Behavioral Activation as a Principle-Based Treatment: Developments from a Multi-Site Collaboration to Advance Adolescent Depression Treatment

Abstract

Adolescent depression is a serious and debilitating disorder associated with lifelong negative outcomes, including heightened risk for recurrence into adulthood, psychiatric comorbidities, and suicide. Among evidence-based treatments for adolescents, psychotherapies for depression have the smallest effect sizes of all psychiatric conditions studied. Advancing care for depression in adolescents is complex due to the heterogeneity in etiology and co-occurring difficulties among youth presenting with depression symptoms. This and a companion paper (Lewandowski et al., 2022) draw on a recent multisite collaboration that focused on implementing depression treatment for adolescents within clinical and research contexts. Specifically, this paper will review our work adapting behavioral activation (BA) as a principle-based framework to improve effectiveness and efficiency of depression treatment used within clinical and research settings in academic medical centers. Piloted adaptations include the use of BA principles to address idiographic drivers of depression and in-session BA “exposures” to illustrate BA principles. Case vignettes illustrate these adaptations of BA to address adolescent depression in the context of co-occurring difficulties.

During adolescence rates of depression rise dramatically (Hankin et al., 1998; Pontes et al., 2020; Thapar et al., 2012), and adolescent onset depression is associated with significantly increased risk for recurrence and death by suicide – the second leading cause of adolescent death (Ivey-Stephenson et al., 2020; Kochanek et al., 2019; Thapar et al., 2012). Even when optimally delivered, psychotherapy for depression produces the smallest effect sizes of any evidence-based treatment for youth psychiatric conditions (Weisz et al., 2017). Among those who do experience remission, 25–50% relapse within 2-years of intervention (Thapar et al., 2012). Heterogeneity among depressed teens may contribute to treatment challenges. Multiple factors contribute to risk of depression including biological and familial vulnerabilities, immediate environmental stressors, past exposure to adversity, and the interactions among these risk factors (Hankin, 2012). Thus, different developmental pathways lead to expression of overlapping core depression symptoms. Furthermore, co-occurring problems are common among depressed youth (Avenevoli et al., 2008; Kessler et al., 2003). For example, depression for one youth may present in the context of anxiety, while another may have symptoms of oppositionality and/or impairments related to attention deficit/hyperactivity disorder (ADHD), and still others may experience all the above or other problems altogether. Improving the effectiveness and efficiency of treatment for depressed adolescents remains a priority of clinical research.

Our group is a national collaborative of clinicians and researchers at the University of Washington (UW), New York University (NYU), and the National Institute of Mental Health (NIMH) engaged with the challenge of improving care for adolescent depression. We adopted behavioral activation (BA) as a focal point of our work after a careful review of the developmental and treatment literature related to depression during adolescence. This national collaborative includes clinicians leading a quality improvement initiative for adolescent depression at an academic medical center (Lewandowski et al., 2022), clinical researchers advancing psychotherapy treatment of adolescent depression, and affective neuroscientists aiming to understand the neurobiological underpinnings of depression. Our collective experiences highlight BA as a flexible, effective, principle-driven psychotherapy treatment with potential for wide-spread dissemination to improve depression care. This paper was informed by clinical work outside of a research protocol as well as treatment done within a formal research study in which consent/assent was obtained from parents and minors who participated. In the following sections we will review the BA model, its theoretical origins and neuroscientific underpinnings, and adaptations our team has developed and piloted. Case examples will illustrate applications of BA as a principle-based treatment that can address the treatment of depression in the context of common co-occurring difficulties.

Why behavioral activation?

While many existing evidence-based treatments for depression exist (e.g., cognitive-behavioral therapy (CBT), interpersonal psychotherapy (IPT); Weisz et al., 2006), no depression treatment has clearly emerged as more efficacious compared to others (Cuijpers et al., 2020). BA was selected as the treatment modality within our national collaborative on the basis of three features: 1) BA is principle-driven and idiographic; 2) it is straightforward and easy to use with teens, parents, and clinicians; and 3) the effectiveness of BA is supported by clinical research and emerging neuroscientific evidence.

Based on a functional analytic model of depression (Figure 1), BA utilizes the principles of operant conditioning (i.e., learning that occurs via the positive and negative associations between behaviors and their consequences) to increase positive reinforcement and decrease barriers to engaging in positively reinforcing behaviors. BA is particularly focused on addressing avoidant behaviors that emerge secondary to anhedonia, or the diminished capacity to feel pleasure or reward. BA clinicians take an idiographic approach to depression assessment that seeks to identify circumstances and actions that contribute to reduced positive reinforcement and maintain depression symptoms within the teen’s context. For example, an adolescent experiencing low mood and anhedonia following a conflict with a friend may engage in avoidant behaviors such as withdrawing from social contact with all friends. Social withdrawal may then lead to an overall reduction in positive reinforcement (e.g., isolation and fewer social invitations) and increased punishment (e.g., criticism from friends) that ultimately worsen depression.

Figure 1.

Behavioral activation: functional analytic model of depression and core principles.

