Acceptance and Commitment Therapy Focused on Repetitive Negative Thinking for Child Depression: A Randomized Multiple‑Baseline Evaluation

[x] Context and Objectives

Childhood depression represents a significant mental health problem affecting 1-3% of the pediatric population, with substantial negative consequences for academic, social, and emotional functioning. Despite this prevalence, the availability of effective evidence-based treatments remains limited. The most widely studied interventions for childhood depression, including cognitive-behavioral therapy (CBT), demonstrate significant but moderate effect sizes overall.

From the perspective of Relational Frame Theory (RFT), repetitive negative thinking (RNT) has been conceptualized as a fundamental process involved in the persistence and exacerbation of depression. Acceptance and Commitment Therapy (ACT), a third-wave psychological intervention, directly addresses psychological suffering through acceptance of aversive thoughts and emotions rather than attempting to eliminate them. However, prior to this study, the efficacy of an RNT-focused ACT protocol had not been specifically evaluated in children with depression.

The primary purpose of this study was to evaluate the efficacy of a brief ACT protocol, specifically adapted for children and focused on reducing repetitive negative thinking, in reducing depressive symptoms and other emotional and behavioral problems. The authors hypothesized that an intervention consisting of only three sessions, grounded in RFT and ACT principles, could produce clinically significant changes in measures of depression, anxiety, stress, and behavioral problems in children with a primary diagnosis of childhood depression.

[x] Method

[x] Participants

The sample consisted of 9 children (4 girls) aged 8 to 13 years (M = 10.22, SD = 2.11). All participants met the primary diagnostic criterion of childhood depression according to the MINI KID (Mini International Neuropsychiatric Interview for Children and Adolescents), a structured clinical interview. Six of the nine participants (66.7%) presented with comorbid disorders: four had Oppositional Defiant Disorder (ODD), three had Attention-Deficit/Hyperactivity Disorder (ADHD), one met criteria for Generalized Anxiety Disorder (GAD), and one had Separation Anxiety Disorder.

An important inclusion criterion was that participants' Verbal Intelligence Quotient had to be at or above 70, as assessed by the Kaufman Brief Intelligence Test (K-BIT), with a sample mean of 100.78 (SD = 18.16), indicating that participants demonstrated normative-range cognitive abilities. Participants were recruited through social media in Bogotá, Colombia, during the data collection period.

[x] Design

A nonconcurrent, multiple-baseline design with three experimental arms (cohorts) was employed. This type of design, also termed a single-case experimental design (SCED), permits evaluation of the functional relationship between the intervention and changes in target behavior. The distinctive characteristic of the nonconcurrent design was that each cohort of participants received the intervention at different times:

Cohort 1 (3 participants): initiated intervention after 4 weeks of baseline
Cohort 2 (3 participants): initiated after 5 weeks of baseline
Cohort 3 (3 participants): initiated after 6 weeks of baseline

This structure allowed for stronger causal inferences, since any change synchronized with intervention initiation and not with external events strengthens the inference that the intervention was responsible for observed change. During baseline periods of different durations, participants completed weekly measurements of psychological change processes.

[x] Intervention

The intervention consisted of a brief protocol of three individual sessions, each lasting 40 minutes. This protocol was specifically adapted for pediatric populations based on Relational Frame Theory (RFT) and Acceptance and Commitment Therapy (ACT) principles. The complete protocol is available in the following open science repository: https://osf.io/ub2n8/.

Session 1 - Differentiation Between Psychological Inflexibility and Psychological Flexibility: Participants learned to identify and differentiate between problematic thoughts (PI) and functional thoughts (PF). Discrimination exercises were employed to strengthen this skill. The metaphor of the "wise king versus slave" was utilized to illustrate the difference between responding from a flexible, reflective perspective versus responding automatically and rigidly to thoughts and emotions.

Session 2 - Identification of Repetitive Negative Thinking Triggers and Cognitive Defusion: Concepts from the previous session were reviewed. Participants identified the specific triggers that activate their repetitive negative thinking. Cognitive defusion techniques were taught to reduce the functional impact of ruminating thoughts. Experiential exercises such as "Where's Waldo?" were used to practice observing thoughts without engaging with them, and a "daydreaming" exercise to differentiate between present-moment awareness and being lost in the mind.

