How to Align DBT and DBT Skills with Adolescent Externalizing Problems

Jakubovic and Drabick (2023) provide an excellent overview of the impact of dialectical behavior therapy (DBT)-based interventions for externalizing problems among adolescents. The authors found small-to-medium sized effects of DBT-based interventions (Hedges’ gs: .32-.54), with significant heterogeneity among studies that was not accounted for by the intervention setting or geographic location, the inclusion of multi-family skill groups or phone/milieu coaching, trial type, the presence of all core components of DBT for Adolescents (DBT-A), or the symptom reporter (e.g., patient, caregiver, clinician, etc.). Interestingly, there was some evidence that interventions of longer duration, but not necessarily those with more or more frequent sessions, were associated with larger improvements in externalizing difficulties. These findings are notable for what they reveal both about the efficacy of DBT for externalizing difficulties and the limitations of research to date on these effects.

For instance, as Jakubovic and Drabick note, the use of a pre-post design by 12 of the 17 studies reviewed limit the ability to distinguish intervention effects from the passing of time or alternative interventions. Further, the median sample size of these studies was relatively small (N = 26), making it difficult to precisely estimate the size of the effects of DBT-based interventions. These study characteristics are somewhat surprising, given that DBT-A was developed 20 years ago. Finally, nine of 17 studies explicitly included adolescents with borderline personality disorder (BPD), somewhat limiting the generalizability of the results. BPD is an excellent starting point for this line of research, given the initial application of DBT-A to BPD and the relatively robust link between BPD and externalizing pathology more broadly (Southward et al., 2022). However, to strengthen this evidence base, we believe it is important to validate the degree to which DBT and DBT skills uniquely impact specific externalizing targets among adolescents with externalizing disorders. Here, we propose a research agenda to align DBT-based interventions with externalizing targets to guide sample selection, mechanism and target verification, and contextual factors to consider.

DBT-Based Interventions for Youth With Externalizing Symptoms Compared to Youth With Externalizing Disorders

As mentioned above, most studies included in Jakubovic and Drabick’s review consisted of samples of youth diagnosed with BPD, whereas researchers only specifically examined effects of DBT-A in youth with externalizing disorders (i.e., attention-deficit/hyperactivity disorder [ADHD] and oppositional defiant disorder [ODD]) in two studies. Thus, although DBT-based interventions appear efficacious for reducing externalizing symptoms in youth with BPD, it is unclear whether DBT-based interventions are effective for youth with externalizing disorders. The overlap in presentation and difficulties between BPD and externalizing disorders seems apparent, as both diagnostic categories present with emotion dysregulation, behavioral impulsivity, and interpersonal dysfunction. Additionally, those with BPD can and do show elevated rates of disruptive, aggressive, and antisocial behavior, although these types of behaviors are more commonly associated with externalizing diagnostic criteria. Yet, determining whether DBT-based interventions can be efficacious for youth with externalizing disorders is a complex task. This is because there is a substantial degree of within and between-person variability in the presentation of externalizing disorders, with several factors possibly contributing to this heterogeneity (e.g., age, callous-unemotional [CU] traits) and thus requiring closer examination when designing and evaluating treatment programs for youth with externalizing disorders.

Age of Onset

One major factor that contributes to heterogeneity within externalizing disorders is age of onset. Whereas BPD is typically conceptualized as beginning in adolescence, two of the most common forms of externalizing disorders (i.e., oppositional defiant disorder [ODD] and conduct disorder [CD]) are diagnosed much earlier in childhood. Developmental psychopathology researchers have shown that the onset of highly aggressive and disruptive behaviors during childhood tend to persist throughout the lifespan relative to those that onset during adolescence, which more frequently desist and taper off as people enter young adulthood (Moffitt, 1993).

Further, patient age has important implications when adapting DBT-based interventions to children with externalizing disorders. Whereas DBT-A modules may translate efficiently to teens and older adolescents with externalizing disorders, modules in their current state may need to be adapted for children and younger adolescents. This extends beyond logistical factors (e.g., reading level). For instance, DBT-A material would likely need to be tailored to be delivered in a more active manner to keep the attention of children with externalizing disorders, going beyond that of occasional role plays or group discussions. Further, the plethora of acronyms seen in DBT (e.g., DEAR MAN, GIVE, IMPROVE) may be overwhelming for youth who already struggle with attention and/or hyperactivity/impulsivity. This is not to state that DBT-based interventions should not be used for younger samples; current treatments for children with ODD and/or CD implement similar techniques as seen in DBT-A skills modules (e.g., “cool down” skills, social skills, accepting). However, for young children with more severe forms of externalizing psychopathology that onset earlier in life, it is worth exploring how current DBT-A skills can be tailored to target the unique challenges that these youth experience.

