The evidence-based CBT programs for the treatment of depressive symptoms and disorders in children and adolescents include the manual used in the Treatment of Adolescents with Depression Study, 35 the Primary and Secondary Control Enhancement Therapy, 36 the Penn Resiliency Program, 37 and the ACTION treatment program. 38 There are several ongoing studies of transdiagnostic CBT programs indicated for the prevention and treatment of anxiety and depressive symptoms in Norway, such as the transdiagnostic EMOTION intervention for children ages 8-12. 39 Evidence for the Effectiveness of Parent Training and CBT for

Behavioral Problems and Disorders in Youth
There are three groups of evidence-based interventions for behavioral problems and disorders (including violence, aggression, and disruptive behavior) in children and adolescents: 1. Parent training programs aimed at helping parents develop their parenting skills, communication, and contact with their child and to reinforce desired behaviors.
2. School-based CBT prevention programs for behavioral problems delivered individually or in groups with and without parental involvement and aimed at improving the child's social communication skills and anger management skills.
3. Programs that combine parent training and child psychotherapy.
Group-based parent training programs for parents of children with problem behaviors or conduct disorders have been extensively investigated. A Cochrane review and meta-analysis 40 compared group-based parent training programs for parents of children ages 3-12 with no treatment or WLC and found significant reductions in child conduct problems in the intervention group whether the changes were assessed by parents or by an independent, blinded assessor. The experimental intervention was also associated with significant improvements in positive parenting skills, as well as reductions in negative or harsh parenting practices, based on the parent and independent observer's reports. There were also small but significant effects on the parents' mental health. Moreover, economic evaluations of the intervention compared to WLC indicated that the cost of bringing the average child with clinically significant levels of conduct problems into the non-clinical range was only EUR 2217 per family. The severity of the child's behavioral problems and the family's socioeconomic status did not moderate the treatment effects. 40 The evidence for school-based indicated prevention programs for children and adolescents when delivered individually or in groups and with or without the involvement of parents have been documented in systematic reviews and meta-analyses. 41,42 The studies found significant positive effects of the school-based programs on behavioral problems, as well as on social skills, social cognition, and adaptation when compared to no intervention. Programs with a strong focus on adaptive social skills training led to greater effects than did programs with a strong focus on anger management. 43 Moreover, lengthy and more complex programs showed lesser effects than brief, more focused programs. 41 Some authors found that individual treatment had a greater impact than group treatment, 43 whereas others authors did not. 44 It seemed to be important that parents and teachers agreed on the need for intervention, as outcomes were significantly worse if only teachers recognized the child's behavioral problems.
A meta-analysis of CBT for behavioral problems in children with externalizing disorders (Attention Deficit Hyperactivity Disorder and Oppositional Defiant Disorder) 45 focused on interventions comprising parent training and child psychotherapy based on CBT-methods. The intervention was compared with WLC or self-help assistance for parents. The study's results indicated significant effects on children's disruptive behaviors, social skills, and parental distress.
Small reductions in problem internalization, aggression, and maternal depressive symptoms were also seen. 45 This evidence suggested the beneficial effects of combined parent training and child psychotherapy for externalizing problems.
In conclusion, parent training programs with and without child psychotherapy (i.e.,

Rationale for Transdiagnostic and Modular CBT
Despite evidence for the beneficial effects of CBT on anxiety, depressive symptoms and disorders, and behavioral problems and disorders in youth, there is a lack of implementation and dissemination of such evidence-based treatments in everyday clinical practice.
These persistent gaps between knowledge and practice call for initiatives to overcome the barriers related to implementing evidence-based programs to address common mental health problems in youth. Important challenges include the broad spectrum of common mental health problems, and the high rate of co-occurring and fluctuating problems in youth. Timely delivery of quality treatment requires the implementation of several disorder-specific programs-each supported by a system that identifies the target group, trains and supervises the therapists, and monitors the treatment effects. Frequent staff turnover in many health services adds to the complexity and expense of this model.
The challenges of implementing several disorder-specific treatment programs have inspired the development and evaluation of transdiagnostic treatment approaches that address multiple problem areas and enable more flexible adaptation to the needs of individual children and families, and thereby increase transferability to various care settings while retaining the benefits of evidencebased and manualized treatments. 55,56 The transdiagnostic treatment approaches have the potential to achieve the following: • Target the comorbidity and fluctuation of psychopathology in youths by tailoring treatment to the individual child.
• Target common key mechanisms of pathology.
• Minimize training demands and costs for clinicians.
• Facilitate the large-scale implementation of evidence-based treatment by targeting a large group of youth.
There are basically two different transdiagnostic approaches to the delivery of CBT for youth: 55 1. A unified approach that targets underlying and common dysfunctions across multiple psychopathologies (e.g., cognitive inflexibility and emotion regulation) to reduce avoidance and other maladaptive behaviors.

