Introduction

Youth anxiety is prevalent and, unless treated successfully, associated with a host of negative outcomes, including chronic anxiety, depressive disorders, risk for suicide and substance use [1,2,3]. Randomized controlled-trials such as the Child Anxiety Multimodal Study (CAMs) have helped establish Cognitive Behavioral Therapy (CBT) and combined treatment of CBT and Selective Serotonin Reuptake Inhibitors (SSRIs) as the gold standard treatment for youth anxiety [4]. CBT has been consistently shown to be effective in a wide variety of youth internalizing disorders [5]; however, as many as 40–50% of youth do not demonstrate optimal response [6] and relapse is common, even among responders [7].

To improve treatment response, researchers have sought to distill the most effective components of treatment [8, 9], and examine treatment augmentations [10, 11]. Through this process, exposure therapy has gained support as the most effective treatment component in CBT for youth anxiety [8, 9]. Exposure refers to having an individual intentionally confront a feared stimulus or situation [12]. There are several theories that describe hypothesized mechanisms of exposure, such as emotional processing theory [13, 14], and learning theories, (i.e. inhibitory retrieval [15, 16]). Central to both theories is the belief that effective exposure is predicated on prolonged and intentional engagement with the feared stimulus [17]. Supporting these theories is evidence that exposure therapy is especially effective when individuals engage with distressing stimuli, and approach and engage with more difficult exposures [18, 19]. Stated simply, effective exposure requires approaching fear provoking stimuli, and persevering in the face of distress [20].

Augmentation for Exposure Therapy

Though exposure appears to be the most effective treatment component for anxiety among youth, treatment response rates and relapse rates remain suboptimal [6, 7]. Thus, it is perhaps unsurprising that numerous augmentations to exposure therapy have been examined. Although a complete review of augmentations for exposure therapy is beyond the scope of this paper, some enhancements examined to date include drug augmentations such as d-Cycloserine, which is thought to enhance memory consolidation [21], as well as scopolamine [22] and caffeine [23]. Additionally, several non-pharmacological augmentations have been explored, including virtual reality [10], aerobic exercise [24], attention training [25], power posing [26], and strategies informed by inhibitory learning/retrieval principles [11, 17]. Among youth, there is limited support for d-Cycloserine for treating anxiety [27]; however, augmentations based on principles of inhibitory retrieval, such as decreasing safety behaviors and violating expectancies, do appear to enhance the effects of exposure [28]. Still, much room remains for optimizing exposures to improve outcomes.

Barriers to Effective Exposure Therapy

Whereas specialty clinics tend to prioritize exposure therapy, community-based clinicians tend to favor relaxation-based strategies when treating anxiety, citing concerns about the acceptability of exposure therapy to patients [29]. However, relaxation strategies do not appear to provide the same benefit as exposure therapy [8, 9, 30]. Community-based clinicians are the first line treatment option for many youth; thus, concerns about the acceptability of exposure therapy among both clinicians and patients must be addressed. Additionally, theory and research suggest that exposure is most beneficial when individuals engage with distressing stimuli, and approach more than one exposure stimulus [16, 18,19,20]. Such effective exposure practice requires commitment from both a patient and a clinician. Enhancements that target acceptability as well as promote engagement with difficult exposures, as has been suggested by Plaisted and colleagues [28], could address current barriers to effective exposure therapy implementation.

Self-Distancing

One potential strategy for increasing youth engagement with exposure is a strategy called “self-distancing”. “Self-distancing” refers to a small shift in the language of self-talk, replacing first-person pronouns (I.e. “I”, “me”) with second or third-person pronouns (I.e. “you”, “she”, “him”, one’s own name [31, 32]). This strategy reflects natural linguistic shifts often observed in difficult situations in the real world [31, 33]. Research on self-distancing demonstrates that this simple shift is associated with improved performance on anxiety-inducing tasks in adults [31] and improved perseverance in children [33]. White et al. [33] asked 4 and 6-year-olds to complete an aversive “go/no-go” task; children who reflected on the task from a 3rd person perspective (e.g., asked themselves “Is Ethan working hard?”) spent more time working on the task than those who used a 1st person perspective (i.e., “Am I working hard?”). This suggests that even young children may be able to increase their engagement with challenging or aversive tasks using self-distancing.