Callous-Unemotional Traits

Beyond when externalizing disorders onset, the presence of callous-unemotional (CU) traits can further complicate the presentation and treatment of externalizing disorders (Frick & Morris, 2004). CU traits are characterized by a lack of remorse, guilt, empathy, reduced care for one’s personal performance, and blunted affective expression. Youth with ODD/CD and concurrent CU traits are at an elevated risk for a host of negative outcomes including substance use, delinquent and/or criminal behavior, higher rates of police contact and arrests, and violent recidivism. Current treatment methods to reduce aggressive and disruptive behaviors in youth with CU traits have shown somewhat mixed results particularly as pertaining to long-term post-treatment outcomes. Specifically, youth with CU traits tend to experience a larger reduction in their symptoms during treatment but remain impaired at post-treatment (Bansal et al. 2019). Although emotion dysregulation (e.g., “big” emotions) is not thought of as a primary deficit in youth with CU traits, it nonetheless is included within various intervention protocols for these youth.

Given the inclusion of emotion dysregulation targets in these protocols, it would be helpful to explore how to adapt DBT-A skill modules for youth with externalizing disorders and CU traits. Indeed, certain modules, such as Emotion Regulation, Distress Tolerance, and Interpersonal Effectiveness, can reduce disruptive and aggressive behavior. However, additional modules may be needed that specifically target and emphasize empathy and prosocial behaviors to get at the unique deficits associated with CU traits. For instance, the recent Coaching and Rewarding Emotional Skills (CARES) targets deficits in child empathy by teaching and improving a child’s emotion recognition skills through (1) emotional labeling of nuanced facial cues; (2) linking emotions to specific social contexts; (3) teaching and increasing prosocial and empathic behaviors through social stories, parent modeling, role play, and positive reinforcement; and (4) increasing a child’s ability to manage difficult emotions through parent modeling, role-playing, and reinforcing child’s use of their learned cognitive-behavioral strategies to decrease the aggressive behaviors. Recently, the CARES module, in conjunction with parent-child interaction therapy, led to reductions in aggressive and disruptive behaviors for young children with externalizing disorders and elevated CU traits (Fleming et al., 2022). This provides a framework for including modules focused on empathy training in comprehensive protocols, like DBT-A, to target the behavioral and emotional difficulties experienced by youth with concurrent CU traits. Quantitative and qualitative mixed methods studies could be particularly useful here to provide a more comprehensive understanding of how DBT-based interventions change symptoms in addition to related functional outcomes.

Conceptualizing Mechanisms of Change in DBT-Based Interventions for Externalizing Problems and Disorders

The heterogeneity among youth with externalizing problems suggests DBT-based interventions may function through unique mechanisms of change for different patient populations. Researchers have identified three broad classes of DBT-related mechanisms of change: (1) emotion regulation and self-control, (2) skills use, and (3) the therapeutic alliance and investment in treatment (Rudge et al., 2017). Given the focus of research to date on youth with BPD, it is necessary to test whether these mechanisms vary between or are consistent across BPD and externalizing disorders. For instance, emotion dysregulation is often considered a core component of externalizing disorders. Emotion dysregulation, in combination with impulsivity, also forms the foundation of the biosocial model that informs DBT-based interventions. However, youth with internalizing and externalizing disorders differ in how frequently they use classes of emotion regulation strategies, with those experiencing internalizing problems primarily using cognitive regulation strategies and those with externalizing problems using behavioral regulation strategies (te Brinke et al., 2021). BPD is characterized by both internalizing and externalizing problems, so it is likely that both sets of skills would provide unique benefits. By contrast, it may be possible to prune DBT-based treatments for externalizing disorders so they focus more on behavioral regulation skills. Tailoring DBT-based interventions this way could streamline treatment and reduce client burden, as full-model DBT-A is relatively intensive and may not be feasible in more time-limited settings. Experimental studies using specific skills or dismantling studies that deliver a subset of skills would be excellent next steps in unpacking which techniques are impactful for which patients and outcomes.