2.
A modular approach comprising problem-specific and generic modules along with decision rules (e.g., guided by flowcharts) that determine the dosing of modules to tailor the treatment to the individual child and family.
The implementation and evaluation of transdiagnostic treatment programs for youths is still in its early stages, but there are promising results from studies of the unified 57,58 and modular approaches 59,60 to the implementation of evidence-based CBT for youths.
The rationale for the development of a new transdiagnostic and modular manual in Denmark was the need for large-scale implementation of evidence-based psychological interventions to a very heterogenous group of youth in a non-specialist community care setting. The transdiagnostic and modular approach made it possible to bring together various evidence-based interventions targeting a broad range of mental health problems and disorders into one manageable "package" of interventions that could be "transported" into this type of setting. In other words: the primary goal was to design a single package of multiple interventions for multiple classes of problems/disorders to "speed up" the dissemination of an evidence-based practice in youth mental health prevention and intervention.
The first empirically supported transdiagnostic and modular intervention for youth with emotional and behavioral problems is the Child "STEPs" using the "Modular Approach to Therapy for Children with Anxiety, Depression, Trauma, or Conduct Problems" (MATCH-ADTC). 60,61 The MMM manual was inspired by the MATCH-ADTC, but our aim, content and structure differ in several aspects. Both manuals are transdiagnostic and modular and include somewhat similar evidence-based behavioral and cognitive behavioral (CBT) processes, methods and techniques for treating different types of anxiety, depressive symptoms and behavioral problems. However, the MMM manual was designed for indicated prevention and treatment of emotional and behavioral problems in help-seeking youths in a non-specialist school-based setting, whereas the MATCH-ADTC targeted children who were clinically referred for treatment in a community mental health setting. Despite the implementation support, the modular approach can be challenging to apply for the therapists. The therapist may experience a constant pressure to make decisions about the focus of the therapy, i.e. what are is the current problems, which one is more important to focus on, and when should the focus change? The MMM training begins with the formulation of a "Top-problem" as a starting point for the case formulation, the setting of goals, and the monitoring of outcome. The MMM manual guides the therapist to explore the problems in a collaborative process with the youth and parent, by which they reach a common understanding of the problem, or the set of problems that currently drive the distress and impairment and that motivate them to seek help. In case of multiple co-occurring problems, the MMM manual still encourages the formulation of one single core problem (called Top-problem) that is written down, using the own words of the youth/parent.

eMethods 4. Management as Usual (MAU)
To reduce the risk of attrition from the MAU group, the MAU treatment was enhanced compared to the regular MAU offered in the municipalities in the following ways: The parents in the MAU group were offered two care-coordination sessions (at week 2 and week 17) by a psychologist (or other local professional). The care-coordinators were not trained in delivering the MMM treatment. The first meeting had to be held no later than 13 days after randomization (i.e., parallel with the first session of MMM therapy). The aim of the first meeting was to review the problems identified in the visitation process of the MMM study, to suggest relevant treatment offers in the municipality, and to help coordinate the interventions in the municipality. The second meeting had to be held no later than week 17 (i.e., parallel to the end of the MMM therapy). The aim of the second meeting was to evaluate the child's current problems and needs. The treatment offers in MAU varied considerably from no intervention to counselling, talk therapy, pedagogical advice, network meetings, and/or individual support in the school setting. Some children were offered CBT interventions, but access to manualized psychological treatment in the municipality was limited.         Missing data were replaced using multiple imputation. Multiple imputations were performed separately for each group using chained equations with predictive mean matching for continuous outcomes and logistic regression imputation models for binary outcomes. Variables in the imputation model included all covariates in the primary analysis model. A total of 100 complete datasets were generated, with treatment effect estimates combined across datasets using Rubin's rules 62 .
Group-wise estimates for continuous outcomes are Least-Squares Means (SE), and dichotomous outcomes are mean numbers (%) at 18 weeks follow-up based on the 100 complete datasets. Estimates for difference between groups are based on Least-Squares Means with corresponding 95% confidence intervals (derived from linear models) for continuous outcomes. For dichotomous outcomes, estimates are odds ratios with corresponding 95% confidence intervals from logistic regression.  † † The Eyberg Child Behavior Inventory. Intensity score . Problem score (0-36). ‡ ‡ Inclusion cut-off: SDQ Total Difficulties score of at least 14, Emotional Problems score of at least 5, and/or Conduct Problems score of at least 3, combined with an SDQ Impact score of at least 1.
Missing data were replaced using multiple imputation. Multiple imputations were performed separately for each group using chained equations with predictive mean matching for continuous outcomes and logistic regression imputation models for binary outcomes. Variables in the imputation model included all covariates in the primary analysis model. A total of 100 complete datasets were generated, with treatment effect estimates combined across datasets using Rubin's rules 62 .
Group-wise estimates for continuous outcomes are mean changes (SDs), and dichotomous outcomes are mean numbers (%) at 26 weeks follow-up based on the 100 complete datasets.
Estimates for difference between groups are unadjusted differences in mean changes with corresponding 95% confidence intervals for continuous outcomes; for dichotomous outcomes, differences are estimated as unadjusted risk ratios with corresponding 95% confidence intervals.