BA has a robust evidence base as an effective treatment for depression in adults. Specifically, BA was shown to be significantly more effective than CBT and comparable to antidepressant medication in reducing depressive symptoms among moderately to severely depressed adults while also demonstrating superior retention rates compared to antidepressant treatment (Dimidjian et al., 2006). Indeed, in treating adult depression, a component analysis found that immediately post-treatment (Jacobson et al., 1996) and at a long-term follow-up (Gortner et al., 1998) the BA-only condition performed as well as the CBT package across self-reported symptoms and diagnostic interviewing measures. A recent meta-analysis (Cuijpers et al., 2020) examining adult depression treatment effectiveness across several types of treatments (i.e., CBT, BA, IPT, psychodynamic therapy, and problem-solving therapy) generally supported Jacobson et al.’s (1996) component analysis, although high levels of publication bias and heterogeneity in effect sizes temper a firm conclusion. The meta-analysis suggests that most therapies were found to effectively treat depression and have comparable effect sizes, which suggests less complicated BA protocols may be as effective as others that are more complex. BA studies also show efficacy across a diverse set of populations including older adults (Sood et al., 2003; Teri et al., 1997), Spanish-speaking Latinos (Collado et al., 2014, 2016; Kanter et al., 2010), and medically complex clients (Armento & Hopko, 2009; Hopko et al., 2005).

While BA has not been as extensively tested in adolescents, adaptations of BA for adolescent depression have demonstrated promising results (Chu et al., 2009; Cuijpers et al., 2007; Jacob et al., 2013; McCauley et al., 2016; Ritschel et al., 2011; Tindall et al., 2017). Specifically, two recent meta-analyses examined the use of BA to treat adolescent depression (Martin & Oliver, 2019; Tindall et al., 2017). While the meta-analytic findings indicated a need for further high-quality randomized control trials (RCTs) that utilize active controls and larger samples, pooled effect sizes support BA as an effective treatment for adolescent depression. Promising adolescent meta-analytic findings, the robustness of the adult literature, and the treatment’s relative simplicity and flexibility (as further outlined below) guided our collaborative’s decision to utilize BA in our work to improve adolescent depression care.

BA is a straightforward treatment approach that takes a non-judgmental view of depression as an understandable response in the face of stressful life events (Martell et al., 2013). It has been successfully administered to adults and adolescents with depression by individuals of varied training backgrounds including mental health trainees and licensed mental health professionals of varied training backgrounds (e.g., social workers, psychologists; Martin & Oliver, 2019). Additional work shows that adult depression care can be effectively delivered by non-mental health specialists (e.g., nurses) and lay people (Dimidjian et al., 2011; Ekers et al., 2013; Singla et al., 2014). A non-inferiority clinical trial of BA delivered by junior mental health-care workers to adults with depression was equally effective to cognitive behavioral therapy (CBT), a costly and complex treatment that requires more extensive psychotherapy training (Richards et al., 2016). Of note, although other approaches, such as problem-solving therapy, may also offer a more straightforward approach to treatment compared to CBT, there are currently no clinical trials examining the use of problem-solving therapy with depressed adolescents. Further, problem-solving skill building is typically incorporated into BA approaches within the context of learning how to overcome avoidance and barriers to goal-setting (Martell et al., 2013; Martin & Oliver, 2019; McCauley et al., 2016). Other studies have found support for BA delivered in primary care, brief group therapy, and via telephone, and research shows that clinicians can be successfully trained via online courses (Dimidjian et al., 2011). Together, this evidence further suggests that BA is a flexible, adaptable treatment well suited to a wide range of settings.

The focus of BA on improving depression symptoms through increased engagement in positively reinforcing activities is supported by the emerging neuroscientific understanding of anhedonia and reward processing. Reward processing refers to many aspects of hedonic functioning including learning associations between a cue or stimulus and a reward, interest in and motivation for pursuing a reward, effort expended to obtain a reward, and enjoyment of the reward once obtained (Rizvi et al., 2016). Anhedonia is an aspect of reward processing with specific subtypes including motivational anhedonia, or the reduced interest, drive, and effort to approach potentially rewarding experiences and consummatory anhedonia, or reduced pleasure when experiencing the reward (Treadway & Zald, 2011). Anhedonia is a core symptom of depression and, as the strongest predictor of increased time to remission and poor depression treatment outcomes, a critical treatment target (McMakin et al., 2012; Pizzagalli, 2014). Data suggest regions of the brain underlying anhedonia and the processing of reward, including the striatum and other basal ganglia structures, reach their developmental peak during adolescence (Urošević et al., 2012), and aberrations in ventral striatum functioning in healthy youth predict subsequent onset of depressive symptoms (Pan et al., 2017; Stringaris et al., 2015). Adolescents may be particularly vulnerable to disruptions in reward processing (Somerville & Casey, 2010; Somerville et al., 2011) and reduced reward responsivity has been implicated in the etiology of adolescent depression (Forbes et al., 2006, 2009; Forbes & Dahl, 2012; Nielson et al., 2021). These findings suggest that targeting reward processing may be important for the treatment of depression in teens. The focus in BA on both improving the motivation and experience of positive reinforcement and decreasing barriers to rewarding activities aligns with this understanding of reward processing and anhedonia in depressed youth.

Development and adaptation of behavioral activation

Our national collaborative met regularly to discuss the implementation and clinical use of BA within a unified system of depression management and a research context. The following sections review two key adaptations to BA derived from our national collaborative to optimize depression treatment in the context of co-occurring difficulties within our respective settings. These include: 1) the deployment of BA as a principle-based framework for treatment of depression in the presence of common co-occurring and contextual difficulties; 2) the augmentation of BA with in-session “exposures” to provide teens with in-session experiments that provide proof of concept for BA principles.