Relatedly, it would be useful to test for potential differences in the functional mechanisms by which DBT-A skills impact treatment targets. For instance, one of the core goals of DBT’s interpersonal effectiveness skills is to manage difficulties in social relationships by balancing objective, relationship, and self-respect priorities. Yet, such difficulties may develop through different functional mechanisms for those with BPD (e.g., due to difficulties tolerating being alone resulting from fears of abandonment and rejection) versus externalizing problems (e.g., defiance, argumentativeness, and vindictiveness), with some tailoring needed to promote effective skill use. Including relatively intensive (i.e., at each session) assessments of putative functional mechanisms in ongoing trials of DBT-based interventions would provide initial data on the strength of the association between changes in these mechanisms and outcomes for youth with BPD compared to those with externalizing disorders. Researchers could then build on these observational associations in follow-up experimental studies to test the degree to which different DBT-A skills act on specific mechanisms.

Lastly, concerning the therapeutic alliance, youth with externalizing problems may have a particularly difficult time establishing enduring rapport with therapists, although a stronger therapeutic alliance later in treatment still seems to be related to better outcomes (Florsheim et al., 2000). This could partly explain Jakubovich and Drabick’s findings regarding the longer duration of treatment being associated with better outcomes. Yet, this finding and its associated implications require further exploration. It is possible that longer-lasting treatment, rather than treatment with more frequent sessions, could facilitate natural and enduring alliance building in youth with externalizing problems. Such theories require further testing, as they could provide crucial information regarding how to optimize these treatments. In particular, researchers should include repeated measures of the alliance from the perspective of youth, caregivers, therapists, and observers throughout treatment to replicate this finding and evaluate its generalizability to DBT-based interventions.

Characterizing the Effects of DBT-Based Interventions Within the Hierarchy of Psychopathology

Jakubovich and Drabick’s findings are particularly applicable to dimensional, transdiagnostic spectra thought to characterize various forms of psychopathology (e.g., hierarchical taxonomy of psychopathology [HiTOP]), and also provide compelling avenues for further extension. For instance, in the HiTOP framework, the five-factor model of personality serves as an overall framework for indicators of a range of psychopathology. Conceptualized in this model, BPD and externalizing problems are both characterized by lower levels of agreeableness and conscientiousness. The transdiagnostic benefits of DBT-based interventions, as suggested by Jakubovich and Drabick (2023), may derive from their impact on these shared personality dimensions, rather than on BPD and externalizing disorders separately. More specifically, researchers have highlighted irritability and impulsivity as particular facets of low agreeableness and conscientiousness, respectively, that may be targeted by DBT-A skills. To connect research on DBT-A to hierarchical models of psychopathology, it would be helpful to include outcome measures of these broad dimensions and specific facets to test whether DBT-A skills target one or a few of these dimensions or facets rather than BPD specifically, given a growing body of research suggesting that DBT-A provides benefits across a wide range of disorders. It is also possible that a “superspectra” level of conceptualization may be needed to capture the benefits of DBT, such that the treatment primarily intervenes on a general “p-factor” that characterizes all types of psychopathology. Studies to date have yet to characterize the nature of the p-factor, with impairment, neuroticism, and impulsivity all preliminarily proposed (Southward et al., 2022). Jakubovich and Drabick’s findings suggest a potentially fruitful way of understanding the nature of the p-factor by parsing outcomes targeted by transdiagnostic DBT skills.

This line of thinking introduces other novel questions regarding the best ways to tailor DBT-based interventions through the omission or addition of certain treatment strategies. On the one hand, BPD, but not externalizing disorders, is characterized by high levels of neuroticism, so DBT-A skills aimed at reducing neuroticism (e.g., ABC PLEASE) may be less beneficial for youth with externalizing problems. It may also be necessary to incorporate or emphasize additional treatment techniques when addressing externalizing problems. For instance, interventions such as family therapy (e.g., parent management training) remain the gold-standard for externalizing problems like ODD but are only incorporated into DBT-A as needed. Framing the targets of DBT-based interventions in terms of hierarchical models of psychopathology facilitates the distinction of treatment components that are impactful across dimensions of psychopathology from those that are more tailored to specific dimensions and can be matched to the presentation of patients and the resources of the healthcare system.

