Acceptance and commitment therapy as a transdiagnostic approach to adolescents with different anxiety disorders: study protocol

According to the World Health Organization [1], 14% of adolescents worldwide experience a mental health disorder, but lifetime prevalence estimates have reached as high as 49.5% [2]. Research consistently shows that anxiety disorders (ADs) are the most common mental health conditions during adolescence [2‐8], with the age of onset usually occurring in childhood or adolescence [9‐11]. The presence of ADs during those periods has been linked to an increased likelihood of experiencing anxiety during emerging adulthood [12], and to a greater risk of presenting other disorders (e.g., substance and alcohol abuse/dependence, or major depressive disorder,[13]. Adolescent ADs have also been linked to significant short and long-term impairments across a broad spectrum of life domains [13‐16]. Experiencing ADs during adolescence can hinder the acquisition of crucial competencies to adolescents overall development that can shape their future lives and lead to negative outcomes in adulthood [17]. Hence, increasing the understanding of the common mechanisms underlying mental health problems, with a particular focus on different ADs in adolescence, and how these mechanisms can be used to sustain efficacious psychological interventions, is a crucial research concern.

Indeed, commonalities in psychological processes underlying human suffering and psychopathology have been increasingly considered, at both conceptual and therapeutic levels (e.g., [18, 19]), as alternatives to considering the distinct mechanisms associated with specific disorders. This shift in perspective has been driven by multiple factors, including empirical evidence from epidemiological studies pointing to high comorbidity rates among mental health diagnoses both in adult [10] and adolescent [8, 20] samples. Additionally, the tendency of mental health problems, including ADs, to predict and cascade both within and between disorders throughout individuals' lives has been consistently reported in the literature (e.g., [14, 21]). While disorder-specific interventions (i.e., interventions focused on a single diagnosis) seem to yield some unintended improvements in non-targeted comorbid disorders [22, 23], they may not comprehensively address the complex interplay and dynamics observed among mental health disorders. This overlap and continuity among and within classes of disorders indicate the presence of common underlying mechanisms, thereby suggesting that treatments should be conceived to address broader processes rather than focusing solely on the specificities of certain disorders.

Transdiagnostic approaches aim to identify common elements that cut across mental health disorders, thus better reflecting the complexity and dimensionality of the human experience, and more accurately representing the reality of mental health problems. These approaches may offer several advantages. For example, transdiagnostic interventions may be more efficacious in the treatment of comorbid disorders and in preventing future mental health conditions (e.g., [24, 25]). As such, a growing body of empirical research points to the efficacy of transdiagnostic interventions in multiple disorders (e.g. [25‐30]) as well as their equivalence to diagnosis-specific interventions [31]. Moreover, transdiagnostic approaches do not rely on specific diagnosis categories outlined in the Diagnostic and Statistical Manual of Mental Disorders (5th ed., DSM-5; [32]), thus allowing to target symptoms associated with subclinical conditions. Transdiagnostic approaches may also have practical and economic benefits, including shorter treatment lengths and reduced need for future access to mental health services (e.g., [25, 33]). Furthermore, it may be easier to train clinicians in transdiagnostic approaches since a single protocol can be applied to multiple disorders, instead of requiring a separate protocol for each disorder [33].