In addition to perseverance, self-distancing has been linked to other positive outcomes including improved performance on anxiety provoking tasks [31, 34]. A full review of the proposed mechanisms linking self-distancing with such positive outcomes is beyond the scope of this paper [32, 35,36,37]. However, several proposed mechanisms (avoidance and adaptive emotion regulation) may be especially relevant to youth anxiety treatment outcomes [8, 28] and are briefly explored below.

Cognitive distraction may support the initial physical approach of anxiety provoking stimuli during exposure; however, it is likely to be counter-productive in the treatment of clinical anxiety, serving to reinforce cycles of avoidance. Although self-distancing has the word “distance” in the description, research does not support the hypothesis that self-distancing typically increases cognitive distraction or avoidance. A study examined whether self-distancing differed from cognitive distraction in supporting recovery from depressed affect in adults [34]. Initially, both strategies were associated with reduced depressed affect; however, after a delay, those in the self-distanced condition had fewer recurring depressive thoughts as those in the distraction condition. The pattern for the distraction condition: initial relief from distress, followed by a rebound, is what we would expect from the self-distanced condition if it too was supporting avoidance.

Although self-distancing has not yet been tested in the context of exposure therapy, some initial research suggests that self-distancing supports adaptive engagement with anxiety-provoking tasks, rather than facilitating avoidance [31, 38]. Kross and colleagues [31] asked college student participants to engage in a task (Study 2) that evoked social anxiety (i.e., making a good first impression on a stranger of the opposite sex). Prior to initiating the task, participants were assigned to reflect on the task from either a self-distanced or immersive first-person perspective. Participants across conditions did not differ in their pre-task levels of anxiety. However, during the task, masked observers rated those in the self-distanced condition as performing better on the task. Following the task, participants in the self-distanced condition also reported greater reductions in anxiety than those in the immersive condition. These results suggest that self-distancing does not promote avoidance during the task but may rather facilitate adaptive emotion regulation leading to better performances and anxiety recovery after the task.

Indeed, there is mounting evidence that self-distancing does support adaptive emotion regulation both among adult and youth populations [37, 39,40,41,42,43,44,45]. Specifically, self-distancing has been shown to facilitate cognitive restructuring [42] and reduce cognitive rumination [44]. It has been additionally been hypothesized to increase goal salience [33]. This emerging work suggests that self-distancing may provide an adaptive avenue for youth to approach and persevere through anxiety-provoking exposure tasks.

Current Study

Increasing youth engagement with anxiety-provoking tasks in the context of exposure therapy could reduce resistance among both clinicians and clinically affected youth, potentially improving treatment acceptability and outcomes. Thus, there is a clear rationale for examining strategies to promote proximal engagement with fear-provoking stimuli in the context of exposure. This study examines whether a self-distancing augmentation for exposure therapy increases engagement with exposure among youth with anxiety. Participants received sessions of both classic exposure [EX] and exposure sessions augmented with self-distancing [EXSD]. During each exposure session, therapists and participants collaboratively selected the exposure, with a goal of choosing the most difficult exposure the participant was willing to do. Primary outcome variables included youth and therapist report of exposure difficulty and exposure engagement, as well as youth and caregiver report of treatment acceptability. Clinical outcomes were also assessed as an exploratory aim. We hypothesized that visual inspection would reveal that (1) participants would complete more difficult exposures (as measured by participant and therapist report of exposure difficulty on a 0–10 scale) and (2) would be more engaged in exposures (as measured by participant and therapist report of exposure engagement on a 0–10 scale) when completing EXSD exposures, as compared to EX. We also hypothesized that (3) EXSD and EX would be acceptable among youth participants and caregivers. Finally, we investigated an exploratory hypothesis that (4) participants would experience improved clinical outcomes by the end of the study (as measured by youth- and caregiver-report).

