Executive Functioning in Pediatric Anxiety and Its Relationship to Selective Serotonin Reuptake Inhibitor Treatment Response: A Double-Blind,Placebo-Controlled Trial

Abstract

Objective:

To characterize executive function in adolescents with generalized anxiety disorder (GAD) and its relationship to treatment.

Methods:

Using data from a double-blind, placebo-controlled trial of escitalopram in adolescents (N = 51) 12–17 years of age with GAD, we used the self-report version of the Behavior Rating Inventory of Executive Function (BRIEF-SR) to assess executive function, at baseline, and examined its relationship to treatment response as measured by the Pediatric Anxiety Rating Scale (PARS).

Results:

For all baseline subscores of the BRIEF-SR, T-scores were significantly elevated in adolescents with GAD compared to an age- and sex-matched normative healthy sample. In escitalopram-treated patients, baseline BRIEF-SR scores for Emotional Control (β = 0.256, 95% credibility interval [CrI]: 0.367 to 0.146, p < 0.001), Working Memory (β = 0.204, CrI: 0.2952 to 0.1134, p < 0.001), Planning/Organizing (β = −0.223, CrI: −0.1021 to −0.3436, p = 0.004), and Task Completion (β = −0.152, CrI: 0.075 to 0.228, p = 0.002) predicted the trajectory of improvement in PARS score over the 8-week trial. For youth who received placebo, only the Inhibit score was significantly, but weakly, associated with response trajectory (β = −0.081, CrI: −0.0167 to −0.1461, p = 0.015). For adolescents who had clinically significant impairment in Emotional Control, Working Memory, Planning/Organizing, and Task Completion (i.e., T-score >65), the trajectory of improvement significantly differed from patients without scores in the clinically significant range.

Conclusions:

Taken together, these findings point to the potential value of assessing executive function in youth with anxiety disorders as one strategy for guiding treatment selection. These data suggest that executive function may predict treatment response to psychopharmacologic treatment and point to numerous avenues for further personalizing treatment.

Introduction

Generalized anxiety disorder (GAD) is one of the most common anxiety disorders in adolescents (Beesdo et al. 2009; Merikangas et al. 2010) and is characterized by uncontrollable, diffuse anxiety and numerous somatic (Crawley et al. 2014) and cognitive symptoms, as well as related functional impairment (Ramsawh et al. 2010). The cognitive symptoms of GAD include executive functioning deficits such as working memory, monitoring, attention bias, planning, and task-switching difficulties (Moran 2016; Sylvester et al. 2016; Castagna et al. 2019; Tobias and Ito 2021). These broad-ranging impairments are likely linked to educational underachievement (Woodward and Fergusson 2001) and strained peer and family relationships common in anxious youth (Tonge et al. 2020).

Impaired executive functioning (e.g., top–down attention, task switching, and error monitoring) in youths with anxiety disorders is related to differences in functional activity within prefrontal cortex (PFC) regions such as the dorsolateral PFC (Fitzgerald et al. 2013), frontal pole, anterior cingulate, and insula (Shanmugan et al. 2016) as well as the organization and connectivity of the ventral attention, default mode (Perino et al. 2021), and cingulo-opercular networks (Shanmugan et al. 2016). In adolescents with GAD, compared to healthy adolescents, the ventrolateral PFC activity is hyperactive and aberrantly connected with other prefrontal regions during attentional tasks with emotional and neutral distractors (Strawn et al. 2012a).

In a heterogeneous group of youths with obsessive compulsive disorder or GAD, the dorsolateral PFC activity decreased during error processing compared to healthy youths (Fitzgerald et al. 2013). Finally, in children with GAD completing a dot-probe task, greater clinician-rated anxiety was associated with greater capture of attention by nonemotionally valanced stimuli. Moreover, children with more difficulty shifting attention exhibited aberrant functional connectivity in the inferior frontal gyrus specific to the ventral attention network along with salience, default mode, and frontoparietal and cingulo-opercular regions (Perino et al. 2021).

Despite recognizing neurofunctional and neurostructural processes in pediatric anxiety disorders that relate to altered executive function (Sylvester et al. 2012; Shanmugan et al. 2016), systematic characterization of executive function in anxious youth (Murphy et al. 2018) and its relationship to treatment have rarely been examined. In fact, we are unaware of any study that has examined executive functioning in anxious youth as a predictor of either psychotherapeutic or psychopharmacologic treatment. Characterizing executive function in pediatric anxiety disorders and understanding how it relates to clinical and demographic characteristics as well as treatment outcomes are important as it offers a strategy for refining diagnostic approaches and tailoring interventions.

This is particularly important given the high comorbidity of anxiety disorders with other disorders that are characterized by impaired executive dysfunction (e.g., mood disorders (Melton et al. 2016) and attention-deficit/hyperactivity disorder (ADHD) (Gümüş et al. 2015), and learning disorders (Nelson and Harwood 2011). With this in mind, we used the self-report version of the Behavior Rating Inventory of Executive Function (BRIEF-SR) (Baron 2000) to examine executive function in unmedicated youth with GAD and then examined the relationship between baseline executive functioning domains and treatment outcome in adolescents with a primary diagnosis of GAD, who were randomized (1:1) to double-blind treatment with escitalopram or placebo.

