CPPNJ - The Center for Psychoanalysis and Psychotherapy of New Jersey


The effectiveness of psychoanalytic therapy: evidence from research

Presented in three parts
By Nancie Senet, PhD

Part 2: Shedler tackles the Dodo’s verdict

As you might remember, we last left the Dodo bird as he sat enjoying the continuing success of his famous verdict. But his joy turned out to be short-lived. Even though research had yielded results upholding his verdict that “ everyone has won..” i.e. that all of the various forms of psychotherapy were equally successful, his ruling was largely ignored. A particular form of psychotherapy, cognitive-behavioral therapies (CBT), was touted as the only psychotherapy to be validated as effective by research evidence. The American Psychological Association declared 2000-2010 to be “The Decade of Behavior.” It issued mandates in a series of Reports for the field of psychology to set upon a course to use and to train psychologists in “empirically supported therapies.” The movement gained traction with momentous consequences. Insurance companies, government entities, schools, universities all jumped on board. Cognitive-behavioral therapists and researchers led the way in amassing “empirical data” to support their treatments, treatments that lent themselves to using simple protocols and a time-limited course of therapy. By the time that psychoanalytically trained therapists began to feel the impact of the changed landscape, there was little that could be done to affect the now widespread belief that only the empirically supported therapies, designated so by current research, were valid forms of treatment.

Previous research such as Luborsky’s and the Consumer Reports’ study were buried under the avalanche of “empirical” data from RCT (randomized controlled trial) methodologies used to validate EST’s (empirically supported therapies). Cognitive-behavioral therapies emerged as the only empirically supported method of treatment.

Drew Westen and his researchers, Novotny and Thompson-Brenner, took on the task of questioning the validity of these RCT methodologies. In their comprehensive 2004 article, “The empirical status of empirically supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials,” they provided evidence that the assumptions used in constructing RCT research were faulty, that they are “neither well validated nor broadly applicable to most disorders and treatments.” (p.632) The assumptions that Westen et.al considered to be non-validated are the following: 1. That psychopathology is highly malleable 2. That most patients can be treated for a single problem or disorder 3. That psychiatric disorders can be treated independently of personality factors unlikely to change in brief treatments, and 4. That experimental methods provide a gold standard for identifying useful psychotherapeutic packages.

Here is a portion of a fairytale with which they begin their article. “Once upon a time, psychotherapists practiced without adequate empirical guidance, assuming that the therapies of their own persuasion were the best…Then a group of clinical scientists developed a set of procedures that became the gold standard for assessing the validity of psychotherapies…the Knights identified interventions for a number of disorders that showed substantial promise…In the excitement, however, some important details seemed to get overlooked…Most of the time, psychotherapy outcome studies supported the preferred position of the gallant knight who happened to conduct them (Sir Grantsalot)…Many graduate programs began teaching new professionals only those treatments that had the imprimatur of Empirical Validation, clinicians seeking licensure had to memorize the tales told by the Knights and pledge allegiance to them on the national licensing examination, and insurance companies used the results of controlled clinical trials to curtail the treatment of patients who did not improve in 6 to 16 sessions, invoking the name of Empirical Validation.” (Pp. 631-632)

By 2010 a rather respectable number of studies had been published that were not conducted under the direction of CBT researchers. Jonathan Shedler took the next step. He conducted a meta-analysis of the existing outcome research that had included participants who had been in psychoanalytic or psychodynamic therapy. There were some noteworthy findings of immense importance to the psychoanalytic community issuing from that data. He began his article, “The efficacy of psychodynamic psychotherapy,” in the following way, telling a story that echoed Westen”s. “There is a belief in some quarters that psychodynamic concepts and treatments lack empirical support or that scientific evidence shows that other forms of treatment are more effective…” (p.98)

Shedler then presented the evidence for a contrary view. Using the amassed research data, he showed that there is considerable scientific support for the efficacy and effectiveness of psychodynamic psychotherapy. He wrote that the “blanket assertions that psychodynamic approaches lack scientific support are no longer defensible.” (p.106) Psychodynamic therapy was found to be effective for a wide range of psychological disorders. Not only was it effective but the benefits from psychodynamic therapy actually increased over time, while those from other forms of “empirically supported” therapies decayed after completion of treatment. Psychodynamic therapy was particularly effective with severe personality disorders. Two recent studies had yielded evidence that long-term psychodynamic therapy provided enduring benefits for borderline personality disorder. No other treatment had produced such benefits - so much for the Dodo’s verdict.

And Shedler did indeed next turn his attention to that verdict. He reviewed the body of literature by researchers who had examined previous empirical outcome studies, studies that have been used to provide the “evidence” in support of the effectiveness for non-psychodynamic therapies. Those researchers concluded that unacknowledged psychodynamic elements were the active ingredients responsible for favorable treatment outcomes in these other therapies. There was an avalanche of evidence indicating that the mechanisms of change in cognitive therapies were not those presumed by their theory. The basis of change did not seem to be cognitive as originally proposed. Instead factors that are core features of psychodynamic therapy, such as the quality of the therapeutic alliance, discussion of interpersonal relations, and exploration of past experiences with early caregivers, were predictive of successful therapy outcomes.

Is there a winner after all? Shedler would say “yes.” He makes the case for it being “psychodynamic process,” be it in psychodynamic therapy or as “a rose by another name” in other therapies.


Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. American Psychologist, 65 (2), 98-109.

Westen, D., Novotny, C.M., & Thompson-Brenner, H. (2004). The empirical status of empirically supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials. Psychological Bulletin, 130, 631-663.

Next in this series on the effectiveness of psychoanalytic therapy: Part 3: Dodo Redux