Method

Participants

Ten participants (nine completers) were recruited via academic and community clinic referrals, posted flyers, online advertisements, and referrals from affiliated studies. Inclusion criteria included clinically interfering anxiety, a historical clinical anxiety diagnosis (within the last year, either within a research or clinical setting), and at least moderate current symptoms of anxiety by a SCARED total score greater than or equal to 25 as reported by caregiver or youth at baseline assessment. Exclusion criteria included: currently receiving psychotherapy, caregiver or youth report of elevated symptoms of bipolar disorders, psychotic disorders, intellectual disability, severe behavioral concerns, autism, active or recent substance abuse/dependence (remission less than one year), and acute risk due to suicidal intention or behavior in the past six months. Participants were not excluded if they were on a stable dose of psychotropic medication. Current or past depression, obsessive–compulsive disorder, and other posttraumatic stress disorders were allowable, but anxiety was required to be the chief complaint and thus appropriate as the primary focus of intervention.

Among 34 individuals who showed interest and completed phone screens for the study, 13 completed evaluations, and 10 met inclusion criteria. One participant (assigned to Group 1) dropped out after two sessions, requesting a different treatment modality. The remaining nine participants attended all eight sessions. The final sample ranged in age from 11 to 17 years old (M = 14 years, Std. Dev = 2.2 years) and included six females (67%) and three males (33%). Eight participants (89%) identified as Not Hispanic/Latino, one (11%) identified as Hispanic/Latino. Seven participants (78%) identified as White, one (11%) participant identified as American Indian/Alaska native, and one (11%) as Black/African American.

Study Design and Intervention

The study employed a case series methodology with an abbreviated (eight-session) study intervention to allow for initial investigation of feasibility, acceptability, and efficacy in increasing youth engagement in exposures. All participants received eight sessions of exposure-based therapy, with and without self-distancing, albeit in varying proportions. We employed a combined within- and between-subjects design such that each participant received exposures both with and without self-distancing with counterbalancing (ABA/BAB). Specifically, both groups received psychoeducation (sessions 1–2). Group 1 (n = 5) received two sessions of EXSD, followed by two sessions of EX, followed by two final sessions of EXSD. Group 2 (n = 4) received exposures in the opposite order: EX-EXSD-EX (See Fig. 1). Participants were assigned by order of enrollment (i.e., non-randomly) to group and without masking.

Fig. 1
figure 1

Study design

Full size image

The intervention was individually administered and consisted of weekly 60-min individual sessions of exposure-focused cognitive behavioral therapy. All participants received a clinical workbook which their study therapist used to guide them through the treatment procedures. Sessions 1–2 included rapport building, psychoeducation about anxiety, rationale for exposure, hierarchy development, and an introduction to self-distancing (see below for additional details). Sessions 3–8 included exposure therapy with and without self-distancing. Session 8 was dedicated to discussing termination as well. Caregivers were invited to attend the majority of the first and final sessions of the study therapy, as well as portions of additional sessions as needed. The study manual was adapted from the manual described in prior work [30]. See manuscript for additional details. Adherence to treatment condition (Group 1 vs Group 2) and session (EXSD vs EX) was maintained through: (1) provision of a participant workbook with clear instructions to both therapist and participant on the assigned exposure procedure for each session; (2) weekly adherence questionnaire (see TEARS below); (3) weekly supervision with the lead author to discuss treatment planning and fidelity; (4) occasional review by lead author of audio-recorded sessions.

Introduction to Self-Distancing

Youth were introduced to self-distancing in session 2. Participants were provided rationale that self-distancing, or talking to oneself in the third person, might allow them to persist in difficult tasks and increase their motivation, especially during exposures. Specifically, the clinical workbook stated: “We’re going to test out whether using self-distancing can help you practice being brave and doing things that make you anxious or afraid”. The therapist would also demonstrate how to use self-distancing in this context; for example, they might say: “During certain exposures I will ask you to talk to yourself, like a friend, as you remind yourself what fear you are going to face. For example, I might say: “Riley is going to face her fear of awkward situations and making mistakes by calling Target and asking for car parts, which she knows they don't have.” Participants were given ample time to practice self-distancing outside of the context of exposures to learn how to do this and did not move forward with the protocol until they demonstrated understanding.

