A meta-analysis of group cognitive–behavioral therapy as an antidepressive treatment: Are we getting better?

Abstract

This meta-analysis examines temporal changes (time trends) in the effects of group cognitive–behavioral therapy (GCBT) as a treatment for unipolar depression. In this exploratory study, 37 studies (comprising 1,016 patients) conducted between 1980 and 2015 were included, and their effect sizes (ESs) were quantified as Hedge’s g based on the Beck Depression Inventory (BDI) and the Hamilton Rating Scale for Depression (HRSD). In the main analyses, within-group (prepost) comparisons (k = 35 for the BDI, and k = 14 for the HRSD) and controlled (between group) comparisons (k = 16 for the BDI) were conducted. The average within-group ES was 1.33 (95% confidence interval [CI] = 1.16–1.19) for the BDI and 1.56 for the HRSD (95% CI = 1.20–1.90). The publication year of each study was examined as a linear metaregression predictor of ES. The results showed a significant increase in effect sizes for GCBT over time, with the BDI as outcome measure. However, this was not found for the HRSD as outcome measure. Subgroup analyses were performed on selected moderator variables to determine whether there was any covariation with effect sizes. These analyses revealed that in trials conducted without following a set manual, effect sizes increased with time—an association not found for trials using a manual. Potential causes and implications are discussed.

Cognitive behavior therapy for adult depressive disorders in routine clinical care: A systematic review and meta-analysis

“…The ES of 1.51 in the present meta-analysis is significantly higher (t(78) = 5.47, p<0.0001) than the one reported by Hans and Hiller (2013b). It is also in line with pre-post ESs for efficacy studies reported by other meta-analyses (Cristea et al, 2017;Johnsen & Thimm, 2018;Rubin & Yu, 2017). This is reassuring to clinicians in terms of choosing CBT for the treatment of depression in real-word clinical settings, despite the fact that more recent publications were associated with lower ES.…”

“…However, this meta-analysis has been criticized on methodological grounds (Cristea et al, 2017;Ljótsson et al, 2017). Also, the same research group did not find year of publication to be a significant moderator in a meta-analysis of mindfulness-based cognitive therapy (Thimm & Johnsen, 2020), and even that the effect size increased significantly over time in a meta-analysis of group CBT for depression (Johnsen & Thimm, 2018). Thus, it is possible that our result for this moderator is a random finding and should be interpreted with caution.…”

Exploring the role of CBT in the self-management of type 2 diabetes: A rapid review

“…Burlingame et al ( 2016 ) conducted a meta-analysis that found no difference in treatment outcomes between individual and group formats. Although the therapeutic factors, such as social support in a group setting (Johnsen & Timm 2018 ), may differ, individual and group formats share the same treatment techniques and philosophy. Therefore, the environment, the needs of the individuals and socioeconomic factors must be considered when choosing an individual or group-based intervention.…”

Untitled

“…It may also be that this is due to habituation or instrumental conditioning since this is often repeated without consequences, which may reduce the use of these procedures. One way to prevent this is to introduce false alarms to disabling the behavior, but this must not be done too often as this can be habituated [16].…”