Interventions Based on Acceptance and Commitment Therapy for Stress Reduction in Children and Adolescents: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Abstract Introduction: Recently, acceptance- and commitment therapy (ACT) gained increasing interest. Studies show good efficacy in the treatment of patients presenting with several psychologic and somatic complaints. The present systematic review and meta-analysis addresses effectiveness of ACT-based interventions to reduce stress in children, adolescents, and young adults compared to control conditions. Methods: The meta-analysis was pre-registered at PROSPERO (CRD42019117440). Randomized controlled trials (RCTs) and quasi-randomized controlled trials (qRCT) in German or English language comparing the effects of ACT-based interventions to control conditions (e.g., treatment as usual, waitlist control) on stress-related outcome measures in youth were considered for inclusion. The target population was subjects 0–18 years of age. The databases PubMed, PsychInfo, Cochrane Database, CINAHL, and Web of Science were searched systematically up to July 2023. A random effect meta-analysis and a risk of bias assessment according to the procedure outlined in the Cochrane Handbook of Systematic Reviews were conducted. Results: The search resulted in 187 studies, of which eight studies with 976 participants were finally subjected to meta-analysis. Studies implemented ACT both in school-based group settings and in single settings and both as a universal and indicated prevention. Analyses yielded a significant main effect (Hedges' g = −0.20; 95% confidence interval [−0.36; −0.05]), indicating that interventions based on ACT resulted in greater reduction of stress compared to control conditions. Conclusion: ACT appears effective at reducing stress in youth. Further research is needed due to methodological shortcomings of existing studies. Small sample sizes, heterogenous studies, methodological shortcomings, and evidence of publication bias limit the conclusions that can be drawn from this meta-analysis.


Introduction
Studies show the severe impact of chronic stress on the development of mental disorders and various somatic illnesses [1,2].It can be assumed that in adolescence both the objective requirements of coping with every-day life and self-perception undergo significant changes.Emancipation from the parents, questions of identity, selfesteem, and first romantic relationships as well as increasing school demands and important decisions concerning the future are just a few examples of the challenges youth are facing during this sensitive phase of development.These high demands on the one side meet still poorly developed skills of coping with stress and emotion regulation [1,2] on the other side.
Studies show that a remarkable number of teenagers suffer from stress-associated symptoms related to school stressors but also stressors within the family, the peer group and in romantic relationships [2].An investigation in a sample of German high school students showed that adolescents are especially affected by stress related to the future and to school [3].The study further found an association between stress, behavioral and emotional problems.The strain caused by stress can lead to symptoms like those described in adults suffering from "burnout."Affected youth describe symptoms of exhaustion, a cynical attitude toward school and low confidence in their own abilities [4].Furthermore, stress among adolescents has been found to be associated with impairments of health and well-being, even depression [5] or anxiety [6], and has recently been shown to be related to substantially elevated healthcare costs [7].These severe consequences of stress in youth emphasize the importance of early and effective prevention (if possible).
Generally, one can differentiate three different types of prevention approaches (10): first, universal prevention addresses a broad population and has the aim to promote general knowledge or skills to maintain health.Second, selective prevention addresses a group of people with an elevated risk of developing a certain physical or mental condition.Third, indicated prevention targets patients who were individually identified as people with a very elevated risk or who already developed some symptoms.In adulthood, manifold approaches in the treatment and prevention of stress-related health impairments have been subject to research [8,9].In comparison, the state of research on stress management and prevention in adolescence seems rather scant.Findings from studies concerning stress in adults in occupational settings suggest that interventions based on cognitive behavioral therapy (CBT) show larger effect sizes than other approaches in reducing stress [10].Since the late 1980s, the cognitive behavioral therapies were extended by approaches of the third wave therapies.One of these approaches is Acceptance and Commitment Therapy (ACT; [11]).An important idea of the concept is that negative thoughts, feelings, and body perceptions are a part of human nature.ACT does not target a specific mental disorder and makes no direct attempt to reduce the symptoms of mental disorders.Instead, the focus is on accepting difficult thoughts and feelings while committing to one's chosen values.Symptom reduction happens simultaneously [12].This transdiagnostic approach makes ACT a possible base for interventions in the prevention and treatment of stressalso in youth (e.g., [13][14][15]).
In 2020, a meta-analysis focusing on the effects of ACT in the treatment of children and adolescents in general was published [14], indicating that ACT is a helpful approach for the treatment of different mental and physical conditions in children and adolescents but not superior to traditional CBT.The authors claim that further investigations are needed to address the differential indication of ACT-based treatments in youth.The authors of the previous meta-analysis last updated their literature search in August 2018, which resulted in three publications of interest less than this meta-analysis.Further, due to the very general research question, the main analysis of the previous publication investigated in the effects of ACT on all kinds of primary outcomes besides stress (e.g., anxiety, depression, wellbeing, etc.).In a second evaluation, only one secondary outcome per study was subject of the investigation.In only one study, the primary outcome was stress.In another study, the primary outcome was the integration of anxiety, depression, and stress (overall value instead of the subscale stress).None of the chosen secondary outcomes was stress.Consequently, the previous work did not allow any specific statement about the effects of ACT on stress.To our knowledge, there is no systematic review or meta-analysis until today, synthesizing existing data on the effectiveness of ACT as a preventive or therapeutic intervention for stress in children and adolescents compared to control conditions.The present article aims to fill this gap.
we made sure that there were no published or registered systematic reviews or meta-analyses addressing the same research question.The details of the protocol for this meta-analysis were preregistered on the "International prospective register of systematic reviews" (PROSPERO) and can be accessed online (CRD42019117440).

