Elsevier

Clinical Psychology Review

Volume 40, August 2015, Pages 1-14
Clinical Psychology Review

The enduring effects of psychodynamic treatments vis-à-vis alternative treatments: A multilevel longitudinal meta-analysis

https://doi.org/10.1016/j.cpr.2015.05.003Get rights and content

Highlights

  • Examined the enduring impact of dynamic treatments versus non-dynamic treatments.

  • Calculated four ESs; targeted, non-targeted, personality, and combined measures.

  • Treatments were not significantly different at post-treatment for all four ESs.

  • Post-treatment slopes for all four ESs were non-significant.

  • Dynamic and non-dynamic treatments were equivalent at post-treatment and beyond.

Abstract

Although evidence suggests that the benefits of psychodynamic treatments are sustained over time, presently it is unclear whether these sustained benefits are superior to non-psychodynamic treatments. Additionally, the extant literature comparing the sustained benefits of psychodynamic treatments compared to alternative treatments is limited with methodological shortcomings. The purpose of the current study was to conduct a rigorous test of the growth of the benefits of psychodynamic treatments relative to alternative treatments across distinct domains of change (i.e., all outcome measures, targeted outcome measures, non-targeted outcome measures, and personality outcome measures). To do so, the study employed strict inclusion criteria to identify randomized clinical trials that directly compared at least one bona fide psychodynamic treatment and one bona fide non-psychodynamic treatment. Hierarchical linear modeling (Raudenbush, Bryk, Cheong, Congdon, & du Toit, 2011) was used to longitudinally model the impact of psychodynamic treatments compared to non-psychodynamic treatments at post-treatment and to compare the growth (i.e., slope) of effects beyond treatment completion. Findings from the present meta-analysis indicated that psychodynamic treatments and non-psychodynamic treatments were equally efficacious at post-treatment and at follow-up for combined outcomes (k = 20), targeted outcomes (k = 19), non-targeted outcomes (k = 17), and personality outcomes (k = 6). Clinical implications, directions for future research, and limitations are discussed.

Introduction

Contentious debates exist regarding the superiority of competing psychotherapy treatments. As an example, proponents of cognitive-behavioral therapy (CBT) have claimed superiority to alternative treatments for the past several decades (Eysenck, 1994, Hofmann and Lohr, 2010, Siev and Chambless, 2007, Siev et al., 2009, Tolin, 2010). These claims are supported by various meta-analyses over the years. Specifically, Shapiro and Shapiro (1982) meta-analyzed 143 comparative studies and found that behavioral treatments were superior to psychodynamic and humanistic treatments. More recently, Tolin (2010) meta-analyzed 26 studies examining the efficacy of CBT vis-à-vis alternative treatments and concluded that CBT was superior to psychodynamic treatments for depressive and anxiety disorders. In a review of meta-analyses, Hofmann and Lohr (2010) claimed that seven meta-analyses found higher response rates for CBT compared to alternative treatments and only one found higher rates for the comparison treatment.

On the other hand, a substantial body of research continues to indicate uniform efficacy of treatments intended to be therapeutic (Baardseth et al., 2013, Benish et al., 2008, Cuijpers et al., 2013, Imel et al., 2008, Luborsky et al., 2002, Wampold and Imel, in press, Wampold et al., 1997). For example, in a reanalysis and extension of the findings of Tolin (2010), Baardseth et al. (2013) found no evidence of the superiority of CBT compared to alternative treatments intended to be therapeutic for anxiety disorders.