Methods

Using data from a federally funded double-blind, placebo-controlled trial of escitalopram in adolescents 12–17 years of age with GAD (Strawn et al. 2020), we examined baseline measures of executive function and their relationship to treatment response. Study details have been described elsewhere (Strawn et al. 2020; Lu et al. 2021). Briefly, this study was approved by the Institutional Review Board (ClinicalTrials.gov identifier: NCT02818751) and conducted at a single U.S. academic site from February 2015 to November 2018. Informed consent and assent were obtained from all legal guardians and patients, respectively.

Patients were recruited from the Greater Cincinnati/Northern Kentucky region (USA), had moderate to severe GAD at baseline, as reflected by a Pediatric Anxiety Rating Scale (PARS) score ≥15. In addition, inclusion and exclusion criteria have been described in detail elsewhere (Strawn et al. 2020). Briefly, however, patients were excluded by the diagnosis of major depressive disorder, bipolar I disorder, posttraumatic stress disorder, or a psychotic spectrum disorder. Patients were randomized by investigational pharmacists 1:1 to escitalopram or placebo with stratification by sex using random number generators. Patients, caregivers, and investigational staff were all blinded to treatment assignment.

Escitalopram was initiated at 5 mg daily and titrated to 15 mg/day by the end of the 2nd week of treatment with the potential to increase to 20 mg at the 4- or 6-week visit. Then, anxiety and depressive symptoms, using the PARS (RUPP 2002) and Children's Depression Rating Scale (Poznanski and Mokros 1996), respectively, were assessed at baseline and weeks 1, 2, 4, 6, and 8, and/or early termination.

Executive function was assessed at baseline using the self-report version of the BRIEF (BRIEF-SR), an 80-item instrument that assesses children and adolescents' perceptions of their executive functions, or self-regulation (Baron 2000). The BRIEF is a reliable and valid measure of everyday executive function (Shear et al. 2002; de Water et al. 2019). Each item assesses specific behaviors, and the patient rates how frequently she experiences those behaviors on a three-point Likert scale: “Never,” “Sometimes,” or “Often.” The items of the BRIEF are categorized into eight clinical scales: Inhibit, Shift, Emotional Control, Initiate, Working Memory, Plan/Organize, Organization of Materials, and Monitor.

The Inhibit scale evaluates inhibitory control and impulsivity (i.e., the ability to resist impulses and stop behavior when appropriate) and includes behaviors such as interrupting, saying inappropriate things, and being restless and intrusive. The Shift scale reflects how one moves from situation-to-situation or activity-to-activity depending on the circumstances and reflects the ability to make transitions, tolerating change, flexibility in problem solving, and alternating attention. The Shift scale can produce subscores related to behavioral shift and cognitive shift (Baron 2000). The Behavioral Shift subscale includes aspects of shifting that pertain to the executive of shifts (e.g., “making transitions and tolerating change”), whereas problem solving flexibly and other cognitive aspects are reflected by the Cognitive Shift subscale (Baron 2000) and assesses an individual's ability to modulate or control his or her emotional responses.

The Emotional Control subscale reflects the “impact of executive function problems on emotional expression” (Baron 2000). The Initiate score represents the ability to initiate tasks and independently generate plans and actions. The Working Memory scale measures “on-line representational memory” (Baron 2000) (i.e., holding information in the service of encoding data or completing a task). How an individual manages “current and future-oriented task demands” (Baron 2000) is reflected by the Plan/Organize Scale and combines a planning component (anticipating future events and goal setting, as well as sequencing) and a developing component. The Organization of Materials scale “measures orderliness of work and storage spaces (e.g., desks, lockers, and backpacks)” (Baron 2000).

The Monitor scale reflects “self-monitoring or interpersonal awareness” (Baron 2000) and describes how the patient sees himself or herself with regard to an awareness that her behavior impacts others. Last, Task Completion represents the outcome of other executive difficulties (e.g., working memory, planning, organization, and inhibitory control) and reflects the “ability to finish or complete tasks appropriately and/or in a timely manner, emphasizing difficulties with the production of work or performance output” (Baron 2000).

Statistical methods

For each BRIEF-SR score, T-scores were determined relative to age- and sex-normed samples using the testing manual (Baron 2000). All analyses were completed on an intent-to-treat basis and analyses of continuous outcome measures included randomized patients with both a baseline and at least one postbaseline value for the variable being analyzed (e.g., PARS score). Imputation occurred by last observation carried forward as previously described. Density estimates for each T-score mean value were computed by Monte Carlo posterior simulation from a Student-t distribution (assuming an uninformative uniform prior density) and posterior p-values for probability of rejection of the null hypothesis (mean T-score = 50) were evaluated from these posterior densities.