Inclusion and Exclusion Criteria
We aimed to identify trials using an intervention based on ACT principles to reduce stress in children and adolescents.To be included in the systematic review and meta-analysis, studies had to meet the following inclusion criteria: Randomized controlled trials (RCT) as well as quasi-randomized controlled trials (qRCT) were eligible for inclusion, whereas observational studies (noncontrolled) and non-randomized trials were excluded.The target population was subjects 0-18 years of age.The intervention delivered in the intervention group had to be based on the ACT principles in any mode of delivery.Treatment as usual (TAU), waitlist control (WLC), or any other type of treatment or intervention (i.e., befriending, CBT, etc.) were eligible as a control group.Only articles written in German or English language were considered eligible for inclusion.The primary outcome of the review was self-or other-reported (i.e., parent or teacher report) stress as assessed by validated questionnaires (i.e., perceived stress scale (PSS; [16]).There was no criterion regarding the length of posttreatment follow-up.There were no restrictions on the setting and type of prevention.

Literature Search
A systematic literature search according to the recommendations of the "Preferred Reporting Items for Systematic Reviews and Meta-Analyses" (PRISMA) 2020 statement [15] was carried out in July 2023.The databases PubMed (last searched July 24, 2023), PsychInfo (last searched July 24, 2023), Cochrane trials (last searched July 24, 2023), CINAHL (last searched July 25, 2023), and Web of Science (last searched July 25, 2023) were searched from 1900 to July 2023 with the following search terms: "acceptance commitment therap*" AND "stress" AND "youth OR adolesce* OR child* OR young people OR teen* OR young adults" AND "random*."The full search strategies for all databases are provided in the appendix.In case of studies including participants younger and older than 18 years, a data request for the data of interest was sent to the authors of the respective publication.If registered trials or published study protocols fulfilled inclusion criteria, a data request was sent to the respective research group.Eligibility of studies was assessed by two independent reviewers (F.B., R.F.) according to the inclusion and exclusion criteria defined above.Discrepancies in the assessment of the evaluators were discussed until consensus was reached.In case no consensus could be reached, a third independent reviewer (J.K.) assessed the study at question for inclusion or exclusion.

Data Extraction and Meta-Analysis
After inclusion, the following data were extracted for each study: authors, year of publication, language of publication, country in which the study took place, sample size (including mean age, sex ratio), sample randomized to intervention arms, complete sample flow (drop-out for all follow-up assessments, compliance with intervention), type of intervention/control, duration and dose of intervention/control, qualification of study personal delivering the intervention, setting of intervention (group, school-based, etc.), mean and standard deviation of any stress-related measure by time of assessment and group and the secondary outcomes assessed of the included studies (including data on any secondary outcome).A meta-analysis was performed using statistics software R (version 2023.06.2+561; packages: metaphor, metaviz, ggplot2).We expected considerable heterogeneity because the included studies differed methodologically (e.g., different stress-related outcomes were used).On that account, we used a random-effects model (RE-Model) to estimate the distribution of effect sizes using a restricted maximum likelihood estimator.The standardized mean difference (SMD) in stress related outcomes comparing treatment and control groups at postline was the primary measure of treatment effect.SMD and 95% confidence intervals (CIs) were calculated.The I 2 Index and Cochrane's Q were calculated to assess heterogeneity across effect sizes.A risk of bias assessment according to the procedure outlined in the Cochrane Handbook of Systematic Reviews was done independently by two assessors (FB, RF) using the Risk of Bias Tool 2 (RoB2; [17]) and visualized using the visualization tool robvis [18].Funnel plots were used to visually identify a possible risk of publication bias.