The debate regarding superiority, at least from a meta-analytic perspective, has focused primarily on outcome measured at one point in time (typically at termination) and has also focused on disorder specific symptom change (Wampold & Imel, in press). Advocates of treatments that are focused on character change rather than on symptoms, such as psychodynamic therapies, suggest that the benefits of such treatments are broader based and longer lasting. For example, Shedler (2010) theorized that:

The goals of psychodynamic therapy include, but extend beyond, alleviation of acute symptoms. Psychological health is not merely the absence of symptoms; it is the positive presence of inner capacities and resources that allow people to live life with a greater sense of freedom and possibility…. Such intrapsychic changes may account for long-term treatment benefits [of psychodynamic treatments].

pp 102, 105

Seeking to produce evidence of the sustained benefits of psychodynamic treatments beyond treatment completion, an increasing number of meta-analyses have indicated that the benefits of psychodynamic treatments at post-treatment are maintained at follow-up, and in some instances increase over time (Abbass et al., 2006, Abbass et al., 2009, Leichsenring et al., 2004, Town et al., 2012). For example, Abbass and colleagues (Abbass et al., 2006, Abbass et al., 2009) conducted a series of meta-analyses of controlled trials of short-term dynamic therapy (STDT) and found that STDT was superior to various types of no-treatment or minimal treatment controls on a variety of outcome measures and that the effects were sustained or grew over time. A number of other meta-analyses have substantiated the enduring effects of psychodynamic treatments (Abbass et al., 2011, Driessen et al., 2010, Town et al., 2011) and some have claimed that the benefits of psychodynamic treatments increase over time (e.g., Leichsenring et al., 2004, Town et al., 2012).

Based on these findings, Shedler (2010) asserted that, “Consistent trend[s] toward larger effect sizes at follow-up suggest that psychodynamic therapy sets in motion psychological processes that lead to ongoing change, even after therapy has ended…. [Whereas] the benefits of other (nonpsychodynamic) empirically supported therapies tend to decay over time for the most common disorders” (pp. 101,102). Shedler's (2010) assertion that the benefits of psychodynamic treatments are longer lasting than non-psychodynamic treatments does not appear to be universally accepted nor is it conclusively supported by empirical research. Although evidence suggests that the benefits of psychodynamic treatments are sustained over time and in some instances increase compared to control groups, it is unclear whether non-psychodynamic treatments produce equivalent sustained benefits beyond treatment completion.

There are few empirical studies that have addressed the question of whether the effects of some types of treatment, such as psychodynamic treatments, are longer lasting than alternative types of treatments. Meta-analyses of studies that do exist have produced mixed findings. For example, Anderson and Lambert (1995) examined the effectiveness of STDT compared to alternative treatments for a variety of disorders and found that STDT was equivalent to alternative treatments at post-treatment, but produced superior benefits compared to alternative treatments at follow-up. However, Keefe, McCarthy, Dinger, Zilcha-Mano, and Barber (2014) recently meta-analyzed the impact of psychodynamic treatments compared to alternative treatments for anxiety disorders and found that psychodynamic treatments did not significantly differ from alternative treatments at short-term follow-up and long-term follow-up.

These mixed findings may be a result of several methodological limitations. Specifically, many of the previous studies used no-treatment or minimal treatment control groups. Additionally, the majority of meta-analyses and clinical trials comparing two or more treatments did not directly compare treatments intended to be therapeutic. For example, in a meta-analysis examining the effect of STDT, Anderson and Lambert (1995) failed to identify treatment comparisons that were intended to be therapeutic: “A treatment was classified as ‘alternative’ only when it was either the usual form of treatment for the disorder or it was expected to produce results similar to STDT.” (p. 505). Operating under this definition, non-bona fide comparison treatments were included in the analysis of STDT vs. “alternative” treatments (i.e. hypnosis, dietary advice, supportive treatments, and mutual-help groups), resulting in a bias for STDT. In order to effectively test the superiority of a particular treatment, studies must implement designs that directly compare two or more treatments intended to be therapeutic (see Wampold and Imel, in press, Wampold et al., 1997).

Additionally, the inconsistent findings regarding the long-term superiority of psychodynamic treatments may be related to the historical focus on disorder specific symptom change in psychotherapy research (Wampold & Imel, in press). In a review of psychodynamic effectiveness, Shedler (2010) posited that the benefits of psychodynamic treatments are not limited to the alleviation of symptoms, but rather simultaneously increase clients' inner capacities and resources. If comparative studies solely assess and report symptom-oriented outcomes, they may be failing to capture the lasting benefits of less symptom-oriented therapies.