COVID-19 Procedures

Participants attended therapy session in person prior to March 16, 2020, when the study transitioned to virtual care. Seven participants completed all study sessions in-person. The final two participants (one each assigned to Group 1 and Group 2) completed one session in-person and the remainder of the treatment via video-conferencing software (BlueJeans).

Procedures

The study was conducted at a large academic medical center in the midwestern United States. All study procedures were approved by the institution’s IRB. A legal guardian provided written informed consent and the youth provided assent. Exposure difficulty and engagement were assessed at the end of each relevant session (sessions 3–8). Acceptability of treatment was assessed immediately post treatment. Anxiety symptoms were assessed pre- and immediately post-treatment. Study treatment was delivered by two masters-level therapists trained in this modality. The therapists were naïve to self-distancing prior to participating in the study, and had zero to one years of experience delivering exposure therapy to youth prior to the study. Weekly supervision was provided by the first author, a board-certified psychologist with expertise in CBT and Exposure Therapy, who also ensured clinician adherence to study protocols and interventions.

Measures

The assessment schedule by timepoint is available in Table 1.

Table 1 Assessment schedule
Full size table

Treatment Engagement and Adherence Rating Scale (TEARS)

The TEARS, adapted from a similar measure used in previous studies [46], was completed by the participant and the therapist after sessions 3 through 8. In-session exposures were rated on an 11-point scale (from 0 to 10) along two dimensions: difficulty (“How difficult was this assignment (0–10)?”/“How difficult (from your perspective) was this assignment for this patient (0–10)?”) and engagement (“How engaged in the exposure practice were you (0–10)?”/“How engaged in the exposure practice was the patient (from your perspective) (0–10)?”). Therapists also rated the youth’s adherence to the assigned exposure condition (EX vs. EXSD), from 0 to 10 the following anchors: 0 = used strategy from other condition; 5 = partially used assigned strategy; 10 = was completely adherent to assigned strategy.” For EX sessions, therapists reported high adherence (M = 9.85, Mode: 10, Std. Dev = 0.78; Range: 5–10). For EXSD sessions, therapists reported moderate adherence (M = 7.43, Mode: 8, Std. Dev = 1.48; Range: 5–10).

Treatment Acceptability Questionnaire

The Treatment Acceptability Questionnaire (developed for this pilot study) was completed by both participants and their caregivers. Caregivers and youth were asked the degree to which they agreed with several statements about treatment (Strongly Disagree to Strongly Agree, 7-point scale). Close-ended questions were rated on a seven-point scale ranging from strongly disagree (1) to strongly agree (7). Example prompts include: “[I/my child] received high quality care” and “I wish [I/my child] had not received this treatment” (reverse coded). The youth version of the questionnaire included five additional questions that assessed their experience of the self-distancing strategy, e.g., “I believe that Exposures with self-distancing helped me to face my fears” as well as open ended questions like, “Please share your thoughts about self-distancing.”

Screen for Child Anxiety Related Disorders (SCARED)

The SCARED [47] is a 41-item measure of anxiety symptoms that was completed by participants and their caregivers about anxiety symptoms during the previous three months. It includes scales for the following anxious domains: panic disorder or significant somatic symptoms, generalized anxiety disorder, separation anxiety disorder, social anxiety disorder, and school avoidance. Prompts are rated on a three-point scale, with responses ranging from not true or hardly ever true (0), Somewhat True or Sometimes True (1) or Very True or Often True (2). Example prompts include: “When [I/my child] feel frightened, it is hard for [me/my child] to breathe,” [I/My child] is a worrier,” and [I/my child] worries about going to school. The measure shows robust psychometrics including good internal consistency and discriminant validity across diverse samples [48].

Analytic Plan

Hypothesis 1

Self-distancing will be associated with completion of more difficult exposures.

Visual analysis was considered the primary method of analysis to examine the study hypothesis that exposure difficulty varies by type of exposure (EXSD vs EX). To achieve this, graphs of estimated marginal means, separated by group and session type, were plotted, allowing for visual inspection of exposure difficulty between groups. Visual analysis facilitates assessment of trajectory of behaviors across conditions, and is especially valuable in experimental therapeutic projects such as this one, as the assigned session type changed within group across treatment [49]. Additionally, to directly compare difficulty of EX vs. EXSD exposures, variables were created reflecting an individual’s average exposure difficulty for all EX and EXSD sessions by reporter (therapist and youth). Paired sample t-tests were conducted to compare average difficulty scores for EXSD sessions and EX sessions to examine whether therapists and/or youth reported that youth complete more difficult exposures during EXSD sessions as compared to EX sessions.