For the continuous outcome measure (PARS score), Bayesian hierarchical models (BHMs) with logarithmic trend individual effects were employed to determine predicted mean outcome values at weeks 0, 1, 2, 4, 6, and 8 (or early termination). BHMs are multilevel models that estimate an individual outcome trajectory for each patient similar to the random effects in a mixed model for repeated measures. Linear models with a single logarithmic trend (i.e., without individual trajectories) were also estimated to evaluate robustness of the results to variation in the model specification. Each model was created with a limited number of covariates (e.g., age, sex) and estimated using the No U-Turn Hamiltonian Monte Carlo sampler as previously described (Strawn et al. 2017; Suresh et al. 2020).

As a further check for robustness, model selection criteria (Akaike information criterion and Bayesian information criterion) were employed to estimate more parsimonious models and to evaluate the effects of inclusion of irrelevant variables (Mills and Prasad 1992). For each score that predicted treatment response, we compared the trajectory of improvement (e.g., PARS score) between patients with and without clinically significant scores (i.e., T-scores ≥65). All analyses were performed in Julia (version 1.6) (Bezanson et al. 2014) and findings were considered statistically significant at the 5% threshold.

Results

Patient characteristics

The mean age of the patients was 14.8 ± 1.6 (standard deviation) years (interquartile range [IQR]: 14 to 16) and baseline PARS score was 17.8 ± 2.3 (IQR: 16 to 19), and 76% of the patients were girls. Of the 51 patients randomized to placebo or escitalopram, BRIEF data were not obtained for one participant who was randomized to escitalopram. Thus, BRIEF data were available, at baseline, for 25 patients who received placebo and 25 patients who received escitalopram. Additional characteristics of the sample have been described previously (Strawn et al. 2020; Lu et al. 2021, 2022).

Executive function scores in adolescents with GAD

For all subscores, at baseline, T-scores were significantly elevated in adolescents with GAD (N = 50) compared to the normative healthy sample whose mean, by definition, is T-score = 50 (Fig. 1). For the Inhibit scale, the estimated mean T-score was 56.3 (95% credibility interval [CrI]: 52.4 to 60.3, p = 0.002), and for the Shift Score, the estimated mean T-score was 64.5 (95% CrI: 60.5 to 68.5, p < 0.001). The estimated mean T-score for Emotional Control was 67.8 (95% CrI: 64.6 to 71.0, p < 0.001) and the estimated mean T-score for Monitor scale was 56.8 (95% CrI: 53.0 to 60.5, p < 0.001). Regarding Working Memory, the estimated mean T-score was 63.2 (95% CrI: 59.2 to 67.1, p < 0.001).

FIG. 1.

Baseline executive function of adolescents with generalized anxiety disorder. Histograms show frequency of T-scores in anxious youth for each subdomain of executive dysfunction compared to compared to age- and sex-matched healthy youths (i.e., T-score 50): (A) inhibit, (B) shift, (C) emotional control, (D) monitor, (E) working memory, (F) planning/organizing, (G) organizing materials, (H) task completion, (I) behavioral shift, and (J) cognitive shift.

For Planning/Organization, the estimated mean T-score was 59.0 (95% CrI: 55.6 to 62.5, p < 0.001), while for Organizing Materials, the estimated mean T-score was 56.2 (95% CrI: 52.4 to 60.0, p = 0.002). For Task Completion, the estimated mean T-score was 63.1 (95% CrI: 59.8 to 66.4, p < 0.001). The estimated mean T-score for Behavioral Shift and Cognitive Shift was 64.7 (95% CrI: 60.5 to 68.8, p < 0.001) and 60.4 (95% CrI: 56.9 to 63.8, p < 0.001) (Fig. 1).

Relationship between baseline executive function domains and treatment outcome

In patients receiving escitalopram (n = 25), baseline scores for Emotional Control (β = 0.256, CrI: 0.367 to 0.146, p < 0.001), Working Memory (β = 0.204, CrI: 0.2952 to 0.1134, p < 0.001), Planning/Organizing (β = −0.223, CrI: −0.1021 to −0.3436, p = 0.004), and Task Completion (β = −0.152, CrI: 0.075 to 0.228, p = 0.002) predicted the trajectory of improvement in PARS score over the 8-week trial (Table 1; Fig. 2). For youth who received placebo (n = 25), only one variable score was statistically significantly associated with response trajectory: Inhibit score (β = −0.081, CrI: −0.0167 to −0.1461, p = 0.015) (Table 2; Fig. 2). Results from single logarithmic trend and individual trend models were remarkably similar (up to at least two significant digits), as were results from more parsimonious specifications obtained by application of model selection criteria.

FIG. 2.

Baseline executive function and trajectory of improvement in adolescents receiving escitalopram (A) and placebo (B). Models are adjusted for age and sex. CrI, credibility interval.

Table 1.