Acceptance and Commitment Therapy (ACT) is a widely researched and empirically sustained transdiagnostic approach to behavior change (e.g., [18, 34‐36]) addressing both psychopathology (suffering) and psychological health. Within this framework ([37]), psychological suffering (and mental health disorders) arises from the interdependent interactions between six core processes (Cognitive Fusion, Experiential Avoidance, Attachment to the Conceptualized Self, Dominance of a Conceptualized Past/Feared Future, Lack of Values Clarity, and Inaction, Impulsivity or Avoidant Persistence,see supplementary material for a detailed description of each process). Together, these processes sustain Psychological Inflexibility (PI), a rigid pattern in which internal experiences (e.g., thoughts, emotions) dominate over contextual cues, undermining the individuals’ ability to persist in and/or change behavior to pursue chosen values ([38]). Problematic behaviors associated with mental health problems may stem from the engagement in psychologically inflexible strategies (e.g., perceiving thoughts as literal representations of reality – i.e., cognitive fusion -, and/or avoiding specific relevant situations to minimize unwanted internal experiences—inaction, impulsivity or avoidant persistence driven by experiential avoidance) that reinforce one another, becoming more rigid and entrenched over-time, and ultimately narrowing the individuals’ lives. ACT aims to reverse PI processes through the cultivation of Psychological Flexibility (PF), which refers to the ability to be in contact with the present moment, regardless of unpleasant internal experiences, while adapting and persisting in behaviors aligned with valued life directions [39]. PF entails six interrelated processes, each representing the opposite of one of the PI processes: Cognitive Defusion, Acceptance, Self as Context, Contact with the Present Moment, Values, and Committed Action [38], see supplementary material for a detailed description of each process). Through the cultivation of PF, ACT does not aim to change the form or frequency of internal events. Rather, ACT intends to promote psychological health by changing how one relates to those events and reduce their behavioral impact, thus laying the foundation for actions that move individuals towards valued and meaningful living [38]. ACT’s perspective on psychological health is closely tied to the process of individual and idiosyncratic growth, which is defined by the gradually increasing choices individuals make to embrace the present and move towards a life worth living [39]. This perspective can be captured by the concept of flourishing, defined as a state of vitality associated with the ability to engage in functional behaviors, in meaningful relationships and to cope with the demands and stresses of life [40]. Flourishing may thus be a relevant construct indicative of desired behavior change towards valued life directions.

The parallel conceptualization of processes sustaining psychopathology (i.e., PI) and, alternatively, mental health and flourishing (i.e., PF) specified by the ACT model offers a functional dimensional approach to mental health problems. Moreover, it provides a model of treatment that enables the identification of different treatment components and of psychological mechanisms that may facilitate change [39]. Multiple meta-analyses with adult populations showed that ACT and ACT-based treatments were as effective as well-established evidence-based interventions (e.g., Cognitive Behavioral Therapy; CBT), and superior to placebo or waiting list control conditions and treatment as usual in treating depression, anxiety disorders, substance use disorders, somatic health problems and other forms of adult psychopathology (e.g., [35, 41, 42]). Regarding the mechanisms underlying change during ACT, a systematic review conducted by Stockton et al. [43] identified a lack of studies on this matter. The authors pointed that few studies examined the role of specific PF processes (e.g., committed action, values, and self-as-context) as potential mediators of change in ACT. Despite the limited research available, findings from this review suggested that PF, acceptance, and cognitive defusion mediated change in mental health outcomes while non-ACT specific processes (e.g., challenging dysfunctional cognitions) did not. Levin et al. [44] conducted a meta-analysis focused on laboratory-based component studies providing evidence for PF components impact on psychological outcomes. Specifically, significant positive effect sizes were observed for conditions promoting acceptance, defusion, present moment awareness, and values, compared to inactive conditions. Further sustaining the PI/PF model assumptions (i.e., the aim is not to directly change the form or frequency of internal events [45, 46]), larger effect sizes were observed for primary targeted outcomes (e.g., willingness to reengage in a difficult task) than for overall non-specific outcomes (e.g., intensity/frequency of negative thoughts and feelings). Network analysis studies also highlighted the theoretically proposed interconnectedness between PI/PF processes [47], and their association with mental health outcomes such as well-being/quality of life, internalizing symptoms [48, 49] and PTSD symptoms [50]. However, findings were inconsistent regarding the centrality of each PI/PF process in the network (e.g., [47, 49]). Considering the cumulative evidence, the ACT model seems to offer a suitable framework to conceptualize and intervene in adults’ suffering and mental health conditions.