BA as a principle-based treatment framework

Lewinsohn (1974) outlined the importance of decreased response-contingent positive reinforcement in the onset and maintenance of clinical depression, a theoretical foundation based on the principles of functional behavior analysis (FBA) and operant conditioning that continues to inform our understanding of BA today. Lewinsohn further highlighted that the amount of positive reinforcement experienced by an individual was impacted by the general availability of reinforcers in a given environmental context and in the individual’s ability to elicit these reinforcers from the environment. Jacobson and colleagues’ body of work (Dimidjian et al., 2014; Jacobson et al., 1996, 2001; Martell et al., 2013) continued to advance and refine BA as part of their component analysis work to elucidate the primary mechanism of change for evidence-based treatments for depression.

While Lewinsohn’s behavioral model of depression involved a complex set of contextual, individual vulnerability, and contingency-related factors, contemporary BA models have been greatly simplified. Specifically, both adult (Dimidjian et al., 2014; Martell et al., 2013) and adolescent (McCauley et al., 2016) BA approaches highlight the following core principles: 1) utilizing FBA to understand what is uniquely reinforcing to the individual and identifying barriers to accessing these incentives and 2) utilizing the principles of operant conditioning by replacing avoidant behaviors, which limit access to reinforcing experiences, with approach-focused goal setting (see Figure 1 for an example of the functional model of depression). Importantly, these principles and the functional model of depression are used to guide the treatment process and tailor the approach to fit each individual’s values and goals based on an individual’s specific context and circumstances.

Although 60% of teens diagnosed with depression experience co-occurring mental health difficulties, including anxiety, oppositionality, and ADHD symptoms (Avenevoli et al., 2008, 2015), most depression intervention approaches do not include provisions for managing impairments related to co-occurring symptoms or for integrating them into the conceptualization of depression treatment. Marchette and Weisz (2017) reviewed several transdiagnostic treatment models that do make such provisions, including common elements approaches, such as MATCH-ADTC: Modular Approach to Therapy for Children with Anxiety, Depression, Trauma, or Conduct Problems (Chorpita & Weisz, 2009) and core dysfunction treatment approaches such as the Unified Protocol for Adolescents (Ehrenreich-May et al., 2017). In contrast, principle-guided treatments focus less on procedures and more on a model of therapeutic change. Specifically, principles must be: 1) central to empirically supported treatments; 2) applicable to multiple problems; and 3) associated with significant treatment benefit (Marchette & Weisz, 2017). These approaches decrease training burden and increase flexibility by focusing on “broadly applicable therapeutic concepts with fewer detailed instructions” (Marchette & Weisz, 2017, p. 976). Although there may be a trade-off between treatment fidelity and implementation burden within principle-based approaches, Weisz and colleagues’ initial data across several clinical trials indicate excellent treatment adherence with minimal implementation burden (Cho et al., 2021; Weisz et al., 2017). Of note, the implementation and training approach outlined in Weisz and colleagues’ work mirrors that of our own approach outlined in our implementation focused companion paper (Lewandowski et al., 2022). While further research is needed, principle-based approaches may improve fidelity, treatment adherence, flexibility, and effectiveness over manualized approaches due to clinicians following a more straightforward, overarching objective throughout care as opposed to rote adherence to a manual.

BA meets criteria for a principle-based treatment, and we began considering the need to deploy BA as such after recognizing the frequency and type of co-occurring mental health symptoms that complicated our standard delivery of typical protocol-driven approaches for adolescent depression treatment. We found that the impairments related to common co-occurring difficulties, including symptoms of anxiety, oppositionality, and ADHD, could be readily organized alongside the depression conceptualization using an FBA model of behavior. Specifically, FBA and the principles of operant conditioning highlight that the onset and maintenance of behaviors can be understood by identifying their antecedents and consequences. FBA highlights that by manipulating antecedents and consequences we can alter problematic behavioral patterns. For example, avoidance of feared situations, anger outbursts in response to requests by authority figures, and avoidance of tasks requiring mental focus are core features of anxiety, oppositionality, and ADHD, respectively (Barlow, 2004; Matthys et al., 2013; Volkow et al., 2011) and are typically addressed by approaches rooted in FBA (Barlow, 2004; Daley et al., 2014; Kazdin, 2008; Lee et al., 2012). Impairments involving heightened avoidant and reduced approach behaviors (i.e., motivational anhedonia) are core treatment targets addressed in BA by approach-focused goal setting. However, to date, manualized depression treatments have not had the theoretical nor practical flexibility to address functional avoidance attributable to co-occurring symptoms that contribute to the onset and maintenance of depression.

While BA has primarily been studied as a treatment for depression, theoretical work utilizing case studies has attempted to extend the use of BA to the treatment of co-occurring anxiety disorders. Hopko, Lejuez and colleagues (Armento & Hopko, 2009; Hopko et al., 2004, 2006) have proposed that depression and anxiety share functional avoidance tendencies. While there are different proposed mechanisms of symptom change for anxiety (i.e., desensitization and inhibitory learning; Craske, 2015; Geller et al., 2019; McGuire et al., 2016) and depression (i.e., increased experience of positive reinforcement; Takagaki et al., 2016), focus on increasing approach behaviors and decreasing avoidance is common to evidence-based treatments for both disorders. There is initial indication that BA may address co-occurring depression and anxiety (Armento & Hopko, 2009; Chu et al., 2009; Hopko et al., 2004; Jakupcak et al., 2006), yet more research is necessary, particularly in relation to the treatment of functional avoidance symptoms characteristic of co-occurring difficulties besides anxiety.