Baseline Executive Function Domains and Change in Pediatric Anxiety Rating Scale in Adolescents Receiving Escitalopram

Score	β	SE	p	CrI
Cognitive shifting	−0.457	0.255	0.0757	0.0434 to 0.9564
Behavioral shifting	−0.164	0.280	0.5584	0.384 to 0.7121
Emotional control	0.256	0.056	<0.001	0.3667 to 0.1458
Inhibit	−0.002	0.036	0.9591	0.0696 to 0.0733
Shift	0.419	0.461	0.3649	1.3219 to 0.4842
Monitor	−0.004	0.037	0.9078	0.0681 to 0.0767
Working memory	0.204	0.046	<0.001	0.2952 to 0.1134
Planning/organizing	−0.223	0.062	0.0004	−0.1021 to 0.3436
Task completion	−0.152	0.039	0.0002	−0.0753 to 0.2284
Organizing materials	−0.015	0.027	0.5779	0.0381 to 0.0684

The model includes age and sex as covariates. Italics indicate findings that are significant at the 0.005 threshold. R ² = 0.596.

CrI, credibility interval; SE, standard error.

Table 2.

Baseline Executive Function Domains and Change in Pediatric Anxiety Rating Scale in Adolescents Receiving Placebo

Variable	β	SE	p	CrI
Cognitive shift	−0.348	0.201	0.0861	0.0466 to 0.7423
Behavioral shift	−0.145	0.219	0.5094	0.2838 to 0.5728
Emotional control	−0.001	0.029	0.9663	0.0547 to 0.0572
Inhibit	−0.081	0.033	0.015	−0.0167 to 0.1461
Shift	0.448	0.362	0.2181	1.1581 to 0.2618
Monitor	−0.01	0.036	0.7825	0.0604 to 0.0803
Working memory	0.041	0.033	0.2214	0.1056 to 0.0242
Planning/organizing	0.066	0.042	0.1188	0.1474 to 0.0163
Task completion	−0.043	0.03	0.1445	0.0146 to 0.1016
Organizing materials	−0.041	0.027	0.136	0.0126 to 0.0948

R ² = 0.336.

Clinically significant alterations in executive function domains and treatment outcome

As previously described, response varied considerably across patients (Fig. 3). For adolescents who had clinically significant impairment in Emotional Control, Working Memory, Planning/Organizing, and Task Completion (i.e., score ≥65), the trajectory of improvement significantly differed from patients without scores in the clinically significant range (Fig. 4). Having a T-score in the clinically significant range for Emotional Control (p < 0.001), Working Memory (p < 0.001), Planning/Organizing (p = 0.004), and Task Completion (p = 0.002) predicted the trajectory of improvement in PARS score over the 8-week trial (Fig. 4).

FIG. 3.

Heterogeneity of trajectory of improvement in escitalopram-treated adolescents. The thick dotted (blue) line represents the average trajectory of improvement in the entire sample of escitalopram-treated adolescents (n = 25). PARS, Pediatric Anxiety Rating Scale.

FIG. 4.

Differences in trajectory of improvement in escitalopram-treated adolescents (n = 25) based on executive function domains that were significant in the multivariate model of response. Working memory, task completion, emotional control and planning/organizing are shown in (A–D) respectively. Models are adjusted for age and sex. Dotted gray lines represent the average improvement in the entire sample of escitalopram-treated adolescents.

Patients treated with escitalopram, with significantly impaired baseline Planning/Organization and Task Completion, experienced greater symptom improvement (as measured by greater reduction in PARS score). Patients with significantly impaired baseline Emotional Control and Working Memory experienced less symptom improvement (as measured by less reduction in PARS score), following with escitalopram. The multidimensional nature of the models (e.g., including multiple covariates) precludes graphical representation of individual patient trajectories without either combining all effects or ignoring additional moderators.

Discussion

Multidimensional executive function deficits among adolescents with GAD may undermine cognitive capacity and impair educational and interpersonal functioning (Woodward and Fergusson 2001; Nelson and Harwood 2011; Tonge et al. 2020). In turn, these experiences may exacerbate anxiety and, over time, contribute to demoralization and secondary depressive symptoms (Dobson et al. 2021; Walkup et al. 2021). Indeed, the complex relationship between affective and anxiety disorders, as well as among anxiety disorders and substance use disorders may result from common disruptions in cognitive processes (Ressler 2020). As such, identifying underlying domains of cognitive impairment may facilitate targeted approaches to treatment of both primary and secondary psychopathology and potentially decrease treatment resistance and accumulated morbidity (e.g., attenuate the risk of substance use in an adolescent with primary anxiety).

Our findings suggest that distinct profiles of executive function may guide experimental medicine approaches to identifying and targeting anxiety-related causal circuits. These findings, similarly, compel examining the neurophysiologic substrates of executive function subdomains. Executive function could be conceptualized as a “fingerprint” of the activation and connectivity in its associated regions and networks. Activity in regions implicated in executive function predicts treatment success and the activity in these regions may change with effective treatment. Yet, most of these regions have yet to be explicitly linked to measures of executive dysfunction (Baumel et al. 2022).