Literature Search and Excluded Studies
Our literature search returned a total of 329 possible matches.After removing duplicates, a total of 188 articles remained.In 135 cases, the abstracts provided the necessary information to exclude these articles.In four cases, registered trials or published study protocols fulfilled the eligibility criteria [19][20][21][22].Consequently, we sent requests to the authors, but no data could be provided (yet).One published study included adolescents but also young adults older than 18 years [23].We requested the data of the subsample which fulfilled the eligibility criteria, but the data could not be retrieved.The full texts of the remaining 47 articles were reviewed in detail.Within this step, 35 publications got excluded.In 25 cases only participants older than 18 years were included or the intervention targeted the parents instead of the children .Further seven studies were excluded because they did not assess stress [49][50][51][52][53][54][55].Although the authors of the multiple baseline-design study by Salazar et al. [55] collected data on stress, the assessment time points of the stress related outcome did not allow an integration to the meta-analysis analog to a WLC group design.Consequently, this trial was excluded.Three possible matches were excluded since the studies were no RCTs [56][57][58].In one case, data was not published in the necessary format for inclusion to the meta-analysis.We sent a request and retrieved the respective data from the authors [59].Figure 1 shows the in-and exclusion of the studies and reasons for exclusion.

Overview of Included Studies
Tables 1 and 2 show the study-and sample characteristics of the ten studies that met inclusion criteria.All studies taken together resulted in a total n of 1,367 participants.Participants' age ranged between 7 and 18 years.Sample sizes varied between n = 32 and n = 359 participants.The majority of studies [1,13,[60][61][62][63][64] took place in a school setting.Six studies were RCTs, whereas four studies were quasi-randomized trials.Five trials provided universal prevention, while the five other studies targeted mental or physical health impairments.Four studies addressed stress as the primary outcome, one trial aimed to reduce aggressive behavior [60] and one trial aimed to improve both self-efficacy and stress [65].The included studies are described in more detail below.
Theodore-Oklota et al. [60] primarily addressed the reduction of aggressive behavior among a sample 7th grade students.Therefore, they piloted a three-session school-based program based on the principles of ACT randomly assigning the classes to either the ACT-or a waitlist-condition.As a secondary outcome, the participants completed the problem-solving subscale of the Responses to Stress Questionnaire (RSQ; [67]) which surveys coping to a stressful situation.The results indicate that the participants in the ACT condition showed more problem-solving coping at the follow-up than the waitlist group.
Livheim et al. [13] investigated the effects of an indicated stress prevention compared to the regular health care of the school where the study took place.Eligible participants were screened for psychosocial problems prior to the training.Half of the adolescents aged 14-15 years received a manualized group program based on ACT which explicitly focused on stress reduction.The authors found a significantly higher improvement of the levels of stress in the intervention group with a large effect size.
Moazezzi et al. [65] examined a sample of children and adolescents diagnosed with Diabetes Mellitus.Half of the sample received therapy sessions oriented on the approach of ACT.The matched control group did not receive any kind of intervention.During intervention all participants continued their regular medical treatment.The intervention resulted in a significant reduction of perceived stress.In-and exclusion of the studies and reasons for exclusion (flow diagram from [15]).Burckhardt et al. [1] examined the effects of a schoolbased program which aimed to foster mental health based on elements of positive psychology and ACT.A sample of adolescents was randomly assigned to either the ACTbased program or their usual "pastoral care" classes.The authors were especially interested in the outcome regarding those students who suffered from heightened symptom scores in the beginning of the study.Within this group, they observed significant reductions in stress in the ACT condition compared to controls.
The same research group conducted another trial testing the feasibility of a study procedure for a universal prevention primarily aiming at the reduction of depression and anxiety [61].A school-based sample of teenagers was quasi-randomized in clusters to either a usual class activity called pastoral care or an ACT-based intervention.Results showed no significant difference between the groups apparently due to the small sample size.However, between-group effect sizes show a large effect in favor of the ACT-condition regarding stress.
Puolakanaho et al. [62] conducted a study investigating the effects of a web-and-mobile delivered intervention.The intervention aimed amongst others at the reduction of stress in adolescents attending 9th grade.The authors found a small but significant reduction of the overall stress level.The effects were largest among those participants especially affected by stress at the baseline.
Bernal-Manrique et al. [63] compared group-based ACT focused on repetitive negative thinking to a WLC group in a sample of adolescents with problems in social and school adaption.The training was effective in improving interpersonal skills as well as in the reduction of stress.
Kallesøe et al. [59] conducted a RCT with a sample of adolescents diagnosed with multiple functional syndromes.In addition to the regular treatment, half of the participants received ACT whereas the other half was randomized to enhanced usual care.Participants in the ACT condition showed a higher reduction of perceived stress than controls.
Knight [64] investigated in the efficacy of a brief universal prevention intervention in a school setting.While half of the participating classes attended their regular lessons, the other half received a brief intervention based on ACT.The intervention was effective in reduction of stress.This study (currently in preparation for publication) was the only included study not published in a peer-reviewed journal.
Guerrini Usubini et al. [73] published preliminary results from a RCT investigating in the effectiveness of ACT in addition to TAU in a of in-patient adolescents with obesity regarding several outcomes.Results indicated no significant effect of the intervention on stress.