Session 3 - Consolidation of Cognitive Defusion Skills and Valued Actions: Cognitive defusion skills learned in previous sessions were consolidated. Emphasis was placed on committing to actions consistent with the child's personal values, independent of the presence of negative thoughts or feelings, as a mechanism for building a meaningful and authentic life.

[x] Measurement Instruments

Process measures (collected weekly throughout baseline and intervention phases):

Acceptance and Action Questionnaire for Youth (AFQ-Y): Assesses the degree of psychological inflexibility, capturing difficulty accepting negative internal experiences. Higher scores indicate greater inflexibility.
Repetitive Negative Thinking Questionnaire for Children (PTQ-C): Evaluates specifically the frequency and intensity of repetitive negative thinking, the central psychological process targeted by the intervention.
Generalized Pliance Questionnaire for Children (GPQ-C): Assesses the degree to which children respond inflexibly to verbal rules and social pressure, a variable related to psychological inflexibility.

Outcome measures (collected at pretest, posttest, and 4-week follow-up):

Depression, Anxiety and Stress Scale for Children (DASS-C): Provides separate assessments of depressive symptoms, anxiety, and stress. This self-report measure was completed by pediatric participants.
Child Behavior Checklist (CBCL): Completed by parents/caregivers, evaluates internalizing problems (depression, anxiety), externalizing problems (disruptive behavior, aggression), and mixed problems (somatic problems, attention problems).
MINI KID Structured Diagnostic Interview: Administered at pretest, posttest, and 4-week follow-up to determine whether participants maintained diagnoses of depression and other comorbid conditions. This measure allowed evaluation of diagnostic remission.

[x] Statistical Analyses

Analyses employed multiple statistical strategies appropriate for single-case designs:

Baseline Slope Analysis: The Theil-Sen slope was used to evaluate trends present during baseline phases. A slope near zero indicates measurement stability and strengthens conclusions about changes due to intervention.
Bayesian Hypothesis Testing for Process Measures: The JZS+AR Bayesian hypothesis test was employed to evaluate evidence for treatment effects on AFQ-Y, PTQ-C, and GPQ-C. Bayes factors (BF) were calculated, where BF > 10 is considered "very strong" evidence that the treatment effect hypothesis is superior to the null hypothesis.
Effect Size Statistic for Single-Case Designs: The design-comparable d-statistic for SCED (d-statistic for single-case experimental design) was calculated for process measures, providing treatment effect magnitude estimates comparable to traditional group studies.
Bayesian Repeated Measures ANOVA: For outcome measures DASS-C and CBCL (measures obtained at multiple time points), Bayesian repeated measures ANOVA was used to evaluate evidence for treatment effects, with Bayes factors interpreted similarly.

[x] Results

[x] Within-Subject Effects on Process Measures

Results demonstrated highly consistent changes on weekly process measures:

Acceptance and Action Questionnaire for Youth (AFQ-Y): All 9 participants (100%) showed evidence of treatment effect with Bayes factors exceeding 10. This perfect 100% consistency across all participants is particularly notable and suggests the protocol was effective in reducing psychological inflexibility across the board.

Repetitive Negative Thinking Questionnaire (PTQ-C): Similarly, all 9 participants (100%) showed very strong evidence of a treatment effect, with BF > 10, indicating substantial reductions in repetitive negative thinking during the intervention.

Generalized Pliance Questionnaire (GPQ-C): Of the 7 participants with computable data, 4 (57.1%) showed evidence of treatment effect. This lower percentage suggests that generalized pliance was less responsive to the intervention than specific psychological inflexibility and repetitive negative thinking, though still a majority showed changes in the expected direction.

[x] Within-Subject Effect Sizes (d-statistics)

The design-comparable within-subject effect sizes were considerably large:

AFQ-Y: d = 3.74, confidence interval [2.43, 5.43]
PTQ-C: d = 3.14, confidence interval [1.88, 4.85]
GPQ-C: d = 1.14, confidence interval [0.01, 2.32]

Effect sizes for AFQ-Y and PTQ-C are classified as "very large" according to meta-analytic conventions, substantially exceeding those reported in meta-analyses of CBT for childhood depression (d ≈ 0.41).