The aberrant neurophysiology in anxiety disorders includes the medial PFC, dorsolateral PFC, and anterior cingulate cortex, as well as multiple networks, including the ventral attention, cingulo-opercular, and fronto-parietal networks (Sylvester et al. 2012; Kujawa et al. 2016; Burkhouse et al. 2017; Gilbert et al. 2020; Perino et al. 2021; Strawn et al. 2012b, 2016, 2020).

As we increasingly appreciate how pharmacotherapies and psychotherapies differentially modify neurofunctional activity in regions implicated in specific aspects of executive function (Maslowsky et al. 2010; Phan et al. 2013; Andreescu et al. 2015; Burkhouse et al. 2018; Lu et al. 2021, 2022; Baumel et al. 2022), we may leverage executive function “fingerprints” in neuroimaging or “outputs” in neuropsychiatric measures to match specific treatments with specific patients (Stancil et al. 2022). For example, an anxious adolescent with more difficulty in planning and organization might benefit more from escitalopram versus cognitive behavioral therapy (CBT). Alternatively, a patient with set shifting/cognitive flexibility deficits derives more benefit from CBT compared to an anxious adolescent who is cognitively flexible, but has difficulty with task persistence and planning. Certainly, CBT focuses on skill building and behavioral rehearsal in ways that might directly affect cognitive flexibility; it may also offer strategies for emotional control, which in this study predicted worse response to pharmacological intervention. These findings raise interesting questions about how baseline cognitive functioning might be used to guide treatment selection. Although further studies stratifying on these variables are needed, this line of inquiry offers a novel approach to personalizing care for youth anxiety disorders.

Notably, baseline executive function in adolescents with GAD predicts response to escitalopram, but not placebo. This finding is intriguing, considering recent evidence that treatment interventions may alter or depend on certain aspects of executive function. For example, in a double-blind treatment study of youths with ADHD, BRIEF scores, at baseline, predicted improvement following trigeminal nerve stimulation (Loo et al. 2021). Moreover, in an open-label comparison of atomoxetine versus methylphenidate for the treatment of children with ADHD both medications improved executive function; however, methylphenidate-treated patients experienced greater improvement in the selection/inhibition domain of executive function (Wu et al. 2021). In addition, SSRIs acutely alter executive function (e.g., reaction times) in anxious adolescents (Lu et al. 2022).

Furthermore, impaired executive functioning may attenuate CBT response in anxious children and adolescents. In this regard, youth with ADHD in the Child and Adolescent/Anxiety Multimodal Study fared worse when receiving CBT compared to patients without ADHD (Halldorsdottir et al. 2015). Taken together, this raises the possibility that more nuanced approaches to treating pediatric anxiety disorders warrant additional study. For example, CBT may require a certain baseline cognitive wherewithal for some patients or based on baseline executive function, CBT approaches might be selectively focused on cognitive or behavioral (e.g., exposure) aspects rather than an equal proportion of both elements.

Our data suggest that the best improvement with escitalopram occurs in those youth with less clinically significant impairment in working memory and emotional control. However, it is noteworthy that those with clinically significant range for planning and organizing and task completion had better responses over time. The heterogeneity of executive function and the way in which executive functioning relates to myriad processes may help to explain what may initially seem like a discordant finding.

Impaired working memory and emotional control, which may represent core features, may globally predict worse outcomes regardless of treatment or, ostensibly, could represent specific deficits that perpetuate nonpharmacologically responsive anxiety symptoms. This would raise the possibility that those with significant dysfunction in these domains may require specific nonpharmacologic or alternative interventions (e.g., executive function coaching, cognitive remediation).

By contrast, youth with difficulty in planning/task completion could reflect those with more cognitive flexibility, who may be more capable to tolerating “messiness,” uncertainty, or ambiguity. How these cognitive features relate to one another and to specific interventions could help clinicians identify which patients would benefit the most from which treatment.

The relationship between executive functioning and treatment response could relate to comorbidity, in the broadest sense. Specific comorbid anxiety symptoms could produce specific executive functioning deficits that drive the results observed herein. For example, the hypervigilance related to social evaluation in social anxiety disorder could affect multiple executive functions, including emotional control or working memory. Similarly, separation anxiety disorder, or subsyndromal social anxiety symptoms may be associated with closer relationships with primary attachment figures that culminate in greater parental involvement in the adolescent's activities and consequently affect several executive functions that were related to outcomes—task completion and planning/organizing.

Furthermore, while our study focused on adolescents with relatively “pure” anxiety disorders, who lacked significant depressive symptom comorbidity, depressive symptoms or subsyndromal depressive disorders could contribute to executive function deficits that were related to treatment response, including working memory, task completion, and planning/organizing. However, baseline depressive symptoms were low, which decreased our ability to explore links between depressive symptoms in anxious youth and executive function. Importantly, understanding how these depressive symptoms in anxious youth impact executive dysfunction could (1) clarify risk factors for the progression from anxiety to depressive disorders and (2) elucidate shared executive dysfunction across these conditions.