Hypothesis 2

Self-distancing will be associated with more engagement in exposures.

Procedures for evaluating Hypothesis 2 are identical to those for Hypothesis 1. To examine the study hypothesis that exposure engagement varies by type of exposure (EXSD vs EX), graphs of estimated marginal means, separated by group and session type, were plotted, allowing for visual inspection of exposure engagement trajectory. Variables were created reflecting average exposure engagement for EX sessions and EXSD sessions for each reporter. Paired sample t-tests were conducted to compare average engagement scores for EXSD and EX sessions to examine whether therapists and/or youth report that youth are more engaged in exposures during EXSD than EX sessions.

Hypothesis 3

Exposures with and without self-distancing will be acceptable to participants and caregivers.

Youth and caregiver responses to the Treatment Acceptability Questionnaire were examined qualitatively to assess whether the EX and EXSD interventions were acceptable. Means, standard deviations, and ranges of responses on relevant questions (i.e., “I believe that Exposures with self-distancing helped me to face my fears”) were assessed and will be reported.

Exploratory Aim: Investigating Clinical Improvement Over the Course of the Study

This study was not powered to examine differential outcomes across the two interventions (EXSD vs EX), thus no intervention specific outcomes are expected. Although the study treatment was abbreviated (8 sessions) and the sample underpowered (n = 9), paired sample t-tests were conducted to explore whether there was a significant change in youth- and caregiver-reported anxiety severity (SCARED total scores) from pre- to post-study.

Results

Hypothesis 1

Self-distancing will be associated with completion of more difficult exposures.

Therapist Report

Visual inspection of the plots of the estimated marginal means of both groups (Fig. 2) indicates that therapists reported that youth completed exposures of greater difficulty during EXSD sessions than during EX sessions. However, there was no significant difference between the means (t(8) = 1.29, p = 0.23; Means and SDs in Table 2) on the paired samples t-test comparing therapist report of average exposure difficulty (EXSD vs EX).

Fig. 2
figure 2

Estimated marginal means of exposure difficulty by therapist and youth report

Full size image
Table 2 Mean difficulty and engagement for EXSD vs EX sessions
Full size table

Youth Report

Visual inspection of the plots of the estimated marginal means of both groups (Fig. 2) indicates that, by youth report, participants completed more difficult exposures during EXSD sessions than EX sessions. However, the paired samples t-test again indicated no significant difference in between the means (t(8) = 1.61, p = 0.15, Means and SDs in Table 2) on the paired samples t-test on youth report of average exposure difficulty (EXSD vs EX). In other words, although it was not significant, youth appeared to complete more difficult exposures by both therapist- and youth-report when using the self-distancing augmentation.

Hypothesis 2

Self-distancing will be associated with more engagement in exposures.

Therapist Report

Visual inspection of the plots of the estimated marginal means of both groups (Fig. 3) suggests that therapists reported higher levels of youth exposure engagement during EXSD sessions than EX sessions. However the paired samples t-test examining therapist-report of average exposure engagement (EXSD vs EX) revealed no significant difference between the means (t(8) = 1.95, p = 0.09, Means and SDs in Table 2).

Fig. 3
figure 3

Estimated marginal means of exposure engagement by therapist and youth report

Full size image

Youth Report

Visual inspection of the plots of the estimated marginal means (Fig. 3) suggests that youth reported high levels of exposure engagement across treatment, regardless of condition. This was further confirmed by the paired samples t-test which revealed no significant difference between the means (t(8) = −0.60, p = 0.56, Means and SDs in Table 2)of youth-reported average exposure engagement (EXSD vs EX). In other words, although it was not significant, youth appeared to be more engaged in exposures when using the self-distancing augmentation according to therapist-report. Youth appeared to have high levels of engagement across both types of exposures (EXSD and EX) according to youth-report.

