by Michael D. Anestis, Ph.D.
I absolutely LOVE it when impressive researchers compare two evidence-based treatments to one another in the same study. Comparative efficacy is a wonderful thing and we don't get nearly enough of it. I just came across a great example of this, as Evan Forman, Jena Shaw, Elizabeth Goetter, James Herbert, Jennie Park, and Erica Yuen just published a follow-up to an earlier comparision of cognitive behavioral therapy (CBT) and acceptance and commitment therapy (ACT) in Behavior Therapy, this time reporting results from an 18-month post-treatment assessment point.
In the earlier study, Forman and colleagues (2007) reported that there were no differences between ACT and CBT immediately post-treatment in a randomized controlled trial (RCT) examining a range of outcomes including depressive symptoms, anxiety symptoms, and overall functioning. This was compelling evidence and supportive of the notion that ACT is a promising emerging treatment capable of producing results comparable to gold standard evidence-based psychotherapy for certain populations. As the authors noted in this study, however, it is always important to ask not only how folks are doing right after treatment, but also whether or not treatment gains are maintained over time.
The total sample included 132 undergraduates seeking treatment at a counseling center. Importantly, participants were not required to meet diagnostic criteria for a DSM-IV-TR disorder. Rather, inclusion criteria centered around moderate levels of depression and/or anxiety as measured by the BDI-II and BAI (scores greater than 9). In most healthy undergradaute samples I've collected, the mean for the BDI-II comes in around 7, with a standard deviation of 7, so I'm not entirely comfortable with considering 9 a measure of moderate depression symptoms (in fact, "moderate" for the BDI-II is typically defined by a score of 20-28 and the maximum score is 63). The authors noted that the goal was to keep inclusion criteria broad in order to maximize external validity (e.g., the degree to which the sample and the results reflect what would typically be seen in the "real world"). That's a valid goal, but it is VERY important to keep the context of the sample in mind when interpreting the findings.
As far as diagnoses go, the authors used structured diagnostic interviews and reported that 49.2% met criteria for an anxiety disorder and 37.1% met criteria for a depressive disorder.
On to the results. 91 participants provided follow-up data and were included in the analyses. That being said, the authors used both intent-to-treat analyses and a second idetentical set of analyses that only considered the completers. I'll present the ITT results here.
The authors found that the group of individuals receiving CBT reported greater maintenance of treatment effects for depression and general functioning than did individuals who received ACT. There was a non-significant trend (p = .08) for quality of life favoring the CBT group and no between group differences on anxiety. Putting these results in another context, the authors ran analyses considering clinical significance by comparing the percentage of individuals in each group (CBT vs ACT) who were "reliably recovered" on each outcome measure (see the original paper for a thorough description of the meaning of that phrase). The found the following:
- Depression (BDI-II): 81.8% of CBT patients vs. 60.7% of ACT patients
- Anxiety (BAI): 72.7% of CBT patients vs. 56.0% of ACT patients
- General functioning: 46.4% of CBT patients vs. 22.6% of ACT patients
- Quality of life: 37.8% of CBT patients vs. 22.9% of ACT patients
So, on the whole, the results pain a rather plainly superior long-term outcome for CBT relative to ACT with this particular population, but it is VERY important to keep our understanding of those results within their context. This is a strong study, but it's only one study. This is a valid sample, but not a particularly severe one. The authors used protocols specific to diagnosis, but not everyone had a diagnosis and results for specific diagnoses were not compared to one another here. Also, the results do not mean that ACT does not work...they simply point towards CBT working better in the long haul. It will be interesting to see if other studies like this emerge and, if so, whether these results will be replicated. If they are, confidence in their meaning will increase significantly. In the meantime, this is something to keep in mind.
Dr. Anestis is an incoming assistant professor in the Department of Psychology at the University of Southern Mississippi
Articles cited in this post:
Forman, E.M., Herbert, J.D., Moitra, E., Yeomans, P.D., & Geller, P.A. (2007). A randomzied controlled effectiveness trial of acceptance and commitment therapy and cognitive therapy for anxiety and depression. Behavior Modification, 31, 772-799.
Forman, E.M., Shaw, J.A., Goetter, E.M., Herbert, J.D., Park, J.A., & Yuen, E.K. (in press). Long-term follow-up of a randomized controlled trial comparing acceptance and commitment therapy and standard cognitive behavior therapy for anxiety and depression. Behavior Therapy.