We propose that BA’s core components of FBA and approach-focused goal setting based on the tenets of operant conditioning could be flexibly used as a principle-based treatment. Specifically, we posit BA principles may effectively address functional avoidance characteristic of depression and other common co-occurring conditions that contribute to the onset and maintenance of depression symptoms. The following sections illustrate ways to apply core BA principles to address symptoms related to disrupted approach and avoidance behaviors observed in conditions that may co-occur with depression including ADHD, social anxiety, and oppositionality (see Table S1 for brief case examples linked to each co-occurring difficulty and the BA Case Example below for an in-depth clinical example).

Clinical assessment

At treatment outset, clinicians use FBA and the BA model of depression (Figure 1) to assess the teen’s context and functional impairments contributing to the onset and maintenance of symptoms. Within a principle-based approach, this assessment additionally includes evaluation of functional avoidance patterns and emotional responses to help manage co-occurring difficulties, such as problems organizing and completing schoolwork, avoidance of peer interactions, or conflict with teachers and parents as highlighted in Table S1. Undergoing a broad assessment of functional avoidance irrespective of diagnostic underpinnings may highlight important onset and maintenance factors that will need to be addressed as part of care. Our national collaborative utilized tools to support broader assessment of functional avoidance patterns and more easily integrate information into a cohesive case conceptualization (see Lewandowski et al., 2022).

BA assessment also includes asking the teen to conduct their own FBA to identify links between specific behaviors and mood using mood and activity tracking. While typically used to assess avoidant and approach behaviors related to depressed mood, teens can be coached to track a variety of behaviors and emotions in order to create personalized treatment goals. As previously noted, functional avoidance associated with co-occurring difficulties, such as avoidance of schoolwork related to inattention or social avoidance related to anxiety, may be important contributors to onset and maintenance of depression. An FBA assessment serves to highlight the interactive effects of low mood and co-occurring symptoms in maintaining avoidant behaviors and preventing positive engagement. Indeed, mood and activity tracking provide the foundation for a discussion about factors the teen believes are most meaningful to improving their mood and serves to facilitate collaboratively identifying a common direction for treatment goals.

Goal- versus mood-directed behavior

BA distinguishes between mood-directed behavior (i.e., when a teen’s behavior follows a mood, whether positive or negative) and goal-directed behavior (i.e., when a teen follows a set goal in line with their values regardless of their current mood). This concept can be adapted to characterize approach and avoidance impairments stemming from conditions co-occurring with depression, including completion of homework at a designated time (ADHD), asking a peer a question (social anxiety) or following a parent or teacher request (oppositionality) with mood being more broadly defined outside of depression.

Short- and long-term consequences of behavior (FBA)

BA draws on FBA to evaluate not only which activities have immediate positive or negative consequences to their mood, but also what are the long-term implications of behavioral choices. Highlighting the short- and long-term positive and negative consequences of behavioral choices guides both the teen and clinician in which behaviors to target that may have the biggest impact on depression symptoms. It further aims to increase adolescent motivation to attempt to address behavior change in the service of improving depression and co-occurring symptoms. As illustrated in Table S1, this FBA approach may be flexibly applied to stressors, emotional experiences, and behaviors underlying common co-occurring symptoms. Evaluation of behaviors related to co-occurring difficulties could also include the short- and long-term consequences of these behavioral choices on depression symptoms.

Goals setting and mini-steps

Appropriately identifying, setting, and following through with approach-focused goals is a core BA skill that can be flexibly applied to treat functional avoidance impairments rooted in co-occurring difficulties common to depression. The clinician aims to help adolescents to choose an attainable goal that is likely to improve their mood. The clinician also assists the adolescent in breaking down this goal into manageable steps that consider impairments related to co-occurring difficulties like disruptions in executive functioning, heightened physiological arousal, and impaired emotion regulation. While goal setting may not appear directly related to treating mood symptoms (e.g., improving quality of homework time), managing impairments related to co-occurring problems is necessary in order to address the broader stress, mood, behavior, consequence cycle contributing to the adolescent’s depression onset and maintenance.

Overcoming avoidance

In BA, avoidance is understood as the primary mechanism for symptom onset and maintenance. Counteracting avoidance, therefore, is the central focus of treatment. Avoidance may take many forms including procrastination, rumination, withdrawal, and emotional outbursts. Thinking through strategies for overcoming avoidance typically occurs in the context of goal setting and identifying barriers to completion, with avoidance considered an “internal barrier” that arises from within the adolescent. In BA, the FBA conceptualization aims to help the adolescent recognize avoidance patterns and to identify goal-directed strategies to improve mood. The clinician’s goal is to support the adolescent in overcoming avoidance using BA skills, such as problem-solving and goal-directed behavior, which readily apply to functional avoidance that characterizes co-occurring difficulties as illustrated in Table S1.

BA as a platform for adjunct treatments

We have highlighted many ways to adapt BA into a principle-based framework to directly address disruptions in approach and avoidance behaviors that cause and maintain depression. Many teens also present with comorbid disorders such as obsessive-compulsive disorder (OCD), eating disorders, and substance use disorders that require alternative or adjunctive treatment outside the scope of BA, such as medication management or other forms of evidence-based psychotherapy. While these may be indicated as a first-line approach dependent on the primary diagnosis, the BA framework may still provide a useful platform for clinicians to conceptualize how comorbid difficulties are impacting depression, provide psychoeducation on alternative evidence-based approaches, and incorporate these interventions into the client’s treatment plan as an approach-focused goal, either through an external referral or by changing treatment focus with the current provider. For example, a clinical assessment using FBA may identify a connection between lack of peer contact and low mood and improved peer relationships with positive mood. An FBA conceptualization could help illustrate the impact of OCD-related contamination fears on the teen’s ability to interact with peers and the corresponding negative consequence for their mood. Specialized, evidence-based interventions for OCD may then be designed to address OCD symptoms that are acting as barriers to the teen’s goal of improved mood. If the adolescent identifies decreasing contamination fears through targeted exposure therapy as a critical approach-focused goal to improve mood, then the clinician and adolescent could work on outlining the necessary steps to achieve this goal, either by outlining steps to find and engage with an external treatment provider or outlining graded exposures with the current provider if the clinician has the necessary expertise to do so.