Lastly, previous meta-analyses have neglected to control for researcher allegiance and treatment dose, potentially contributing to inconsistent findings. Researcher allegiance refers to a researchers preference for a particular treatment and results in better outcomes for the preferred treatment (Munder et al., 2013, Wampold and Imel, in press). Similarly, allegiance may impact the effect of the non-preferred alternative treatment as a result of researchers poorly implementing the non-preferred treatment (Munder, Gerger, Trelle, & Barth, 2011). Treatment dose refers to the amount of sessions or treatment “dose” received by a patient. Treatment dose has been found to be significantly and positively related to treatment outcomes, as such it is imperative to control for differences in dosage in comparative analyses. To the best of our knowledge, the majority of meta-analyses examining the effectiveness of psychodynamic treatments fail to control for differences in treatment dose and explore the effect of researcher allegiance on treatment outcome, resulting in potentially biased results.

The inconsistent evidence pertaining to the long-term benefits of psychodynamic treatments suggests that additional meta-analyses addressing previous limitations are needed. As such, the purpose of the current study was to test the growth of the benefits of psychodynamic treatments compared to alternative treatments beyond treatment completion. Specifically, the present meta-analysis strictly included clinical trials that directly compared at least one bona fide psychodynamic treatment to at least one bona fide non-psychodynamic treatment for a variety of disorders. Non-psychodynamic treatments were not further classified into categories of treatments, as we were specifically interested in testing the lasting impact of psychodynamic treatments compared to non-psychodynamic treatments. Multilevel longitudinal analyses were run to perform a significance test of the growth of the impact of psychodynamic treatments compared to alternative treatments beyond treatment completion for four categories of outcome measures (i.e., all outcome measures, targeted outcome measures, non-targeted outcome measures, and personality outcome measures). Informed by the sizeable body of research finding uniform efficacy at post-treatment, we hypothesize that psychodynamic treatments and bona fide non-psychodynamic treatments will not significantly differ at post-treatment on all outcome measures. Additionally, we hypothesize a significant and positive growth (i.e., slope coefficient) in the impact of psychodynamic treatments compared to non-psychodynamic treatments from post-treatment to follow-up on all types of outcome measures. This second hypothesis is based on evidence that the benefits of psychodynamic psychotherapy at post-treatment increase at follow-up compared to control groups.

Section snippets

Inclusion criteria

For studies to be included in the current meta-analysis they needed to (a) be published in an English-printed peer-reviewed journal, (b) have utilized randomized clinical designs, (c) have examined treatments of adult patients, (d) have utilized direct comparisons of at least two bona fide therapeutic treatments, one of which was psychodynamic and one which was not, (e) have reported outcome data at post-treatment and at least one follow-up assessment, (f) have reported the necessary statistics

Psychodynamic survey

Of the total of 800 IEDTA (n = 600) and SEPI (n = 200) members who were contacted via email, approximately ten percent (9.63%, n = 77) responded. Of the 77 respondents, 75% (n = 58) identified their primary theoretical orientation as psychodynamic and therefore were included in the coding of psychodynamic treatments. The following data is reported for the 58 psychodynamic raters. The majority of survey participants (48%, n = 28) indicated Ph.D. as the highest level of degree completed, followed by

Discussion

The present meta-analysis examined the impact of bona fide psychodynamic treatments compared to bona fide non-psychodynamic treatments both at post-treatment and, most interestingly, beyond post-treatment. In particular, the growth of the impact of bona fide psychodynamic treatments compared to bona fide non-psychodynamic treatments was examined beyond the end of therapy. As hypothesized, psychodynamic and non-psychodynamic treatments were equally efficacious at post-treatment on combined

Role of funding sources

Funding for this study was provided by the Society of Psychotherapy Research (SPR) (Fund-Project: 133-PRJ74RA) as a Small Research Grant in 2013. SPR had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

Contributors

Maleeha Abbas, Brian T. Pace, Noah E. Yulish, Joel G. Thomas, and Megan M. Cullen assisted in the literature search, coding identified studies, and editing the manuscript. D. Martin Kivlighan, III and Bruce E. Wampold designed the study. D. Martin Kivlighan, III wrote the first draft of the manuscript. D. Martin Kivlighan, III, Simon B. Goldberg, and Christoph Fluckiger conducted the statistical analyses. All authors contributed to and have approved the final manuscript.