[x] Diagnostic Outcomes

The most striking diagnostic outcome was that none of the 9 participants maintained the diagnosis of childhood depression at immediate posttest or at 4-week follow-up. This represents a 100% diagnostic remission rate. Regarding comorbid disorders:

Only one participant (P6) retained diagnoses of ADHD and ODD at posttest
At 4-week follow-up, no participant maintained comorbid diagnoses
This pattern suggests changes were not limited to depression but generalized to other conditions

[x] Emotional Symptoms (DASS-C)

Bayesian repeated measures analysis showed very strong evidence for a treatment effect on the total DASS-C scale and all subscales:

DASS-C Total: BF = 53,465.77 (very strong evidence)
Depression Subscale: BF = 172.95
Anxiety Subscale: BF = 67.42
Stress Subscale: BF = 4,163,000 (extraordinarily strong evidence)

Effect sizes from pretest to 4-week follow-up were:

DASS-C Total: d = 2.12, CI [0.89, 3.32]
Depression: d = 1.22
Anxiety: d = 1.18
Stress: d = 3.11

These very large effect sizes indicate substantial reductions in negative emotional symptoms.

[x] Parent-Reported Behavioral Problems (CBCL)

Bayesian analysis showed strong evidence for a treatment effect:

CBCL Total: BF = 21.50
Internalizing Scale: BF = 29.62
Externalizing Scale: BF = 10.04
Mixed Scale: BF = 22.22

Effect sizes from pretest to 4-week follow-up were:

CBCL Total: d = 1.21
Internalizing: d = 1.35
Externalizing: d = 1.06
Mixed: d = 1.22

These large effect sizes indicate that parents observed considerable reductions in behavioral problems, both internalizing and externalizing.

[x] Discussion and Conclusions

This study constitutes the first trial of an RNT-focused ACT protocol for childhood depression. The findings are notable for several reasons. First, the 100% diagnostic remission rate stands in sharp contrast to remission rates reported in CBT studies for childhood depression, which typically range from 40-60%. Second, the observed effect sizes, particularly for process measures (AFQ-Y d = 3.74, PTQ-C d = 3.14), are dramatically larger than those reported in CBT meta-analyses (d ≈ 0.41).

The authors argue that these results, while preliminary, represent a promising step in developing psychotherapeutic interventions for childhood depression. The brevity of the intervention (only three sessions) makes this approach particularly attractive from implementation and accessibility perspectives. However, the authors acknowledge numerous limitations that must be considered when interpreting these findings.

Recognized Limitations:

Design: Although the nonconcurrent multiple-baseline design provides greater experimental control than a single case, it lacked the highest level of experimental control (random assignment to control and true control conditions).
Self-Report Measures: Many process measures (AFQ-Y, PTQ-C, DASS-C) relied on self-report, vulnerable to biases such as social desirability and changes in perceived sensitivity.
Blinded Assessment: The MINI KID was not administered by an evaluator blind to intervention status, potentially allowing expectancy bias to influence diagnostic assessment.
Single Therapist: A single therapist administered all interventions, potentially confounding treatment effects with idiosyncratic characteristics of that therapist.
Small Sample Size: With only 9 participants, although appropriate for a single-case design, generalization is limited.
DASS-C Collection: The DASS-C was administered at pretest, posttest, and follow-up but not weekly like other process measures, providing fewer temporal data points for this important outcome measure.

[x] Significance and Contribution

This study contributes significantly to the literature on psychological interventions for childhood depression. It presents the first systematic evaluation of an extremely brief ACT protocol (3 sessions) specifically designed for children with depression, with particularly striking results: 100% diagnostic remission rate, very large effect sizes (AFQ-Y d = 3.74; PTQ-C d = 3.14), and extraordinarily strong Bayesian evidence of changes in proposed psychological mechanisms. These findings are clinically significant given the prevalence of childhood depression and the need for efficient interventions. Effects generalized to comorbid disorders and behavioral problems observed by parents, suggesting that the transdiagnostic approach grounded in psychological flexibility may be particularly potent. Although requiring replication in randomized controlled trials with blinded assessors and multiple therapists, this preliminary study offers a promising alternative for childhood depression treatment.