In sum, we propose that BA can function as a principle-based treatment that addresses anhedonia and impairments in approach-related behaviors that characterize depression in the context of co-occurring problems. When the needs of a teen extend beyond the scope of FBA or approach-based goal setting to address functional avoidance, BA can offer particularly useful tools for identifying the most urgent adjunct needs and defining steps for incorporating additional interventions into a treatment plan.

In-session behavioral activation “exposures”

There is burgeoning support that BA may target the brain’s reward system by focusing on approach-based goals which reinforce behavioral change leading to improved mood (Nagy et al., 2020). Preliminary evidence suggests that BA may induce a change in reward-related neural systems of depressed individuals (Dichter et al., 2009) that can serve to increase learning and promote generalization (Kanter et al., 2008). Given the similarities and interconnection between anxiety and depression (Merikangas et al., 1994; Mineka & Gilboa, 1998), BA treatment may operate in a manner analogous to exposure treatment in anxiety, but largely through the behavioral principle of differential reinforcement rather than inhibitory learning and habituation to anxiety (Hopko et al., 2004).

Behavioral processes have long been understood to mediate CBT (Powers et al., 2017) as reflected in the anxiety literature: exposure leads to inhibitory learning and/or habituation, cognitive reappraisal, and improved self-efficacy (Craske, 2015; Foa & Kozak, 1986; Kindt, 2014). Complementary processes may pertain to the treatment of depression (Kanter et al., 2008). Exposures, as used in CBT treatment for anxiety, facilitate the client’s successful treatment progression (Brown et al., 2001; Franklin & Foa, 2008), and trials can occur in session, thus enabling the clinician to directly observe, manipulate and reinforce the targeted behavior (Kanter et al., 2008). Repeated exposures can then be assigned as homework for completion between sessions, and successful exposure trials may generalize to other situations due to cognitive changes that underlie the effectiveness of the exposure (Kendall et al., 2005). For treatment of either depression or anxiety, overcoming avoidance through behavioral confrontation of avoided stimuli is central to treatment’s success.

We defined in-session BA “exposures” as completion of approach-based goals aligned with a teen’s values, mood improvement aims, and overall formulation. We piloted exposures in-session to explore feasibility and whether they would augment between-session mood-boosting activities to enhance treatment response. Practically, we know from exposure work in anxiety treatment that exposures are most effectively implemented in-session with the therapist. Similarly, completion of in-session BA exposures may facilitate the following: a) allow for a temporally proximal practical demonstration of principles that embody the work and facilitate learning; b) serve to directly tackle the avoidance common to completing approach-based BA goals with the added benefit of increasing the individual’s sense of self-efficacy, mastery, or enjoyment; c) give the opportunity to titrate the BA exposures, which supports treatment response (Busch et al., 2010; Kanter et al., 2008; Nagy et al., 2020); and d) provide the ability to identify safety behaviors or cognitions that could “undo” the work.

Each site in our national collaborative utilized in-session BA exposures to varying degrees depending on clinical and research aims. The specific exposures were chosen based on the adolescent’s idiographic case conceptualization and stage of treatment, with the overarching objective of overcoming avoidance by engaging in an approach behavior during session that aligned with the adolescent’s values and treatment goals. For example, an adolescent with the goal of seeing friends more frequently outside of school might decide on a proposed friend and activity and during the therapy session text that friend. The teen and clinician would then discuss the next step or role-play engaging in that step, such as asking a parent for a ride or replying to subsequent texts to increase likelihood of follow through. When in-session avoidance of or difficulty with the exposure was encountered, the clinician drew from concepts such as engaging in value-based goal- versus mood-directed behavior, weighing the short- and long-term consequences of the behavior, or reevaluating and prioritizing mini-steps to ensure mastery and progress toward the goal. In addition, clinicians underscored an “experimental” approach by noting that observing changes in emotions while taking this action step provides “data” that will help the teen and therapist better understand the impact of the behaviors on mood and functioning.

Given the NIMH site’s research interest in reward processing differences that underlie depression (Stringaris et al., 2015), the NIMH site incorporated client ratings before- and after-BA exposures to examine whether reward prediction error (i.e., difference between expected and experienced enjoyment; Keren et al., 2018) was associated with mood ratings. For example, during in-session BA exposures at the NIMH site, teens were asked to rate their mood (on a 0–10 scale, 10 being best) and rate their expected feelings of enjoyment, anxiety, and satisfaction following the exposure (each on a 0–10 scale, 10 being most). Following the exposure, teens then were asked to rate their experienced mood and feelings of enjoyment, anxiety, depression, and satisfaction. Clinically, if the experienced enjoyment differed from the expectation (i.e., prediction error), the therapist explored the violation of expectancy with the teen by inquiring whether there was anything within the teen’s control, such as focus of attention, that may have impacted enjoyment. This might be followed by questions about what the teen could do for homework to replicate or build on the exposure learning and increase the likelihood of success.