Meta-Analysis
Prior to meta-analysis, we reversed the means to negative values in case of one study [60].Unlike the other included studies, coping was assessed instead of stress, i.e., higher levels indicating an improvement.Random effect meta-analysis showed that interventions based on ACT were superior to control conditions (Hedges' g = −0.75;95% CI [−1.44; −0.06]).We detected significant heterogeneity across reported effect sizes (Q = 258.22,df = 9, p = 0.000, I 2 = 95.55%).A second calculation after removal of two outliers (15, 64, see Fig. 2) resulted in a significant decrease of heterogeneity across effect sizes (Q = 5.53, df = 7, p = 0.596, I 2 = 4.31%).The adjusted meta-analysis included 976 instead of 1,367 participants and still resulted in a significant but smaller effect in favor of ACT (Hedges' g = −0.24;95% CI [−0.46; −0.02]) as illustrated in Figure 3.The funnel plot illustrated in Figure 2 shows a tendency of publication bias.

Risk of Bias and Quality Assessment
In six studies, there are some concerns of a risk of bias arising from the randomization process.Theodore-Oklota et al. [60] did not provide detailed information on the randomization procedure, reported relevant baseline differences between the conditions, and used a quasi-randomized design.Two more studies reported significant baseline differences between intervention and control group in relevant aspects [13,62].Further three studies quasi-randomized participants in classes or tutorial groups [1,61,64].
None of the included studies reported a blinding of participants (which we consider not possible in this type of intervention study) and only one study reported a blinding of assessors and study therapists regarding the aims of the study [73].Two studies monitored the correct implementation of the intervention using assessments of adherence [1,62].Consequently, in these two studies, the risk of bias due to deviations from the intended intervention was rated as at low.In the other cases, we rated the risk as unclear.
Three studies were rated at risk of bias due to missing outcome data.One study did not provide data from all participants and was consequently rated at high risk of bias [65].Furthermore, two studies [1,62] reported significant differences between completers and dropouts.In both cases, these differences include the levels of stress resulting in a high risk of bias regarding the outcome of interest.Apart from one exception [73], all studies raised some concerns regarding a possible bias in measurement of the outcome since they did not report a blinding of the data assessors.One study reported that they did not ensure blinding of outcome assessment [1], whereas the other studies provided no information concerning this issue at all.Five studies were rated as low regarding the risk of bias in selection of the reported results [13,59,62,64,73].The remaining five studies did not mention a prespecified plan of the study details (study protocol or trial registration).
Overall, in four cases, we rated a high risk of bias, in six cases some concerns about a risk of bias.No study had a low over all risk of bias. Figure 4 shows the summarized assessment of the studies regarding their risk of bias.

