Acceptance and Commitment Therapy versus Traditional Cognitive Behavioral Therapy: A Systematic Review and Meta‑analysis of Current Empirical Evidence

Full reference: Ruiz, F. J. (2012). Acceptance and Commitment Therapy versus Traditional Cognitive Behavioral Therapy: A Systematic Review and Meta-analysis of Current Empirical Evidence. International Journal of Psychology and Psychological Therapy, 12(2), 333-357.

Type of work: Systematic review and meta-analysis

Central thesis and objectives

The central thesis of Ruiz is that significant controversy remains regarding the empirical status of Acceptance and Commitment Therapy (ACT) and its presumed distinct characteristics relative to traditional cognitive behavioral therapy (CBT). The author argues that a rigorous systematic review and meta-analysis is necessary to clarify whether ACT provides something genuinely new in relation to traditional CBT or merely represents a reformulation of the same techniques. The stated objective is to conduct a quantitative synthesis of studies that have empirically compared ACT versus CBT in terms of outcomes and processes of change, using full face-to-face treatment protocols (excluding self-help and component studies).

The fundamental research question is whether ACT is more efficacious than CBT, whether it works through different processes of change (as the ACT model proposes), or whether both treatments are equivalent. Ruiz argues that this question is "somewhat sterile" unless systematic comparison and analysis of potential mediators and moderators are conducted.

Theoretical framework

Ruiz presents a detailed historical and theoretical analysis of the evolution from pure behavioral therapy (based on Skinner, 1953; Wolpe, 1958) to the second generation of behavioral therapies that integrated cognitive techniques (1960s-1980s), culminating in the "third generation" of behavioral therapies emphasizing acceptance and mindfulness. ACT is identified as "probably the most representative" of these third-generation therapies.

The theoretical framework of ACT, according to Ruiz, is explicitly based on Hayes's (1993) Functional Contextualism (FC), which is a pragmatic philosophy of science and behavior. Unlike traditional CBT, which assumes that therapy must change the content of cognitions, ACT emphasizes changing the individual's relational context with private events. The argument is that within FC, there is no attempt to directly control or change thoughts and beliefs; rather, the goal is to alter their functions through contextual changes.

Ruiz describes ACT as eclectic but always guided by the implementation of psychological flexibility. This is defined as the ability to be in contact with present-moment private experiences without needing to avoid them, while adjusting behavior according to what the situation requires to pursue valued ends. The model proposes six interrelated processes of change: acceptance, defusion, self as context, contact with the present moment, values, and committed action.

Traditional CBT, in contrast, according to the author, is based on a diverse set of techniques and models that share the fundamental principle that modification of the content of cognitions (dysfunctional thoughts, automatic attitudes, beliefs) is the central mechanism of therapeutic change.

Argument development

Ruiz conducted a systematic search in PsycInfo and Scopus through July 2012 using the terms "(acceptance and commitment therapy) AND (cognitive behavior therapy OR behavior therapy)." A total of 352 and 332 documents were identified respectively. Three unpublished studies were obtained from the ACT Listserv. Inclusion criteria were established: outcome or mediation/moderation studies comparing complete face-to-face treatments of ACT versus some form of traditional CBT.

Of 23 documents describing empirical comparisons of ACT versus CBT, five were excluded for methodology (self-help or component studies). The remaining 16 studies were included in meta-analysis. The total number of participants at baseline was N = 954. Separate meta-analyses were conducted for each g index (effect size) at different study phases according to outcome measure to avoid problems of statistical dependence.

Main meta-analysis results:

On primary outcome measures (combined pre-treatment to post-treatment and follow-up), ACT surpassed CBT in all cases. Specifically, the combined effect size magnitude was small-to-medium (g = 0.40, 95% CI: 0.16, 0.64; Z = 3.23, p = .001). At post-treatment, ACT showed advantage (g = 0.37, p < .01), and this advantage was maintained at follow-up (g = 0.42, p < .01). There were no significant differences for anxiety symptoms, where a positive trend for ACT was observed (g = 0.25 at post-treatment). For depression (g = 0.27, p = .12 at post-treatment), differences were not statistically significant.

On process measures (processes of change proposed by ACT versus CBT), ACT showed greater impact on its proposed processes (g = 0.38, p < .01) but no differences in CBT's proposed processes (g = 0.05, p = .05, nonsignificant).

Regarding heterogeneity, heterogeneity between studies was expected. The author applied moderator analyses (categorical variables using ANOVAs in mixed effects models). The presence of cognitive techniques in CBT packages showed that effects did not differ significantly between CBT with cognitive techniques (g = 0.39) and CBT without techniques (g = 0.34; p = .85). Treatment format (group versus individual) did not show significant differential effects although the mean was greater with group interventions (g = 0.50) versus individual interventions (g = 0.31). Geographic region (North America: g = 0.31; Europe: g = 0.58) showed differences, with larger effects in Europe.

Meta-regression of continuous variables revealed that number of sessions, total hours, gender, mean age, publication year, and pre-treatment g index did not reach statistical significance in prediction.

Implications

Ruiz argues that the results have several important implications. First, ACT appears to generally achieve better results than CBT when comparing primary outcomes. Second, ACT reduces anxiety symptoms to the same degree as CBT, and possibly produces greater reductions of depression symptoms. Third, ACT shows greater impact on its proposed processes of change, while both treatments impacted CBT's proposed processes similarly.

The author emphasizes that although significant heterogeneity exists, ACT shows "an interesting pattern" in that it operates through its proposed processes to a greater degree than CBT, yet both produced equivalent symptom reductions in many cases. This suggests that multiple types of treatments can impact ACT outcomes, but the factor differentiating ACT might be the mediators through which it operates.

Regarding outcome measurement, Ruiz notes that it is notable that current ACT studies have been mostly compared in relation to symptom reduction or CBT-related measures. However, he argues that the meta-analysis results remain "basically the same" when applying the Duval and Tweedie procedure.

Significance and contribution

This work constitutes the first comprehensive systematic review and meta-analysis comparing ACT and traditional CBT. It provides a rigorous quantitative synthesis of the empirical literature addressing whether ACT represents a distinctive contribution within the broader framework of cognitive-behavioral therapy. The primary contribution is both methodological (rigorous application of meta-analysis with publication bias control) and substantive (demonstration that ACT appears to operate through different mechanisms of change than CBT, although both produce symptom improvements). The findings suggest that ACT may not be simply a reformulation of CBT but an approach with distinctive characteristics regarding processes of change, which has implications for understanding third-generation behavioral therapies.