Nevertheless, the role of PI/PF in adolescents’ mental health remains poorly understood, and little is known about the efficacy of ACT in this age group. Previous studies support the role of PI processes in adolescents’ suffering and psychopathology (e.g., [49‐54]), as well as the role of PF in adolescents’ well-being [55, 56] and meaning in life [57]. These studies relied mostly on the Avoidance and Fusion Questionnaire—Youth Version (AFQ-Y), which is a widely used measure to assess PI and explore its relationship with psychopathology related variables (e.g., [58]). However, the AFQ-Y assesses PI uniquely in relation to cognitive fusion and experiential avoidance [59], and so the literature has been amiss in considering all six core processes of PF/PI. Failing to consider all these processes may result in a less specific and idiosyncratic perspective on the theoretical and applied suitability of PI/PF to adolescents' mental health.

At the therapeutic level, a meta-analysis by Fang & Ding [34] examined 14 studies with children and adolescents, and found support for the efficacy of ACT in multiple mental health problems. ACT outperformed waiting list control conditions and treatment as usual, and did not significantly differ from evidence based interventions (i.e., CBT) in improving negative outcomes such as anxiety and depression. Promising results have been found regarding ACT efficacy on adolescents mental health problems such as ADs [60], depressive symptoms [45, 46, 61, 62], obsessive–compulsive disorder [27‐30, 63], social and school adaptation [64].

Despite the potential of ACT for promoting adolescents’ mental health, further research is needed to confidently establish its efficacy. Specifically, there is a scarcity of methodologically robust designs, including Randomized Controlled Trials (RCTs), investigating ACT interventions in this population (see [27‐30, 65] for a review). Also, adolescence is a period marked by significant and rapid physical, psychological, and social changes [66], which underscores the importance of studies considering extended post-intervention follow-up periods to fully assess the utility of ACT for this age group. Moreover, because ACT's efficacy on different non-comorbid adolescent mental health problems is not clearly established, there is a gap in our understanding of ACT's transdiagnostic intervention principles (i.e., that targeting the same processes is helpful for different mental health difficulties). This gap hinders our ability to further address other complex issues in adolescent mental health, such as comorbidity and the cascading effects of mental health problems.

In light of this, it is important to understand how conceptualizations can better inform interventions to increase the well-being of adolescents with diverse ADs. Particularly, SAD and GAD are ADs that hold significant prevalence rates in adolescence: 12-month prevalence estimates of 4.8% [20] and point prevalence estimates of 9.4% [67] have been reported for SAD,12-month prevalence estimates of 2.1% [20] and 6-month prevalence estimates of 9.7% [3] have been reported for GAD. Both these disorders tend to reveal a chronic course that may fluctuate by evolving into other ADs or mood disorders in adulthood [21, 68, 69]. However, transdiagnostic approaches to these disorders in adolescence, particularly GAD, are still poorly investigated at both conceptual and therapeutic levels. ACT has been proven effective for SAD and GAD treatment in adults [68‐73]. To our knowledge, only few studies pointed to ACT efficacy for SAD [60, 74, 75], and only one considered a sample that included adolescents with GAD [60]. Thus, research on the efficacy of ACT on adolescent SAD and GAD is still largely missing, and mechanisms underlying change have not yet been investigated.

This project is funded by the Portuguese Foundation for Science and Technology (Identifier: 2022.13986.BD) and is registered at ClinicalTrials.gov (Identifier: NCT05906849). All procedures were approved by the Ethics Committee of the Faculty of Psychology and Educational Sciences of the University of Coimbra. Eventual protocol amendments will be communicated to this Committee.

Adolescence is a critical phase of human development where mental health problems, particularly ADs, are highly prevalent (e.g., [3, 4]) and often persist and/or develop into other disorders throughout an individual's life (e.g., [14, 21]). Even if growing evidence points to the importance of understanding the common mechanisms underlying mental health conditions (e.g., [31]), research focusing on adolescents is worryingly limited. Indeed, while PI/PF processes have been largely investigated and ACT has demonstrated to be effective for multiple conditions in adults (e.g., [35, 42]), adolescent populations have been consistently overlooked (e.g., [27‐30]). Thus, this project aims to explore the conceptual foundations and therapeutic application of ACT in this pivotal phase of human development. We will delve into the relationships between PI/PF processes, anxiety and flourishing in various adolescents’ samples, and investigate the efficacy of ACT for GAD and SAD in this developmental life stage.

