Category: Adult Depression / Dysthymia

PS3- #B46 - Systematic Review and Meta-Analysis of the Effects of Behavioral Characteristics on Behavioral Activation

Friday, Nov 17
11:00 AM – 12:00 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Behavioral Activation | Therapy Process | Depression

Behavioral activation (BA) for depression aims at reducing depression through changing some behavioral characteristics, such as environmental reward, goal-directed behavior, and avoidance (Manos et al., 2010). Although BA is a more effective treatment of depression than control group (Okajima et al., 2011), little is known about how the degree of the change in behavioral characteristics affects BA. The purpose of this study is to examine the effects of changes in behavioral characteristics on BA using meta-analysis.

We searched previous literature using PsycINFO and PubMed, and the terms used “behavioral activation”, “behavioural activation”, and depress* in the Title/Abstract (until April, 2016). In total of 877 articles, the following criteria were finally met by eight papers in this meta-analysis: 1) treatment of BA (Jacobson et al., 1996: Lejuez et al., 2003: Martell et al., 2003), 2) RCT, 3) measures of behavioral characteristics in pre- and post-treatment. Risk of bias about these articles was estimated using CCAR. To estimate the effect size and heterogeneity, we used the “meta” package of R version 3. 3. 2.

Risk of bias was calculated to be 3.11 point on an average. Five articles measured environmental reward, three measured goal-directed behavior, and four measured avoidance. BA was a more effective treatment to change of environmental reward (Hedges’ g = 0.51, 95%CI = from 0.30 to 0.73), goal-directed behavior (Hedges’ g = 0.46, 95%CI = from 0.12 to 0.81), and avoidance (Hedges’ g = -0.67, 95%CI = from -1.20 to -0.15) on post treatment rather than control group, and all the effect sizes were medium. Low heterogeneity was observed in environmental reward (I 2 = 0%, p = .57) and goal-directed behavior (I 2 = 36%, p = .21), whereas high heterogeneity observed avoidance (I 2 = 78%, p = .01).

Results of meta-analysis revealed that BA had a medium effect size on each changes of behavioral characteristic. It is considered that BA treatment process is statistically consistent with the behavioral theory of depression (Manos et al., 2010). Results of environmental reward and goal-directed behavior are low in heterogeneity and the improvement of these characteristics is related to BA (Kanter et al., 2009). On the other hand, the result of avoidance has high heterogeneity. Modifying the avoidance is central elements of format the Martell et al. (2003) rather than the other format. Future research should concentrate on examining the effects of BA divided by the format of BA. 

Shuntaro Aoki

a doctoral student
Graduate School of Psychological Science, Health Sciences University of Hokkaido; Research Fellow of Japan Society for the Promotion of Science
Sapporo, Hokkaido, Japan

Suguru Iwano

Faculty of Welfare and Health Science, Oita University

Koki Takagaki

Department of Psychiatry and Neurosciences, Hiroshima University

Yuta Takano

Graduate School of Psychological Science, Health Sciences University of Hokkaido

Yuji Sakano

School of Psychological Science, Health Sciences University of Hokkaido