Hypothesis 3

Exposures with and without self-distancing will be acceptable to participants and caregivers.

Mean responses to the Treatment Acceptability Questionnaire by youth- and caregiver-report were examined to assess acceptability of treatment overall, exposure broadly, and exposure with self-distancing specifically (see Table 3 for all means and standard deviations). Youth and caregivers reported high levels of acceptability for the treatment overall. Specifically, youth and caregivers reported that youth had received high quality care (mean score corresponding with Strongly Agree), that the study treatment was helpful (mean score of Agree and Strongly Agree for youth and caregivers respectively), that they were glad that youth had received the treatment (mean score corresponding with Agree and Strongly Agree for youth and caregivers, respectively), and that they would recommend the treatment to other youth with anxiety (mean score corresponding with Strongly agree). On average, youth and caregivers disagreed or strongly disagreed (respectively) with a statement expressing a desire for the youth to have not received the study treatment.

Table 3 Treatment acceptability questionnaire results
Full size table

Exposure Acceptability

Youth reported believing that exposures helped them face their fears (mean score corresponding with Agree) and indicated that they planned to continue using exposures after treatment (mean score corresponding with Agree).

Self-Distancing Acceptability

Youth reported believing that self-distancing helped them face their fears (mean score corresponding with Agree); however, there was some variability in responses (Std. Dev = 1.72). Youth were mixed on whether they found self-distancing to be awkward and unhelpful (mean score corresponding with Neither agree nor disagree; Std. Dev = 2.18), and whether they would continue to practice exposures with self-distancing after treatment (mean score corresponding with Neither agree nor disagree; Std. Dev = 1.86). Open ended responses to a question about self-distancing (“Please share your thoughts about self-distancing (e.g., what you think the purpose is, what you would like other people to know about it, etc.)”) indicated (1) good understanding of the purpose of the skill, and (2) that many participants felt that self-distancing could be helpful in specific situations or for certain people but may not be helpful for everyone all the time (See Table 3 for all open ended responses to this item).

Given the novel approach to using self-distancing during exposures, we also documented spontaneous observations several participants made about the self-distancing approach, throughout the study period. Some observations were positively-valanced, like “[Self-Distancing] makes me think about my more improved self. Makes me think about the person I want to be,” and “I think that it was easier to have my mind be able to focus on something different, instead of just the exposure and all [anxiety’s] spiels.” Other youth reported that they did not enjoy the self-distancing enhancement, including one who noted, “I’m so happy I don’t have to do that anymore.”

Exploratory Aim

Investigating clinical improvement over the course of the study. Paired sample t-tests were conducted to examine change in anxiety severity on the SCARED total score from pre- to post-study. Youth and caregivers reported no significant change in anxiety from over this interval (Youth: pre-treatment M: 41.56, post-treatment M: 38.22; t(8) = 1.35, p = 0.21; Caregivers pre-treatment M: 38.4, post-treatment M: 32.67; t(8) = 2.13, p = 0.07; Hedges g = 0.41).

Discussion

This manuscript presents pilot data examining whether a self-distancing augmentation for exposure therapy can increase exposure engagement among youth. We found initial, although not conclusive, support for the hypotheses that participants completed more difficult exposures and were more engaged in exposures during sessions with self-distancing than those without. We also found support for our hypothesis that the treatment would be acceptable to youth and to their caregivers. Families were highly engaged in the study therapy and, aside from one participant who dropped out before starting exposures, all participants completed all study sessions. Responses to the treatment acceptability questionnaire were also favorable. However, exploratory analyses did not reveal significant clinical improvement during the course of the study.

Exposure Difficulty

Youth and therapists both appeared to indicate that youth completed more difficult exposures during sessions when the self-distancing augmentation was implemented. The counterbalanced cross-over (ABA/BAB) design allowed for inspection of the sensitivity of the self-distancing manipulation on reported exposure difficulty; indeed, exposure difficulty appeared to increase during blocks when SD was present and decrease when SD was not present. When averaged together youth did not complete significantly more difficult exposures on EXSD sessions as compared to EX sessions; the direction of the findings supported the hypothesis, but the difference was not significant (Table 2). This non-significance may have been due to the small sample in the current trial (n = 9). Taken together, these results provide some initial support for our hypothesis that the self-distancing augmentation supports youth and therapists to select and complete more difficult exposures than classic exposure therapy. If supported in future research, this finding has important clinical implications, given that more difficult and complex exposures are associated with better treatment outcomes (e.g., Peris et al. [19]). Indeed, future research should examine whether self-distancing might allow youth to reach treatment goals more quickly or completely if they are able to approach more difficult targets.