To the best of our knowledge, this is the first project that proposes to comprehensively examine all PI/PF processes in testing ACT assumptions as accurate theoretical conceptualizations of anxiety and flourishing in community adolescents, resorting to network analysis (Study I). Drawing on the theoretical propositions and studies using network approaches to explore associations between PI/PF and mental health outcomes (e.g., [47, 49]), we expect PI processes to display interrelationships and associate positively with anxiety and negatively with flourishing in adolescents from the community. Similarly, PF processes are expected to interconnect with each other and associate negatively with anxiety and positively with flourishing in that community sample. By exploring the strongest and most influential ACT processes in the networks, we will be able to better understand the unique contribution of each PI/PF process within a system, either to flourishing or to anxiety. For example, understanding which process may influence a greater number of other processes, or more strongly impact specific mental health outcomes (e.g., anxiety, flourishing), may help clinicians to increase interventions’ impact. Furthermore, differences in the paths linking PI/PF processes to anxiety and flourishing in clinical (SAD and GAD) and mentally healthy samples of adolescents (Study II) will be explored by testing model invariance. We expect similar pathways linking PI/PF processes to anxiety and flourishing across all groups (i.e., SAD clinical, GAD clinical and mentally healthy groups), with distinct mean levels observed between the clinical and the mentally healthy groups. This would be in line with existing research with adolescents supporting the associations between PI/PF and positive (e.g., [55, 57]) and negative (e.g., [52, 54]) mental health outcomes in clinical (e.g., [53]) and non-clinical samples (e.g., [51]), though not specifically applied to GAD and SAD in adolescence. Moreover, if confirmed, these results will further sustain the role of PI/PF towards adolescents’ mental health, also under a transdiagnostic framework. By investigating specific associations between all PI/PF processes and anxiety as well as flourishing across clinical samples, this research project will provide valuable insights into the necessity for process sensitivity assessment, which may contribute to the development of more tailored and, hopefully, more impactful and idiosyncratic interventions for adolescents.

Additionally, this project will investigate the efficacy of ACT delivered online via videoconferencing to adolescents with SAD and GAD, compared to a control group. The online delivery of ACT is a relevant strength of this research project. The efficacy of full synchronous online ACT interventions delivered to adolescent SAD and GAD have not yet been investigated. However, this method of delivering interventions may be particularly relevant to adolescents as it is more accessible and it may help to circumvent obstacles to treatment seeking such as autonomy (i.e., being dependent on parents/adults to drive them to clinics), time and money spent on commuting and fear of stigmatization when seeking in-person psychological support [94]. Moreover, adolescents spend more time online than any other age group, and so online interventions have the potential of being a well-received method, thus enhancing the delivery of mental health care within this adolescence [95].

By using a robust RCT design to investigate the efficacy of ACT, this project will address the scarcity of methodologically robust designs examining that topic specifically in relation to mental health conditions in adolescents (e.g., [27‐30, 65]). We will also compare the efficacy of the same ACT intervention between GAD and SAD clinical groups, expecting similar effects on outcome measures. Moreover, we will investigate the maintenance of intervention gains over time, which will offer a better understanding of the utility of ACT during a period of significant and rapid physical, psychological, and social changes. If the intervention is found to be efficacious, this project will support the future delivery of ACT interventions to adolescent ADs as a reliable transdiagnostic alternative, and as an added option for those who do not achieve desired results from standard treatments. This may offer clinicians the opportunity to choose interventions based on adolescents needs, while adhering to evidence-based treatments and benefiting from the practical and economical implications of transdiagnostic approaches in clinical practice.

