Johathan Shedler

Page 5 of 8 Previous page Next Page

The Efficacy of Psychodynamic Psychotherapy

Jonathan Shedler
University of Colorado Denver School of Medicine

February–March 2010 American Psychologist
© 2010 American Psychological Association 0003-066X/10/$12.00
Vol. 65, No. 2, 98–109 DOI: 10.1037/a0018378

A Rose by Another Name: Psychodynamic Process in Therapies

The “active ingredients” of therapy are not necessarily those presumed by the theory or treatment model. For this reason, randomized controlled trials that evaluate a therapy as a “package” do not necessarily provide support for its theoretical premises or the specific interventions that derive from them. For example, the available evidence indicates that the mechanisms of change in cognitive therapy (CT) are not those presumed by the theory. Kazdin (2007), reviewing the empirical literature on mediators and mechanisms of change in psychotherapy, concluded, “Perhaps we can state more confidently now than before that whatever may be the basis of changes with CT, it does not seem to be the cognitions as originally proposed” (p. 8).

There are also profound differences in the way therapists practice, even therapists ostensibly providing the same treatment. What takes place in the clinical consulting room reflects the qualities and style of the individual therapist, the individual patient, and the unique patterns of interaction that develop between them. Even in controlled studies designed to compare manualized treatments, therapists interact with patients in different ways, implement interventions differently, and introduce processes not specified by the treatment manuals (Elkin et al., 1989). In some cases, investigators have had difficulty determining from verbatim session transcripts which manualized treatment was being provided (Ablon & Jones, 2002).

For these reasons, studies of therapy “brand names” can be highly misleading. Studies that look beyond brand names by examining session videotapes or transcripts may reveal more about what is helpful to patients (Goldfried & Wolfe, 1996; Kazdin, 2007, 2008). Such studies indicate that the active ingredients of other therapies include unacknowledged psychodynamic elements.

One method of studying what actually happens in therapy sessions makes use of the Psychotherapy Process Q-Sort (PQS; Jones, 2000). This instrument consists of 100 variables that assess therapist technique and other aspects of the therapy process based on specific actions, behaviors, and statements made during sessions. In a series of studies, blind raters scored the 100 PQS variables from archival, verbatim session transcripts for hundreds of therapy hours from outcome studies of both brief psychodynamic therapy and CBT (Ablon & Jones, 1998; Jones & Pulos, 1993).10

In one study, the investigators asked panels of internationally recognized experts in psychoanalytic therapy and CBT to use the PQS to describe “ideally” conducted treatments (Ablon & Jones, 1998). On the basis of the expert ratings, the investigators constructed prototypes of ideally conducted psychodynamic therapy and CBT. The two prototypes differed considerably.

The psychodynamic prototype emphasized unstructured, open-ended dialogue (e.g., discussion of fantasies and dreams); identifying recurring themes in the patient’s experience; linking the patient’s feelings and perceptions to past experiences; drawing attention to feelings regarded by the patient as unacceptable (e.g., anger, envy, excitement); pointing out defensive maneuvers; interpreting warded-off or unconscious wishes, feelings, or ideas; focusing on the therapy relationship as a topic of discussion; and drawing connections between the therapy relationship and other relationships.

The CBT prototype emphasized dialogue with a more specific focus, with the therapist structuring the interaction and introducing topics; the therapist functioning in a more didactic or teacher-like manner; the therapist offering explicit guidance or advice; discussion of the patient’s treatment goals; explanation of the rationale behind the treatment and techniques; focusing on the patient’s current life situation; focusing on cognitive themes such as thoughts and belief systems; and discussion of tasks or activities (“homework”) for the patient to attempt outside of therapy sessions.11

In three sets of archival treatment records (one from a study of cognitive therapy and two from studies of brief psychodynamic therapy), the researchers measured therapists’ adherence to each therapy prototype without regard to the treatment model the therapists believed they were applying (Ablon & Jones, 1998). Therapist adherence to the psychodynamic prototype predicted successful outcome in both psychodynamic and cognitive therapy. Therapist adherence to the CBT prototype showed little or no relation to outcome in either form of therapy. The findings replicated those of an earlier study that employed a different methodology and also found that psychodynamic interventions, not CBT interventions, predicted successful outcome in both cognitive and psychodynamic treatments (Jones & Pulos, 1993).

