Which Type of Therapy Works Best? New Study Calls It a Draw

A head to head comparison of psychodynamic versus cognitive-behavioral therapy

Posted Jun 04, 2017

One of the eternal debates in mental health is over which type of psychotherapy works best.  The argument has focused mainly on two types of therapy:  psychodynamic versus cognitive-behavioral.  While I’m sure to be criticized over my attempt to conceptualize each in one sentence, psychodynamic therapy emphasizes the value of gaining insights into one’s unconscious conflicts and the less than ideal strategies we have developed to deal with our past.  Cognitive-behavioral therapy (CBT), by contrast, focuses more on the here and now in trying to help acquire new skills to master maladaptive thoughts and behaviors. 

For quite some time, advocates of CBT have invested quite a bit of effort into demonstrating that it works through the use of randomized clinical trials.  Psychodyamic therapy, with some notable exceptions, has lagged behind in this effort, prompting critics to claim that the treatment was not an “evidence-based therapy.”  More recently, however, a number of research studies have been published to test the efficacy of psychodynamic therapy against both no treatment and against other treatments such as CBT or medications. 

A new study just published in the American Journal of Psychiatry now attempts to combine all of these studies in a procedure called a meta-analysis to provide a more definitive answer regarding the relative efficacy of psychodynamic therapy versus other treatments. Others have tried this before and the results have actually been somewhat inconsistent, but the authors of this study try to argue that theirs is the first attempt to test directly the hypothesis that psychodynamic therapy is equivalent in efficacy to other treatments.  This gets a bit technical, but the authors try to make the point that failing to show that two treatments are not statistically different from each other (which is what others have done) is not the same as directly testing whether or not two treatments are essentially equivalent to each other.  

The researchers were able to find 23 different studies, encompassing 2,751 adults, that attempted to compare psychodynamic therapy to something else (in the vast majority of cases it was CBT but in two studies in was medications).  The trials were done for many different types of psychiatric disorders but most commonly depression, anxiety disorders, personality disorders, and eating disorders

Another interesting wrinkle to this meta-analysis is that the authors also tried to rate each study as to its overall quality and also the degree to which the authors of the individual studies might have been biased towards one type of therapy over another, something called “researcher alliance.”

Overall, the researchers found that the psychodynamic therapy worked about as well as CBT based treatments, regardless of diagnosis, study quality, or even researcher allegiance.  Interestingly, they did find a slight edge for CBT across studies, but the difference was quite small and within their predefined zone of equivalence.  They  concluded that their analyses suggest that psychodynamic therapy is equivalent to other types of therapy and is deserving of being called “evidence based,” at least for the conditions they examined. 

Will this end the debate about psychotherapy or the snubbing of psychodynamic therapy as being non evidence-based?  It’s unlikely.  While people use techniques such as meta-analysis to combine the results of multiple studies into one and provide a clearer answer to an important question, there are still enough moving parts in these kinds of studies for different researchers to arrive at different conclusions.  As mentioned, this particular question has been examined before through the use of meta-analysis, and different research groups have come up with different answers.  The authors of this study attempt to argue that their analyses are more valid than some previous attempts, and some of their points do seem to have some merit.

As a child psychiatrist, I also need to point out that all of the studies used in this article came from adults.  For adolescents and especially for younger children, there are many studies supporting the efficacy of CBT but so few good studies of psychodynamic therapy that a meta-analysis like this can’t even be done yet.  I’m personally skeptical that traditional psychodynamic therapy will hold up as well in pediatric samples but this study suggests that it is at least worth a look.  In the discussion section of this article, the authors also suggest that it may be more productive to think about which therapy works best for which people rather than looking at more global measures.  Finally, research has suggested that often what matters more is the individual characteristics of the therapist rather than the specific modality being used.  As The Who once sang, “It’s the singer not the song that makes the music move along.”

@copyright by David Rettew, MD

David Rettew is author of Child Temperament: New Thinking About the Boundary Between Traits and Illness and a child psychiatrist in the psychiatry and pediatrics departments at the University of Vermont College of Medicine.

Follow him at @PediPsych and like PediPsych on Facebook.


Steinert C, Munder T, et al.  Psychodynamic Therapy: As Efficacious as Other Empirically Supported Treatments? A Meta-Analysis Testing Equivalence of Outcomes.  American Journal of Psychiatry 2017. epublication ahead of print.