Accounting for Cultural and Contextual Factors in the Applicability of DBT Skills

A relatively unexplored line of research related to DBT more generally concerns whether skills are equally efficacious across various demographic groups. Some researchers have explored cultural differences in the effectiveness of DBT skills, with heterogeneity emerging in the usefulness of dialectical thinking and acceptance across different cultures. Cultural adaptations have been proposed to related interventions, like CBT, and it may be worthwhile to explore whether analogous adaptations to DBT-based interventions, across both externalizing problems and BPD, may be necessary to facilitate culturally competent treatment, facilitate skill use, and improve the therapeutic relationship. Very little work has been done on this topic, with corresponding results being equivocal, yet this remains an important consideration before generalizing DBT to other populations. We encourage researchers to actively recruit more diverse samples to increase their statistical power to test for these cultural differences.

Contextualizing DBT-Based Interventions

It will also be important for future researchers to consider the context or setting in which DBT-based treatments for externalizing problems could occur and be most effective. More specifically, it remains unclear whether DBT-based interventions are best delivered as a stand-alone treatments, or whether certain modules could be incorporated into well-established treatment protocols like anger control training, behavioral therapy, parent management, or parent-child interaction therapy. For instance, anger control training targets emotion regulation by teaching youth with externalizing disorders how to recognize, label, and manage their anger through role plays and behavioral rehearsal of anger-provoking situations (see Eyberg et al., 2008). This description is similar to that of the DBT-A distress tolerance module. However, unlike DBT-A, anger control training is predominately delivered to the youth whereas parents are not active participants in treatment. This target of delivery contrasts with research suggesting that the greatest reductions in disruptive and aggressive behavior occurred in settings where treatment was delivered to children and their parents, such as for youth with externalizing disorders and concurrent CU traits (Bansal et al., 2019). DBT-A appears to fit well into this context, given that its core components include individual therapy and group-based skills sessions that involve both the adolescent and their caregivers. That is, treatments involving current evidence-based skills, along with DBT-based techniques from the distress tolerance module, such as TIPP or Self-Soothing with Six Senses, could be provided to youth and their caregivers and facilitate additive benefits.

In an adult program, DBT is typically implemented in an office setting. However, interventions for youth with externalizing disorders that aim to incorporate DBT-A skills may benefit from delivery in naturalistic settings that cater to the challenges seen in youth with externalizing disorders. For instance, summer research programs provide children the opportunity to engage in structured sports activities and free play while trained staff are implementing behavioral management techniques (e.g., point system, positive reinforcement) throughout every day and providing weekly caregiver training sessions (Bansal et al., 2019). The inclusion of distress tolerance, emotion regulation, and interpersonal effectiveness DBT-A modules could be worthwhile in encouraging youth to manage their regulation processes in real-world settings, in addition to relying on the behavioral management strategies that are currently in place. This strategy may be particularly helpful for addressing externalizing problems involving high levels of hyperactivity and impulsivity, such as ADHD. Ultimately, despite the factors contributing to the heterogeneity in disruptive and aggressive disorders, the underlying mechanisms of change of DBT-A appear to map onto the difficulties experienced by youth with externalizing disorders. Special consideration is required before introducing DBT-A modules into existing treatments; however, it does appear that the inclusion of various DBT-A skills could lead to even greater positive outcomes for youth and their families.

Conclusion

Overall, the evidence provided by Jakubovic and Drabick (2023) suggests that implementing DBT-based skills and approaches may provide benefits for youth experiencing externalizing problems. These findings highlight several areas in need of future study. In particular, we emphasize the importance of evaluating the generalizability of DBT-based interventions to specific externalizing populations, clarifying potentially transdiagnostic mechanisms of change, and determining the best practices for incorporating DBT-based techniques into currently existing externalizing problem treatment protocols. We included specific research designs to test each area and better facilitate the translation of areas of need into areas of study. Such work could provide critical insights to build on Jakubovic and Drabick’s work, and ultimately inform evidence-based practices that use the most effective components of DBT-based interventions to improve quality of care for youth with externalizing problems.