An independent team of investigators using different research methods also found that psychodynamic methods predicted successful outcome in cognitive therapy (Castonguay, Goldfried, Wiser, Raue, & Hayes, 1996). The study assessed outcomes in cognitive therapy conducted according to Beck’s treatment model (Beck, Rush, Shaw, & Emery, 1979), and the findings had been reported as evidence for the efficacy of cognitive therapy for depression (Hollon et al., 1992).12

Investigators measured three variables from verbatim transcripts of randomly selected therapy sessions in a sample of 64 outpatients. One variable assessed quality of the working alliance (the concept working alliance, or therapeutic alliance, is now widely recognized and often considered a nonspecific or “common” factor in many forms of therapy; many do not realize that the concept comes directly from psychoanalysis and has played a central role in psychoanalytic theory and practice for over four decades; see Greenson, 1967; Horvath & Luborsky, 1993). The second variable assessed therapist implementation of the cognitive treatment model (i.e., addressing distorted cognitions believed to cause depressive affect). The third variable, labeled experiencing, beautifully captures the essence of psychoanalytic process:

At the lower stages of [experiencing], the client talks about events, ideas, or others (Stage 1); refers to self but without expressing emotions (Stage 2); or expresses emotions but only as they relate to external circumstances (Stage 3). At higher stages, the client focuses directly on emotions and thoughts about self (Stage 4), engages in an exploration of his or her inner experience (Stage 5), and gains awareness of previously implicit feelings and meanings [emphasis added] (Stage 6). The highest stage (7) refers to an ongoing process of in-depth self-understanding. (Castonguay et al., 1996, p. 499)

Especially noteworthy is the phrase “gains awareness of previously implicit feelings and meanings.” The term implicit refers, of course, to aspects of mental life that are not initially conscious. The construct measured by the scale hearkens back to the earliest days of psychoanalysis and its central goal of making the unconscious conscious (Freud, 1896/1962).13

In this study of manualized cognitive therapy for depression, the following findings emerged: (a) Working alliance predicted patient improvement on all outcome measures; (b) psychodynamic process (“experiencing”) predicted patient improvement on all outcome measures; and (c) therapist adherence to the cognitive treatment model (i.e., focusing on distorted cognitions) predicted poorer outcome. A subsequent study using different methodology replicated the finding that interventions aimed at cognitive change predicted poorer outcome (Hayes, Castonguay, & Goldfried, 1996). However, discussion of interpersonal relations and exploration of past experiences with early caregivers—both core features of psychodynamic technique—predicted successful outcome.

These findings should not be interpreted as indicating that cognitive techniques are harmful, and other studies have reported positive relations between CBT technique and outcome (Feeley, DeRubeis, & Gelfand, 1999; Strunk, DeRubeis, Chiu, & Alvarez, 2007; Tang & DeRubeis, 1999). Qualitative analysis of the verbatim session transcripts suggested that the poorer outcomes associated with cognitive interventions were due to implementation of the cognitive treatment model in dogmatic, rigidly insensitive ways by certain of the therapists (Castonguay et al., 1996). (No school of therapy appears to have a monopoly on dogmatism or therapeutic insensitivity. Certainly, the history of psychoanalysis is replete with examples of dogmatic excesses.) On the other hand, the findings do indicate that the more effective therapists facilitated therapeutic processes that have long been core, centrally defining features of psychoanalytic theory and practice.

Other empirical studies have also demonstrated links between psychodynamic methods and successful outcome, whether or not the investigators explicitly identified the methods as “psychodynamic” (e.g., Barber, Crits-Christoph, & Luborsky, 1996; Diener, Hilsenroth, & Weinberger, 2007; Gaston, Thompson, Gallagher, Cournoyer, & Gagnon, 1998; Hayes & Strauss, 1998; Hilsenroth, Ackerman, Blagys, Baity, & Mooney, 2003; Høglend et al., 2008; Norcross, 2002; Pos, Greenberg, Goldman, & Korman, 2003; Vocisano et al., 2004).


10 The cognitive therapy study was a randomized controlled trial for depression; the psychodynamic therapy studies were panel studies for mixed disorders and for posttraumatic stress disorder, respectively. See the original source for more detailed descriptions (Ablon & Jones, 1998; Jones & Pulos, 1993).

11 See the original source for more complete descriptions of the two therapy prototypes (Ablon & Jones, 1998).

12 The study is one of the archival studies analyzed by Jones and his associates (Ablon & Jones, 1998; Jones & Pulos, 1993).

13 Although the term “experiencing” derives from the humanistic therapy tradition, the phenomenon assessed by the scale—a trajectory of deepening self-exploration, leading to increased awareness of implicit or unconscious mental life—is the core defining feature of psychoanalysis and psychoanalytic therapy.