Exposure Engagement

Through visual inspection, we found that therapists, but not youth, reported that participants were more engaged with exposures on sessions where the self-distancing augmentation was implemented. Specifically, exposure engagement increased when the self-distancing augmentation was implemented and decreased when self-distancing was not implemented. Again, this finding was not supported by statistical analyses. By therapist-report, youth were not significantly more engaged in EXSD sessions as compared to EX sessions, although the means supported this hypothesis (Table 2).

Youth did not report increased engagement when implementing self-distancing. Rather, youth-reported exposure engagement increased sharply for both conditions early in treatment (by session 5) and stayed elevated for the duration of treatment, possibly indicating a ceiling effect. Together, this provides modest support for the hypothesis that youth would be more engaged in exposures augmented with self-distancing than traditional exposures.

Engaging with and attending to the exposure stimulus is a necessary condition for inhibitory retrieval/learning, a leading theory guiding exposure work across the developmental trajectory [15]. Increasing engagement with the target stimulus during exposures should in turn reduce avoidance, and potentially have a downstream impact on treatment outcome. Supporting this theory is literature demonstrating that avoidance during exposure hinders treatment outcomes for youth with OCD [50]. Self-distancing may be a novel strategy for increasing engagement in exposures. However, additional work is needed to further confirm this relationship.

Treatment Acceptability

In line with study hypotheses, youth and caregivers reported high levels of treatment acceptability. Specifically, caregivers and youth reported liking exposure therapy and indicated that they would recommend it to other youth with anxiety. Youth reported that exposures without self-distancing were very helpful, and all participants reported that they planned to use them in the future. This is meaningful as there is a documented hesitation among some patients and even clinicians about pursuing exposure-focused therapy [29, 51, 52]. Indeed, one participant in this study dropped out prior to exposures due to concerns about the treatment. However, those participants who initiated exposures were able to complete all sessions and reported that the treatment was both acceptable and helpful. Exposure therapy may suffer from a parallel process of fear and avoidance, in that therapists may simultaneously fear the distress experienced by their patients and subsequently avoid implementing exposure techniques. Strategies like self-distancing could encourage clinicians to approach rather than avoid exposure, thus allowing youth to further approach anxiety-provoking stimuli in turn.

Regarding acceptability of the self-distancing augmentation itself, some youth reported that it was awkward to use self-distancing aloud (i.e., saying out loud “Andrew is going to give a speech”), but did not refuse to do it. In general, youth appeared to believe that using self-distancing was designed to increase courage, allowing them to stick with challenges to achieve their goals. One youth explicitly stated that it helped them think about “my more improved self” during exposures. Youth provided mixed reports about whether they would keep using self-distancing after the treatment was over. Some participants said they would continue using self-distancing, while others noted that they would reserve it for more difficult exposures. Others reported that they thought self-distancing might be helpful for other youth but wasn’t as suited to them.

These results suggest that while acceptability of exposure therapy was high, acceptability of the self-distancing augmentation was moderate. Youth were all willing to try self-distancing, all appeared to understand the purpose of the intervention, and some were interested in continuing to use it in the future to assist them in completing more difficult exposures. This finding was also reflected by our adherence data; therapists reported that some youth had higher levels of adherence during EX sessions (M = 9.85; Std. Dev = 0.78) than during EXSD sessions (M = 7.43; Std. Dev = 1.48). The varying levels of acceptability and adherence for self-distancing may indicate that self-distancing is best suited for specific youth, perhaps those with low willingness to engage in exposures. Future research should examine individual differences into who benefits most from the self-distancing augmentation. Additional research might investigate whether self-distancing is as effective when it is conducted silently (i.e., child states to themselves “Jenn, you’re going to give a speech”) rather than out loud, and whether that increases acceptability, and natural uptake of and adherence to this skill.