Summary of Evidence
In recent years, stress in childhood and adolescence has gained increasing research interest.ACT is a comparably new approach in the treatment and prevention of stress at a young age.There are only few studies that investigated the efficacy of ACT in this context.Moreover, many of them use ACT as a basis for parenting interventions in burdened families (for an overview see [74]).To our knowledge, this article is the first to provide a systematic review and meta-analysis addressing the efficacy of ACT compared to control conditions as an approach to target stress in children and adolescents.Following a systematic search of the literature, ten studies meeting our inclusion criteria were included in the metaanalysis, two of which [13,64] had to be excluded for being identified as outliers.Results of this meta-analysis yielded a significant effect in favor of ACT with a small effect size, indicating that in controlled conditions, ACT can be effective in the reduction of stress in this age group.
Methodological heterogeneity across studies was high mainly due to the different outcome measurements included in the studies.Four studies primarily addressed the reduction of stress [1,13,61,62] using either the DASS-21 [1,61,69], the PSS [13,16], or a single-item measurement [62] as the primary outcome.The other included studies assessed stress or the improvement of Fig. 2. Meta-analysis funnel plot.The studies far left outside the gray-shaded area refer to the studies [13,64] and have been excluded from analysis.skills to cope with stress as a secondary outcome using the PSS [16,59,64,65], DASS [63,69,73], or the RSQ [60,67].
Among the eight finally included studies, four initially reported significant advantages of the ACT-based condition compared to control conditions.Burckhardt et al. [1] and Moazezzi et al. [65] both found significant reductions of stress in the intervention group compared to the control group.Puolakanaho et al. [62] reported a small but significant effect in the reduction of overall stress between the intervention group and the control group.Bernal-Manrique et al. [63] targeted interpersonal skills in their trial resulting in a large effect.Further, the training was effective in the reduction of stress with a large effect size.
Theodore-Oklota et al. [60] primarily addressed the reduction of aggression in their trial, which did not result in a significant effect.However, they did find a significant impact on the participants' coping in stressful situations.
Three studies did not find significant main effects on the respective primary outcome nor on stress-related secondary outcomes.Kallesøe et al. [59] did not find a significant effect of ACT on the primary outcome physical health at the primary endpoint.However, they found a faster improvement in physical health in the intervention group.The trial did not yield a significant effect regarding stress measured via PSS.Guerrini-Usubini et al. [73] defined psychological wellbeing as the primary outcome of their RCT.The study did neither result in significant differences between conditions regarding the main outcome nor in stress.Burckhardt et al. [61] reported no significant main effect of the ACT-based training.Since the trial was a feasibility study with a small sample size, they deduced a tendency that the program might be successful in the reduction of stress from a comparison of the between-group effect sizes.
It is important to consider the methodological differences between the included studies.All but three studies [62,65,73] offered the intervention in a group format, mostly in class or other school-based tutorial groups.The intensity of the intervention varied to a considerable degree.Puolakanaho et al. [62] offered a 5-week mobile-and web-delivered program where the participants could decide individually how many exercises they wanted to finish.In addition, half of the intervention group received individual support by a coach.This trial did not yield a  [65] received ten therapy sessions in a single setting resulting in a significant effect and the largest effect size of all included trials.Burckhardt et al. [61] conducted a seven-week group-based training with a total duration of 4.6 h that resulted in a larger effect than the other trial by this research group [1] despite of a higher intensity of the training with 8 h.The shortest intervention of the included trials which consisted of three sessions that lasted 144 min in total [60] resulted in a significant effect regarding the stress-related measurement.
All studies reported data that were collected immediately after intervention, but only two studies assessed follow-up data 5 months after the baseline assessment [61] and 2 weeks, 3 months, and 7 months, respectively [59].Burckhardt et al. [61] found further improvement in the intervention group compared to controls in the follow-up assessment.Kallesøe et al. [59] only assessed stress at baseline and at the end of treatment.
The study that resulted in the largest effect of the finally included studies [65] differed from the other trials in at least one relevant aspect.With an average age of just under 12 years, the sample was the youngest.
One interesting study in the context of this review had to be excluded from analysis since the single case design did not allow an inclusion in the computation of the effect sizes in a meta-analysis.Salazar et al. [55] investigated in the effects of ACT in the treatment of nine children aged 8-13 years diagnosed with depression using a multiplebaseline design.The ACT-based intervention focused on the reduction of repetitive negative thinking and assessed stress as a secondary outcome.The authors report very large SMD effect sizes for stress measured via DASS posttreatment and at the 4-week-follow-up.
Compared to other approaches aiming at the reduction of stress-related symptoms in children and adolescents, the overall effect size of this meta-analysis appears rather small.A meta-analysis that integrated the results of different school-based approaches to reduce stress in adolescents (relaxation training, social problem solving, social adjustment, emotional self-control, or combined interventions) found a large overall effect in the reduction of stress-related symptoms [75].A meta-analysis on mindfulness-based stress reduction in children and adolescents found a small to moderate effect size in the reduction of stress [76].Kallapiran et al. [77] investigated the effectiveness of mindfulness addressing different health-related outcomes at a young age and provided additional evidence for the effectiveness of Mindfulnessbased-Stress-Reduction (Hedges' g = 0.31).Furthermore, they found a moderate to large effect in reducing stress using other mindfulness-based interventions.ACT also seems to yield smaller effects in the reduction of stress compared to its effectiveness regarding other outcome measures.Fang and Ding [14], who investigated ACTbased interventions targeting not only stress but several primary outcomes such as depression and anxiety, found a medium effect size over all studies comparing ACT to TAU or WLC.
Also in adults, the state of research regarding efficacy of ACT on stress in adults does not permit clear conclusions.Graham et al. [78] describe the state of research on ACT as a field that grows fast but lacks studies with a high quality.In existing reviews, the construct of stress is often defined very broadly.In a meta-analysis on the effects of ACT in adults with chronic conditions, most of the included studies showed improvements in distress (including anxiety and depressive symptoms) following ACT compared to control conditions [78].Han and Kim [79] investigated in the efficacy of internet-based ACT in adults on various outcomes, including stress.Like the present study, they conclude that more high-quality studies are necessary to make a firm statement.Overall, they found a small effect in favor of ACT regarding stress reduction.In subgroup analyses, ACT was superior to passive control groups, but there was no significant effect in the comparison to active control groups.ACT was not superior to control groups at follow-up.Han et al. [80] found small effects of ACT on stress in family caregivers compared to control groups.