To build off of reported positive therapist experiences with BA exposures, we plan to further evaluate the effectiveness of in-session exposures, such as through improvements in mood or violation of expectancies. Parallel to the inhibitory learning processes underlying exposure treatment for anxiety (Craske, 2015), we also aim to evaluate the underlying mechanism linking use of BA exposures to depression treatment response. The example that follows presents a case study that utilizes BA exposures as part of care. Based on this limited experience, further exploration of in-session BA exposures and its impact on treatment response and outcomes is warranted.

Addressing challenges with motivation for treatment

While motivation to engage in treatment is a difficulty experienced across diagnoses, treatment modality, and treatment setting (Olfson et al., 2003, 2009), motivation to engage in depression care poses a unique challenge given depression’s core features of anhedonia, apathy, and diminished self-efficacy. Several core components of BA may be inherently useful to engaging depressed adolescents in care.

Several BA strategies overlap with aspects of motivational interviewing – an evidence-based technique to improve treatment engagement and adherence (Flynn, 2011; Lundahl et al., 2010) – and may be useful when clinicians encounter client resistance. BA principles allow clinicians to take an idiographic, client-centered approach that aims to use the client’s perspective to build awareness of the challenges arising from depression while simultaneously normalizing and empathizing with these difficulties. Initial FBA assessment work feeds into building the teen’s BA model of depression and assists the teen and clinician in identifying a treatment focus that is most important and meaningful to the teen. Once these directions are identified, BA supports self-efficacy to engage in change behaviors through the concept of mood versus goal-directed behavior while reinforcing any effort toward goal-directed behavior change. Collaboratively generating short- and long-term consequences of behavioral choices further allows clinicians to help the teen to elicit change talk and develop awareness of the discrepancy between their current functioning and stated goals as identified in the FBA. In-session exposures provide additional opportunities to illustrate concepts, which may build self-efficacy, confidence, and motivation to try goal-directed behaviors outside of session. While BA is not meant to serve as a standalone treatment engagement approach, these overlapping principles and strategies may support clinicians in better understanding and overcoming low treatment motivation.

Telehealth considerations

The rapid increase in telehealth delivery of psychosocial interventions due to the COVID-19 pandemic provided an opportunity to capitalize on the benefits of digitally delivered therapy (Torous et al., 2020). While telehealth delivery may preclude the inherent activation of leaving the house to attend therapy, clinical observations drawn from our national collaborative and colleagues suggest that many in-person activation strategies can be readily adapted to a telehealth format. For instance, clinicians may help teens complete a joint in-session BA exposure, eliciting the client’s mood rating before and after to highlight to the client the connection between activation and mood. Mood-boosting activities on telehealth may include online two-person games, learning a new skill such as knitting or another craft, completing a shared activity such as drawing, watching a mood-boosting video or listening to energizing music, or asking the teen to share with the clinician a pet or object of special interest. Additionally, clients can be supported in-session to take incremental steps toward goals, such as by scheduling a movie-watching party, completing an internet search related to hobbies or online groups of interest, or scheduling task completion for a school project. There may even be added benefits of in-home telehealth delivery such as supporting the teen to gather materials in the home that are necessary for goal completion such as laying out exercise clothes and shoes for convenient access and making a reinforcement plan to encourage follow through. When clients have outdoor space to talk without concerns of breaching confidentiality, some clinicians have scheduled virtual sessions by smartphone so that clients can use the therapy session as impetus to get outdoors and explore whether changing one’s environment may improve mood.

Overall, there are numerous creative ways that telehealth can facilitate what is central to BA – bolstering clients’ experience of active coping rather that passive or avoidant-based coping. In-session tasks can be planned to support the client in countering rumination through goal-directed action whether the session is conducted in-person or by telehealth.

Behavioral activation case example

The following case example highlights BA as an idiographic and client-centered treatment. This client, whom we will refer to as “Sarah,” was a research participant in a longitudinal study of adolescent depression. The study enrolled 11–17 year-olds in longitudinal neuroimaging research in which adolescents with clinical depression could elect to receive a standardized, yet client-tailored, course of BA. Sarah and her parents responded to recruitment materials; Sarah met criteria for past and current major depression, and she and her parents consented to and enrolled Sarah into the longitudinal research study that included a 13-week course of BA. All aspects of the research and treatment programs received approval from the Institutional Review Board.

Demographic data

Sarah was a 14-year-old, gay, White, cisgender female who lived with her biological parents, three siblings, and family dog in an upper-middle class suburb. Sarah’s parents noted that prior to experiencing depression Sarah was a conscientious student and athlete, socially reserved but always with a couple of close friends. Sarah’s medical history included good overall health with recurrent major depressive episodes as of age 11, and past and ongoing generalized and social anxiety disorder. Family mental health was significant for parental depression, anxiety, and attention deficit disorder, and sibling anxiety, learning disability and disruptive mood dysregulation disorder. Sarah described precipitants to depression as moving homes and changing schools, which resulted in increased anxiety and isolation. Additionally, she believed that her parents were not accepting of her queer identity, as their response to her initial disclosure was that she was just trying to fit in with a queer peer group at her new school.

Risk and safety assessment

Prior to enrolling in the study, Sarah’s parents had learned that Sarah had engaged in self-injury by cutting her thighs when overwhelmed. Sarah’s parents sought supportive counseling and psychiatric services for Sarah at that time, and she was started on a selective serotonin reuptake inhibitor (SSRI), escitalopram, which was titrated to 20 mg daily.