Conflict of interest

All authors declare that they have no conflicts of interest.

References (46)

  • B.E. Wampold et al.

    Evidence-based treatments for depression and anxiety versus treatment-as-usual: A meta-analysis of direct comparisons

    Clinical Psychology Review

    (2011)
  • A.A. Abbass et al.

    Short-term psychodynamic psychotherapies for common mental disorders

    Cochrane Database of Systematic Reviews

    (2006)
  • A. Abbass et al.

    Short-term psychodynamic psychotherapy for somatic disorders: Systematic review of meta-analysis of clinical trials

    Psychotherapy and Psychosomatics

    (2009)
  • A. Abbass et al.

    The efficacy of short-term psychodynamic psychotherapy for depressive disorders with comorbid personality disorder

    Psychiatry: Interpersonal and Biological Processes

    (2011)
  • M. Borenstein et al.

    Introduction to meta-analysis

    (2009)
  • P. Crits-Christoph

    The efficacy of brief dynamic psychotherapy: A meta-analysis

    The American Journal of Psychiatry

    (1992)
  • P. Cuijpers et al.

    A meta-analysis of cognitive–behavioural therapy for adult depression, alone and in comparison with other treatments

    The Canadian Journal of Psychiatry / La Revue Canadienne De Psychiatrie

    (2013)
  • C. Flückiger et al.

    The sleeper effect: Artifact or phenomenon—A brief comment on Bell et al. (2013).

    Journal Of Consulting And Clinical Psychology

    (2015)
  • C. Flückiger et al.

    How central is the alliance in psychotherapy? A multilevel longitudinal meta-analysis

    Journal of Counseling Psychology

    (2012)
  • L.V. Hedges et al.

    Statistical methods for meta-analysis

    (1985)
  • S.G. Hofmann et al.

    To kill a dodo bird

    The Behavior Therapist

    (2010)
  • T.B. Huedo-Medina et al.

    Assessing heterogeneity in meta-analysis: Q statistic or I2 index?

    Psychological Methods

    (2006)
  • Z.E. Imel et al.

    Distinctions without a difference: Direct comparisons of psychotherapies for alcohol use disorders

    Psychology of Addictive Behaviors

    (2008)
  • Cited by (29)

    • Fifteen years after the INSERM report. Psychoanalysis's efficacy reevaluated

      2021, Evolution Psychiatrique
      Citation Excerpt :

      Meta-analytic operations allow us to obtain a series of independent results of a higher level. Meta-analyses have also shown that psychoanalytic therapies have an equal clinical efficacy compared to all concurrent approaches and for all pathologies, in the short as well as in the long-term [12,13,22,111,114,120–122].13 Compared to other forms of treatment, psychoanalytic therapies prove to be just as effective as other therapies for:

    • Short-term psychodynamic psychotherapy for functional somatic disorders: A systematic review and meta-analysis of within-treatment effects

      2021, Journal of Psychosomatic Research
      Citation Excerpt :

      STPP may yield important relational and personality changes that prevent relapse after treatment [70]. There is some evidence sustained benefits may be seen in other treatments, including CBT [27,40,74]. There are limitations of the literature and our review of it.

    • Is there a sleeper effect of exposure-based vs. cognitive-only intervention for anxiety disorders? A longitudinal multilevel meta-analysis

      2019, Clinical Psychology Review
      Citation Excerpt :

      There is, therefore, no evidence to favor a sleeper effect in the current work. In fact, there is marginal empirical evidence to support a sleeper effect in comparative psychotherapy research in general (Flückiger et al., 2012; Flückiger et al., 2014, 2015; Flückiger & Del Re, 2017; Kivlighan III et al., 2015). As in the case of the current work, none of the previous longitudinal meta-analyses found a growth in relative efficacy at follow-up irrespective of what form of psychotherapy was investigated.

    View all citing articles on Scopus
    View full text