Clinical Severity

Exploratory examinations of clinical outcome indicated that caregivers and youth did not report significant reductions in anxiety symptoms from pre- to post-treatment. This may be due to the small sample size (n = 9) and the short course of study therapy (eight sessions). While this design facilitated study goals related to efficiently examining initial feasibility and efficacy of using a novel augmentation to increase youth engagement in difficult exposures, it does not allow us to make conclusions about the downstream impact on clinical outcomes.

Limitations

The current manuscript describes a pilot study that incorporated feasibility and acceptability data. There were several significant limitations that may limit the generalization of findings. First, this study omitted many elements of a traditional randomized controlled trial: there was no randomization to group, evaluators and participants were not masked to condition or study purpose, there were no objective measures of primary outcomes, and acceptability measures were not validated and were completed only at the end of the study. Additionally, all participants were informed about the hypothesized rationale for self-distancing at the beginning of treatment, which may have been leading to participants.

These limitations should be considered when reviewing the results of this study. For example, it is possible that therapists and youth rated EXSD exposures as more difficult because they were aware that self-distancing was supposed to increase willingness to complete more difficult exposures. Objective raters and measures could verify these findings. Additionally, the current study did not define exposure “engagement” for therapists and participants, and it is not known how they interpreted this term as they rated engagement in this study. Future work should provide operationalized definitions of exposure behavior (i.e., approach vs. avoidance) for any informant ratings. Future investigations should also include validated and more frequent measures of acceptability, which could improve interpretability and reduce hindsight bias. It is also recommended that future research employ objective ratings of outcome variables, and potentially include behavioral ratings of youth behavior during exposure, such as the codes validated by Benito and colleagues in the Exposure Process Coding System [53].

Finally, the sample size was very small. The small pilot nature of the trial allowed us to investigate whether there was any potential relationship between self-distancing and exposure difficulty and engagement through visual inspection of plots, but the sample was not sufficient to detect significant differences between groups. A larger scale RCT could better answer the question of whether a full course of EXSD (as compared to a course of exposure focused CBT) increases engagement, approach, and youth ability to complete more difficult exposures, as well as the impact on treatment outcomes.

Self-distancing appears to have potential as an augmentation to exposure for youth. However, mechanisms driving this effect remain unexamined. Future work should examine cognitive factors that could potentially drive these results, such as cognitive distraction, salience of goals, and self-efficacy. Although prior research on self-distancing indicates that it is not likely to facilitate avoidance [34, 36], this requires confirmation within the context of exposure. Evidence is accumulating that self-distancing may instead support adaptive regulation by increasing salience of goals and reducing rumination and distraction [33, 45]. Self-efficacy remains untested in the self-distancing literature, but could also be a mechanism driving perseverance and engagement and is especially relevant to test given emerging evidence of relevance to exposure outcomes [54, 55].

Summary

Exposure is the most effective treatment component for youth anxiety but is not associated with optimal outcomes for all youth. Self-distancing, or shifting self-talk to the second or third person perspective, has been shown to increase perseverance with difficult tasks among youth [33] and engagement with anxiety provoking tasks [31] among adults, but has not been examined in the context of exposure therapy. This pilot study examines self-distancing as an augmentation to exposure therapy. Employing a counterbalanced cross-over (ABA) design, we examined self-distancing’s impact on both the difficulty of exposure tasks and youth engagement with exposure. Although differences were not significant, visual inspection revealed that youth completed more difficult exposures when self-distancing was implemented and less difficult exposures when self-distancing was removed. The same was true for therapist-reported exposure engagement. Youth and caregivers generally found the interventions to be acceptable and reported that self-distancing in particular might be useful in the context of certain exposures, or for certain people. Youth and caregivers did not report a significant improvement in anxiety symptoms in the current trial; this should be further examined in future studies with a larger sample, full course of treatment, and RCT design. Overall, this study provides initial proof of concept for using self-distancing to augment youth engagement with difficult exposures; if this finding is confirmed in future research, this could be one avenue for improving the efficacy of exposure therapy for youth.