While this is the first examination of executive functioning in youth with GAD and the only study to examine executive functioning as a predictor of SSRI treatment response, there are several limitations. First, we used a self-report measure and, as such, a patient's insight may affect the validity of the scores; however, the BRIEF has high reliability for the parent and self-report versions (Baron 2000). Second, the sample size was small and did not provide the power to explore the neurobiological correlates of baseline executive function domains and improvement in anxiety. Third, the sample was homogeneous, which limited our ability to explore many patient-specific characteristics with respect to outcomes.

This study had a high proportion of females (76%), which limits our power to probe gender differences that may be present in the executive function architecture of youth with GAD and differential impacts of executive function on treatment response across genders. Fourth, this study examined adolescents with GAD (mean age 14.8 years); however, executive function in patients with GAD may differ across the lifespan (e.g., childhood vs. adolescents vs. adult vs. elderly patients) or with cumulation of years with GAD. Fifth, whether the impairment in working memory and emotional control are part of GAD processes or whether these impairments are related to “comorbidity” in the broadest sense of the term remains to be determined. However, if the former were true, higher symptom severity—reflected by PARS score—might be expected.

Interestingly, our examination of depressed and anxious youth participating in randomized controlled treatment trials suggests that greater symptom severity is related to outcomes (Strawn et al. 2022). Last, given the small sample size, we were unable to examine how these executive function domains relate to executive function networks or the extent to which examining executive function could probe the underlying circuits and be used to tailor psychotherapeutic or psychopharmacologic treatment.

Conclusion

This study presents a rare characterization of executive function in adolescents with GAD and examines how baseline executive function predicts treatment response in a double-blind placebo controlled trial. Executive function is disrupted across multiple domains in anxious youth, and disruptions in particular subdomains predict improvement in escitalopram-treated patients, but not in those who receive placebo.

Clinical Significance

Executive function deficits may—through cognitive mechanisms—contribute to distress and functional impairment. These findings speak to the importance of assessing cognitive functioning alongside other social/contextual variables. In addition, they provide evidence that executive function may predict treatment response.

Footnotes

Disclosures

J.R.S. has received research support from the National Institutes of Health (NIMH/NIEHS/NICHD) as well as AbbVie, Neuronetics, and Otsuka. He has received material support from and provided consultation to Myriad Genetics. He also is a consultant to the U.S. Food and Drug Administration and receives royalties from the publication of two texts (Springer). J.R.S. serves as an author for UpToDate and an Associate Editor for Current Psychiatry. J.R.S. also received research support from the Yung Family Foundation, the National Institute of Environmental Health and Safety, and the National Institute of Mental Health. J.A.M. has received research support from the Yung Family Foundation and the University of Cincinnati. T.S.P. and J.R.S. has received research support from the NIMH and PCORI. The other authors report no biomedical conflicts of interest.

References

Andreescu

, Sheu

, Tudorascu

, Gross

, Walker

, Banihashemi

, Aizenstein

: Emotion reactivity and regulation in late-life generalized anxiety disorder: Functional connectivity at baseline and post-treatment. Am J Geriatr Psychiatry, 23:200–214, 2015.

Baron

: Behavior rating inventory of executive function. Child Neuropsychol J Norm Abnorm Dev Child Adolesc, 6:235–238, 2000.

Baumel

, Lu

, Huang

, Drysdale

, Sweeny

, Gong

, Sylvester

, Strawn

: Neurocurcity of treatment in anxierty disorders. Biomark Neuropsychiatry, 2022 (in press).

Beesdo

, Knappe

, Pine

: Anxiety and anxiety disorders in children and adolescents: Developmental issues and implications for DSM-V. Psychiatr Clin North Am, 32:483–524, 2009.

Bezanson

, Edelman

, Karpinski

, Shah

: Julia: A fresh approach to numerical computing. SIAM Review, 59:65–98, 2014.

Burkhouse

, Kujawa

, Hosseini

, Klumpp

, Fitzgerald

, Langenecker

, Monk

, Phan

: Anterior cingulate activation to implicit threat before and after treatment for pediatric anxiety disorders. Prog Neuropsychopharmacol Biol Psychiatry, 84:250–256, 2018.

Burkhouse

, Kujawa

, Klumpp

, Fitzgerald

, Monk

, Phan

: Neural correlates of explicit and implicit emotion processing in relation to treatment response in pediatric anxiety. J Child Psychol Psychiatry Allied Discip, 58:546–554, 2017.

Castagna

, Calamia

, Roye

, Greening

, Davis

: The effects of childhood inattention and anxiety on executive functioning: Inhibition, updating, and shifting. Atten Defic Hyperact Disord, 11:423–432, 2019.

Crawley

, Caporino

, Birmaher

, Ginsburg

, Piacentini

, Albano

, Sherrill

, Sakolsky

, Compton

, Rynn

, McCracken

, Gosch

, Keeton

, March

, Walkup

, Kendall

: Somatic complaints in anxious youth. Child Psychiatry Hum Dev, 45:398–407, 2014.