Limitations
Many of the studies published so far in this field are pilot trials including only small samples [13,60,61].Due to methodological shortcomings, especially insufficient randomization and blinding of the outcome assessors, the results of this meta-analysis need to be interpreted with caution.Another important limitation is the heterogeneity of the included studies.Although the calculated heterogeneity was small enough to allow a summary by means of a meta-analysis, sub-analyses regarding variables such as single-versus group intervention, targeted versus universal intervention, etc. would be desirable.The small number of existing trials is not sufficient for these calculations.Furthermore, there is some evidence of publication bias, indicating that the existence of unpublished trials showing an opposite pattern of results cannot be completely ruled out despite of the screening of trial registers.Further, this meta-analysis cannot draw any definite conclusions concerning the long-term effects of ACT-based stress prevention programs in children and adolescents since only one trial provided follow-up data on stress.

Future Directions and Conclusion
Future studies should aim at a higher methodical rigor.Only one study showed no deficiencies regarding the randomization procedures [13].None of the studies provided an appropriate blinding procedure of participants, personnel, and assessors.Especially, the lack of blinding of assessors (who were in many cases the same persons who provided the intervention and who might have an implicit interest in the effectiveness of the program) causes a high risk of bias.Future studies should provide a procedure to guarantee the independence of the data collection from the implementation of the training.Some of the studies emphasize that especially those adolescents most affected by stress at the baseline measurement had the highest gains from the intervention.This finding is in line with the results of a meta-analysis schoolbased intervention which found that targeted interventions were more effective in the reduction of stress than universal programs [81].Puolakanaho et al. [62] conducted a moderation analysis which showed that those participants who reported more stress at the baseline had higher benefits in their stress levels due to the intervention.Livheim et al. [13] investigated in an indicated prevention approach that resulted in a large effect size in favor of the ACT-based training.However, this finding still needs to be replicated because the results may be biased due to the small sample.Still, an indicated stress prevention especially targeting the more severely stressed adolescents appears to be a promising approach.In conclusion, the present metaanalysis lends support that ACT is as an effective intervention to reduce stress in youth.

Fig. 1 .
Fig.1.In-and exclusion of the studies and reasons for exclusion (flow diagram from[15]).

Table 1 .
Characteristics of the included studies Systematic Review and Meta-Analysis: ACT for Stress Reduction in Youth

Table 2 .
Sample characteristics