At the time of study enrollment, Sarah reported that she had been free from self-injury for over two months. She endorsed weekly thoughts of wanting to die but denied suicidal ideation, plan, preparations or intent, citing the protective factor of her friends and her dog. Given the history of self-injury, the thoughts of wanting to die, and the additional risk factors of sexual identity and lack of family support, Sarah’s initial treatment focus was to develop a safety plan with her clinician that together they shared with her parents, and subsequent parent- and parent–child sessions were planned. Her clinician helped Sarah to recognize the precipitants to thoughts of wanting to die or urges to self-harm, worked with her parents to reduce access to impulsive means, and worked together with Sarah and her parents to create opportunities for alternative behaviors when feeling overwhelmed, unhappy, or lonely. The parent-only sessions focused on improving parental support and validation of Sarah’s identity and strengths, and the parent–child sessions helped Sarah practice communicating her feelings and needs and supported her parents in hearing her feelings and responding with validation.

Course of treatment

Sarah completed 13 sessions of BA based on the Adolescent Behavioral Activation Program (McCauley et al., 2016). Symptoms were monitored through self-report measures including: the Mood and Feelings Questionnaire-Short Form (MFQ-SF; Costello & Angold, 1988), a 13-item questionnaire, each item scored 0–2, with a clinical cutoff of 12 for depression; the Snaith-Hamilton Pleasure Scale (SHAPS; Snaith et al., 1995), a 14-item questionnaire, each item scored 0–3, with higher scores suggestive of higher levels of anhedonia; the Self-Report for Anxiety Related Emotional Disorders (SCARED; Birmaher et al., 1997), a 41-item questionnaire, each item scored 0–2, with a clinical cutoff of 25 for generalized anxiety; and the Liebowitz Social Anxiety Scale (LSAS; Liebowitz, 1987), a 24-item questionnaire, each item scored rated 0–3, with a clinical cutoff of 30 suggestive of social anxiety. At treatment outset, Sarah’s self-report scores were all above established thresholds for impairment: MFQ = 15 (moderate depression), SHAPS = 25 (elevated anhedonia), SCARED = 42 (elevated anxiety), and LSAS = 74 (elevated symptoms of social anxiety).

Functional behavioral analysis (FBA)

Sarah and her clinician discussed, in adolescent-friendly language, the precipitants and functions of her past and current coping behaviors, and the short- and long-term consequences. Sarah’s clinician validated Sarah’s experience and feelings, and simultaneously taught Sarah to examine the immediate and subsequent impacts of her actions (i.e., whether her behavioral coping strategies were helping her to access mood-boosting reinforcement or punishment).

Sarah identified contextual triggers to her depression, distressing emotions, and avoidance-based coping responses. She came to recognize that while avoidance served her in the short-term (e.g., by wearing ear buds at school she felt less self-conscious avoiding interactions with peers, or by spending hours on her phone she avoided the frustration of challenging homework), in the long-term these behaviors resulted in increased stress as she was isolated from peers, was getting poor grades and was increasingly in conflict with her parents and siblings, which led to the “downward spiral” of loneliness, guilt and shame and behavioral withdrawal. Additionally, when reviewing her risk history, she identified that cutting her thigh with scissors had been an attempt to “feel less overwhelmed” but actually made her feel more anxious and ashamed. Better understanding the connections between her feelings and behaviors assisted in the creation of core treatment goals as outlined below.

Goal setting, mini-steps, and in-session exposures

Through the BA therapy sessions, Sarah worked to replace avoidant behaviors with approach-focused goal setting. She identified her core values and her associated goals of having more time with friends, completing school assignments, and improving relationships with family members. Each goal was broken down into achievable mini-steps and practiced in multiple ways in- and between-sessions. For example, to increase time with friends, Sarah had to first talk with her parents about scheduling and transportation, and then had to initiate contact with peers. Because she was anxious about talking with her parents, something that often resulted in misunderstanding and argument, a joint parent–child session was scheduled to facilitate the conversation. Likewise, as she was anxious about texting peers, this was incorporated into a session. Additionally, she practiced role-plays with her clinician to prepare for follow-up interactions when she encountered peers at school.

Barriers to the mini-steps were also discussed and together Sarah and her clinician planned in- and between-session practices to overcome the barriers. For example, barriers to talking with peers at school included her anxiety in noisy places, and her fear of making eye contact and talking. During sessions she identified mini-steps of walking around the bustling clinical center while 1) not wearing headphones, 2) making eye contact and saying, “Hi,” to strangers, and 3) approaching and asking general questions of staff.

Through identifying goals and mini-steps, Sarah helped to develop her own between-session practice assignments; for example, to help her with her time management and schoolwork goals, she planned mini-step assignments such as breaking homework into smaller chunks, building in reinforcers, and packing her lunch the night before school to help her with her school morning routine. Practice assignments were written down in-session then were reviewed or practiced at the start of the following session to reinforce compliance and motivation. When Sarah reported that low motivation was a barrier to homework compliance, she was able to discuss this with her clinician and they problem-solved together by exploring and targeting other factors that were impacting her motivation (e.g., frustration with parents, low self-efficacy with siblings, poor sleep hygiene and fatigue, disorganization with homework) and, at other times, by reassessing and refining her goals.

As Sarah accomplished her mini-steps and ultimately her goals, she reported improved motivation, along with improved mood and confidence. Sarah reported that she began to feel better about herself through completion of between-session homework, and she and her parents reported improvement in her family relationships and approach to schoolwork and peers.