Finally, to the best of the authors' knowledge, this project will be the first to explore mechanisms of change following an ACT intervention for adolescents with SAD and GAD. We will consider all PI/PF processes, providing a more comprehensive examination of potential change mechanisms. This finding will further support the theoretical consistency of the ACT model and shed light on the processes responsible for change, thereby expanding the reach and benefits of ACT interventions. Based on previous research on the relevance of the ACT processes [43, 44] and on the transdiagnostic model underlying ACT (e.g., [37, 38]), we expect changes to those processes to predict changes in flourishing and anxiety symptoms similarly in both clinical intervention groups. From an ACT perspective ([39, 38]), adaptative behavioral change results from reducing inflexible patterns in which private events heavily influence action taking, distancing individuals from a valued living (PI) through cultivating the ability to fully contact the present moment as it is and persist in and/or change behavior in ways that bring meaning and vitality to one’s life (PF). Therefore, we expect decreases in PI processes and increases in PF processes to account for higher levels of flourishing and lower levels of anxiety following the intervention. This expectation aligns with the clinical features of many anxiety disorders, such as GAD and SAD, where avoidance of anxiety-triggering situations and rumination about future events are key maintenance factors encompassed within PI processes such as inaction, impulsivity, or avoidant persistence, and dominance of the conceptualized past/feared future. About specific PI/PF process, equal importance is attributed to them within the ACT framework, as the inflexible/flexible patterns of responding to internal experiences and contextual cues and engaging in unworkable/workable behaviors result from the interdependent interactions among them that sustain overall PI/PF ([38]). For instance, helping adolescents see their thoughts as mental events rather than literal representations of reality (cognitive defusion), may help foster present moment awareness which, in turn, creates space for them to choose to avoid less anxiety triggering situations and take more contextual useful and values guided actions [96]. At the empirical level, it remains unclear the extent to which each individual PF/PI process may be more or less important to promote change, particularly while accounting for all model processes (see [43] for a review). Thus, we did not make specific predictions about which PI/PF processes would most influence change in anxiety or flourishing, as all processes are expected to contribute to change.

This study presents some foreseen limitations that are, nevertheless, necessary for its viability. Firstly, it focuses solely on SAD and GAD, disregarding other ADs that also present significant prevalence rates and impairments during adolescence. Additionally, the control group will not receive intervention within this project. To ensure these adolescents receive appropriate care, they will be assessed after completing the control condition and referred to school psychology services if difficulties persist. They will also be explicitly informed about the possibility of discontinuing their participation in the study to seek treatment, without any consequences. It is also important to note that the control group may not be entirely comparable to either of the intervention groups, which could lead to inherent disparities and the introduction of potential confounding variables. Based on ACT transdiagnostic premises, we believe that these differences will not significantly influence our findings. Nevertheless, the authors have devised a contingency plan to address this potential limitation, if necessary (i.e., gathering two independent control groups: one consisting solely of adolescents with SAD and another comprising only adolescents with GAD). We also acknowledge that by only using self-report measures, data collection may be susceptible to the effects of social desirability, and only a limited comprehension of the psychological constructs assessed is provided. Moreover, we anticipate dropouts to be a significant challenge in this study. To prevent its impact, the research team plans to recruit more participants than the minimum needed for each condition. This approach will allow us to account for potential attrition and ensure that sample size remains adequate throughout the study, maintaining statistical power. Efforts will be made to engage and motivate participants throughout the duration of the study to maximize retention rates (e.g., by sending thank you messages after each assessment moment). Lastly, we acknowledge that by not considering a group with comorbid GAD and SAD, we cannot fully elucidate the applicability of transdiagnostic principles to adolescents. Future research should consider comorbidity and/or cascading effects of mental health difficulties as relevant outcomes to better address the complex issues related to ACT as applied to adolescent mental health.

Albeit these anticipated limitations, we believe that this project holds great relevance in advancing knowledge of transdiagnostic processes underlying psychological functioning in adolescents. It also has the potential to inform and enhance the efficacy of interventions for this population. ACT shows promise as a therapy for adolescents with ADs (e.g., [60]), as its underlying transdiagnostic premises may help mitigate the impact of present and future mental health conditions. By exploring PI/PF processes in both clinical and non-clinical adolescent populations and testing the efficacy of ACT along with mechanisms of change, we will gain a better understanding of how (i.e., through which specific PI/PF processes) and for whom (i.e., GAD and SAD) ACT can contribute to adolescents’ mental health. This knowledge will ultimately contribute to the development of more effective and tailored interventions for this population, with the potential to improve their well-being and long-term mental health outcomes.