10.

de Water

, Curtin

, Zilverstand

, Sjödin

, Bonilla

, Herbstman

, Ramirez

, Margolis

, Bansal

, Whyatt

, Peterson

, Factor-Litvak

, Horton

: A preliminary study on prenatal polybrominated diphenyl ether serum concentrations and intrinsic functional network organization and executive functioning in childhood. J Child Psychol Psychiatry Allied Discip, 60:1010–1020, 2019.

11.

Dobson

, Croarkin

, Schroeder

, Varney

, Mossman

, Cecil

, Strawn

: Bridging anxiety and depression: A network approach in anxious adolescents. J Affect Disord, 280:305–314, 2021.

12.

Fitzgerald

, Liu

, Stern

, Welsh

, Hanna

, Monk

, Phan

, Taylor

: Reduced error-related activation of dorsolateral prefrontal cortex across pediatric anxiety disorders. J Am Acad Child Adolesc Psychiatry, 52:1183.e1–1191.e1, 2013.

13.

Gilbert

, Perino

, Myers

, Sylvester

: Overcontrol and neural response to errors in pediatric anxiety disorders. J Anxiety Disord, 72:102224, 2020.

14.

Gümüş

, Çakin Memik

, Ağaoğlu

: Anxiety disorders comorbidity in children and adolescents with attention deficit hyperactivity disorder. Arch Neuropsychiatr, 52:24–27, 2015.

15.

Halldorsdottir

, Ollendick

, Ginsburg

, Sherrill

, Kendall

, Walkup

, Sakolsky

, Piacentini

: Treatment outcomes in anxious youth with and without comorbid ADHD in the CAMS treatment outcomes in anxious youth with and without comorbid ADHD in the CAMS. J Clin Child Adolesc Psychol, 44:985–991, 2015.

16.

Kujawa

, Swain

, Hanna

, Koschmann

, Simpson

, Connolly

, Fitzgerald

, Monk

, Phan

: Prefrontal reactivity to social signals of threat as a predictor of treatment response in anxious youth. Neuropsychopharmacology, 41:1983–1990, 2016.

17.

Loo

, Salgari

, Ellis

, Cowen

, Dillon

, McGough

: Trigeminal nerve stimulation for attention-deficit/hyperactivity disorder: Cognitive and electroencephalographic predictors of treatment response. J Am Acad Child Adolesc Psychiatry, 60:856.e1–864.e1, 2021.

18.

, Li

, Baumel

, Mills

, Cecil

, Schroeder

, Mossman

, Huang

, Gong

, Sweeney

, Strawn

: Acute neurofunctional effects of escitalopram during emotional processing in pediatric anxiety: A double-blind, placebo-controlled trial. Neuropsychopharmacology, 47:1081–1087, 2022.

19.

, Mills

, Li

, Schroeder

, Mossman

, Varney

, Cecil

, Huang

, Gong

, Ramsey

, DelBello

, Sweeney

, Strawn

: Acute neurofunctional effects of escitalopram in pediatric anxiety: A double-blind, placebo-controlled trial. J Am Acad Child Adolesc Psychiatry, 60:1309–1318, 2021.

20.

Maslowsky

, Mogg

, Bradley

, McClure-Tone

, Ernst

, Pine

, Monk

: A preliminary investigation of neural correlates of treatment in adolescents with generalized anxiety disorder. J Child Adolesc Psychopharmacol, 20:105–111, 2010.

21.

Melton

, Croarkin

, Strawn

, McClintock

: Comorbid anxiety and depressive symptoms in children and adolescents. J Psychiatr Pract, 22:84–98, 2016.

22.

Merikangas

, He

, Brody

, Fisher

, Bourdon

, Koretz

: Prevalence and treatment of mental disorders among US children in the 2001–2004 NHANES. Pediatrics, 125:75–81, 2010.

23.

Mills

, Prasad

: A comparison of model selection criteria. Econom Rev, 11:201–233, 1992.

24.

Moran

: Anxiety and working memory capacity: A meta-analysis and narrative review. Psychol Bull, 142:831–864, 2016.

25.

Murphy

, Luke

, Brennan

, Francazio

, Christopher

, Flessner

: An investigation of executive functioning in pediatric anxiety. Behav Modif, 42:885–913, 2018.

26.

Nelson

, Harwood

: Learning disabilities and anxiety: A meta-analysis. J Learn Disabil, 44:3–17, 2011.

27.

Perino

, Yu

, Myers

, Harper

, Baumel

, Petersen

, Barch

, Luby

, Sylvester

: Attention alterations in pediatric anxiety: Evidence from behavior and neuroimaging. Biol Psychiatry, 89:726–734, 2021.

28.

Phan

, Coccaro

, Angstadt

, Kreger

, Mayberg

, Liberzon

, Stein

: Corticolimbic brain reactivity to social signals of threat before and after sertraline treatment in generalized social Phobia K. Biol Psychiatry, 73:329–336, 2013.

29.

Poznanski

, Mokros

: Children's Depression Rating Scale–Revised (CDRS-R) Manual. Los Angeles (California), Western Psychological Services, 1996.

30.