Symptom monitoring and overcoming avoidance

Within 5 weeks of starting treatment, Sarah’s symptoms of depression and generalized anxiety had improved based on standardized measures and she continued to refrain from self-injury. She endorsed ongoing social anxiety which impeded her enjoyment of rewarding activities, such as meals out with family or friends and ordering at restaurants. After assessing functional avoidance related to symptoms of social anxiety, Sarah and her clinician returned to her FBA and identified rumination and withdrawal as her primary ongoing forms of avoidance. Sarah began to understand the role of anxiety and avoidance within her BA/FBA framework, learned to anticipate and recognize her triggers, and developed a plan for alternative coping behaviors. She implemented problem-solving and mood/anxiety management strategies and identified a range of behavioral alternatives to rumination and withdrawal, such as taking her dog for a walk, having a video call with a friend, or moving toward anxiogenic but valued actions such as enrolling in a new after-school activity. With her success she sought other social opportunities, such as meals out with friends, which in turn increased her enjoyment and sense of competence.

Conclusion of treatment

At treatment termination in week 13, Sarah’s depression and generalized anxiety disorder were in remission, and social anxiety was in partial remission (MFQ = 5 [normal range/subthreshold for depression], SHAPS = 4 [hedonic/low anhedonia], SCARED = 14 [normal range/subthreshold for generalized anxiety], LSAS = 41 [low range for probable social anxiety symptoms]). She had refrained from self-injury throughout treatment, and, as of week-5 through treatment completion, no longer endorsed thoughts of wanting to die nor urges for self-harm. She continued her involvement in enjoyable extracurricular activities and planning time with friends. Clinical observation suggested that Sarah had gained greater resilience to adversity. For instance, when her dog died toward the end of treatment, she was able to use BA strategies to reach out to friends and family for support rather than isolate – which may serve to buffer against future depressive relapse. Treatment concluded by helping Sarah to recognize her gains and new repertoire of skills, make plans to prevent relapse and to continue to address symptoms of social anxiety through a referral for CBT for anxiety. At her 1-year follow-up research visit, Sarah had sustained remission from depression and continued to endorse positive connection to friends, school, and family.

Future directions

The work and developments from our national collaborative of clinicians and researchers described in the present and companion paper (Lewandowski et al., 2022) are considered initial steps toward optimizing adolescent depression treatment outcomes. Clinical observations from our work offer several foci of future research that are necessary for determining whether the use of BA as a principle-based framework and in-session BA exposures improve intervention outcomes.

Randomized clinical trials are necessary to understand whether either a principle-based BA intervention framework or adding structured in-session BA exposures optimizes outcomes for adolescents diagnosed with depression. While there is evidence for the efficacy of BA in adolescents (i.e., performance of BA under ideal circumstances), there are currently no effectiveness, or pragmatic, clinical trials (PCT) of BA with adolescents to our knowledge. Previous work on the population impact of psychotherapy has highlighted the importance of studying psychotherapy interventions in more representative samples (Rothwell, 2005; Tunis et al., 2003; Zatzick et al., 2011). This is an important gap as efficacy trials utilize homogeneous sampling and strict adherence to session-by-session protocols that hinders the ability to generalize findings to real-world clinical populations that are comprised of adolescents with depression and a range of co-occurring problems (Liebowitz, 1987). One barrier to conducting PCTs may be the lack of protocols that allow clinicians to flexibly address impairments related to common comorbidities. It is possible that PCTs of BA may be optimized by utilizing a flexible, principle-based approach as outlined that would allow clinicians to address related comorbidities with greater ease. Importantly, PCTs should measure improvements across a range of symptoms and functional impairments outside of depression symptom improvement. It may be particularly useful to conduct this research within the context of applying BA as a principle-based framework to a unified system of depression management as described in the companion paper (Lewandowski et al., 2022). Relatedly, while we have begun this work as part of our implementation model, further development and research examining a principle-based training approach to BA will be a necessary concurrent step to this proposed work.

Using BA as a principle-based framework requires the underlying assumption that shared impairments related to approach and avoidance behaviors are key markers for improvement in depression and related comorbidities in adolescents. However, this has not yet been empirically tested. Additional or concurrent work examining behavioral and neurobiological mechanisms of treatment response related to approach and avoidance behaviors is needed to determine whether BA functions to address these underlying, shared, risk mechanisms and whether improvement in approach and avoidance behaviors is a marker for treatment response. In particular, there is a need to empirically test whether or not inclusion of in-session BA exposures improves depression outcomes specifically via increased approach behaviors. It will be important to address whether approach behaviors are a mechanism of BA treatment response across multiple levels of analysis by including: 1) assessment of approach and avoidance behaviors across self- and parent-report; 2) neural function in regions of the brain associated with reward processing; and 3) in-session measurement of physical, social, and activity engagement. It may be particularly useful to draw on the ubiquity of mobile technology use among adolescents (Madden et al., 2005; Onnela & Rauch, 2016) to measure social (i.e., texts, calls, app usage) and physical (i.e., steps, activity, and sleep) to improve ecological validity of the assessment of “real world” approach and avoidance behaviors. Mobile technologies may also be useful to determine what adolescents are doing (or not doing) on a “real time” basis across multiple contexts via ecological momentary assessment surveys.

Conclusion

There is a clear need to improve depression management for adolescents. The developments stemming from our national collaborative, including the use of BA as a principle-based treatment framework and addition of in-session BA exposures, highlight new practical applications for clinicians and health systems. Our future work will aim to establish empirically whether these approaches improve depression treatment for adolescents.

Supplementary Material

Supplemental data for this article can be accessed on the publisher’s website.