Ramsawh

, Chavira

, Stein

: Burden of anxiety disorders in pediatric medical settings: Prevalence, phenomenology, and a research agenda. Arch Pediatr Adolesc Med, 164:965–972, 2010.

31.

Research Units on Pediatric Psychopharmacology (RUPP). The Pediatric Anxiety Rating Scale (PARS): Development and psychometric properties. J Am Acad Child Adolesc Psychiatry, 41:1061–1069, 2002.

32.

Ressler

: Translating across circuits and genetics toward progress in fear—And anxiety-related disorders. Am J Psychiatry, 177:214–222, 2020.

33.

Shanmugan

, Wolf

, Calkins

, Moore

, Ruparel

, Hopson

, Vandekar

, Roalf

, Elliott

, Jackson

, Gennatas

, Leibenluft

, Pine

, Shinohara

, Hakonarson

, Gur

, Satterthwaite

: Common and dissociable mechanisms of executive system dysfunction across psychiatricdisorders in youth. Am J Psychiatry, 173:517–526, 2016.

34.

Shear

, DelBello

, Lee Rosenberg

, Strakowski

: Parental reports of executive dysfunction in adolescents with bipolar disorder. Child Neuropsychol J Norm Abnorm Dev Child Adolesc, 8:285–295, 2002.

35.

Stancil

, Tumberger

, Strawn

. Target to treatment: a charge to develop biomarkers of response and tolerability in child and adolescent psychiatry. Clin Transl Sci, 15:816–823, 2022.

36.

Strawn

, Bitter

, Weber

, Chu

, Whitsel

, Adler

, Cerullo

, Eliassen

, Strakowski

, Delbello

: Neurocircuitry of generalized anxiety disorder in adolescents: A pilot functional neuroimaging and functional connectivity study. Depress Anxiety, 29:939–947, 2012a.

37.

Strawn

, Cotton

, Luberto

, Patino

, Stahl

, Weber

, Eliassen

, Sears

, Delbello

: Neural function before and after mindfulness-based cognitive therapy in anxious adolescents at risk for developing bipolar disorder. J Child Adolesc Psychopharmacol, 26:372–379, 2016.

38.

Strawn

, Mills

, Cornwall

, Mossman

, Varney

, Keeshin

, Croarkin

: Buspirone in children and adolescents with anxiety: A review and bayesian analysis of abandoned randomized controlled trials. J Child Adolesc Psychopharmacol, 28:2–9, 2017.

39.

Strawn

, Mills

, Schroeder

, Mossman

, Varney

, Ramsey

, Poweleit

, Desta

, Cecil

, DelBello

: Escitalopram in adolescents with generalized anxiety disorder: A double-blind, randomized, placebo-controlled study. J Clin Psychiatry, 81:20m13396, 2020.

40.

Strawn

, Mills

, Suresh

, Peris

, Walkup

, Croarkin

: Combining selective serotonin reuptake inhibitors and cognitive behavioral therapy in youth with depression and anxiety. J Affect Disord 298(Pt, A):292–300, 2022.

41.

Strawn

, Wehry

, Delbello

, Rynn

, Strakowski

: Establishing the neurobiologic basis of treatment in children and adolescents with generalized anxiety disorder. Depress Anxiety, 29:328–339, 2012b.

42.

Suresh

, Mills

, Croarkin

, Strawn

: What next?. A Bayesian hierarchical modeling re-examination of treatments for adolescents with selective serotonin reuptake inhibitor-resistant depression. Depress Anxiety, 37:926–934, 2020.

43.

Sylvester

, Corbetta

, Raichle

, Rodebaugh

, Schlaggar

, Sheline

, Zorumski

, Lenze

: Functional network dysfunction in anxiety and anxiety disorders. Trends Neurosci, 35:527–535, 2012.

44.

Sylvester

, Hudziak

, Gaffrey

, Barch

, Luby

: Stimulus-driven attention, threat bias, and sad bias in youth with a history of an anxiety disorder or depression. J Abnorm Child Psychol, 44:219–231, 2016.

45.

Tobias

, Ito

: Anxiety increases sensitivity to errors and negative feedback over time. Biol Psychol, 162:108092, 2021.

46.

Tonge

, Lim

, Piccirillo

, Fernandez

, Langer

, Rodebaugh

: Interpersonal problems in social anxiety disorder across different relational contexts. J Anxiety Disord, 75:102275, 2020.

47.

Walkup

, Friedland

, Peris

, Strawn

: Dysregulation, catastrophic reactions, and the anxiety disorders. Child Adolesc Psychiatr Clin N Am, 30:431–444, 2021.

48.

Woodward

, Fergusson

: Life course outcomes of young people with anxiety disorders in adolescence. J Am Acad Child Adolesc Psychiatry, 40:1086–1093, 2001.

49.

, Shang

, Lin

, Gau

SSF

: Differential treatment effects of methylphenidate and atomoxetine on executive functions in children with attention-deficit/hyperactivity disorder. J Child Adolesc Psychopharmacol, 31:187–196, 2021.