LETTERS TO THE EDITOR
Design Considerations Related to Short-Term Psychodynamic Psychotherapy To the Editor: e read with interest the article by Abbass et al.1 who reported metaanalytic data on short-term psychodynamic psychotherapies (STPPs) in youth. We applaud their adherence to contemporary meta-analytic standards, and their report provides valuable and clearly presented information that merits attention. Within the study strengths, we have 3 concerns and suggest pathways forward. First, limited focus was given to the negligible overall between-groups effects. This is particularly troublesome because most included studies did not compare STPP to interventions that are designated empirically supported treatments. In addition, specific effect sizes were reported in the abstract for within-subjects effects, but not for between-subjects effects. The investigators are balanced in their discussion of the preliminary nature of these data, and the nascent approach to using randomized controlled trials with STPP can be challenging in the context of meta-analysis; nevertheless, the proportion of text and quantitative data devoted to withinsubjects effects is disproportional to this crucial null finding. Second, interventions and outcomes should be specific and falsifiable. Psychodynamic psychotherapy was characterized as consisting of a “focus on emotion, exploration of attempts to avoid distressing thoughts and feelings, identification of patterns, discussion of past experience, focus on interpersonal relationships, focus on the therapy relationship, and exploration of wishes and fantasies” (p. 863).1 However, this is not unique to any psychosocial treatment, and this depiction is essentially unfalsifiable. For instance, by virtue of treating anxiety and depression, a clinician must identify these conditions as emotional states and by discussing them will show attempts to avoid distress and discuss past experiences, focus on how these affect a patient’s social life, make sure the patient “buys in,” and consider treatment objectives in the context of future goals. Cognitive-behavioral therapy is not immune to these difficulties, and many patients who purportedly receive cognitive-behavioral
W
therapy for anxiety do not receive the essential ingredient of exposure treatment (despite many myths such as exposure being harmful having been long since debunked).2 Indeed, the term cognitive-behavioral refers only to addressing thoughts and behaviors, but not with a falsifiable “how” to do this. Although many of the outcomes are better defined in this meta-analysis than in older trials, the intervention itself is not defined in specific or falsifiable terms. Third, this article highlights the difficulty in determining the evaluation criteria for preliminary data from new treatment approaches. We would like to introduce 2 options to evaluate incremental utility: one would be a refinement and concentration of specific procedures, and the other would be to claim unique variance in outcomes (e.g., a new approach is more effective with nonresponders from a gold-standard treatment, or perhaps it works more quickly than existing interventions). For instance, the investigators referenced a falsifiable “sleeper effect” hypothesis in STPP, which merits further testing with larger samples. However, in addition to considering “what treatment works,” we believe it is fair to consider “how much” and “how quickly” and “at what relative cost in time and expenses”; in the present meta-analysis, STPP frequently required more than the standard 12 to 16 sessions used in anxiety/depression trials for cognitive-behavioral therapy, with no observed gain in outcomes, and even singlesession exposure protocols have shown efficacy for specific phobias.3 It was stated in the abstract that STPP “may be effective in children and adolescents”1; because the null hypothesis cannot be proved, this statement is true for all interventions, so we would like to consider “how much” better and “how quickly” such results are achieved. We are grateful to Abbass et al.1 for bringing data to the discussion of how to best help struggling youth. We believe an intervention that is specified in falsifiable and specific terms and shows incremental improvement over current practice through the distillation of components or claiming of new outcome variance can achieve these ends, which will require further work. Given the importance of remaining focused on outcome, new options that meet these criteria should be welcomed.
JOURNAL OF THE AMERICAN ACADEMY OF C HILD & ADOLESCENT PSYCHIATRY VOLUME 52 NUMBER 11 NOVEMBER 2013
www.jaacap.org
1213
LETTERS TO THE EDITOR
Alessandro S. De Nadai, M.A. Eric A. Storch, Ph.D. University of South Florida Tampa
[email protected]
Disclosure: Dr. Storch has received research funding in the past 2 years from the National Institutes of Health, Centers for Disease Control, the Agency for Healthcare Research and Quality, All Children’s Hospital Guild Endowed Chair, and Janssen Scientific Affairs. He receives textbook honoraria from Springer, the American Psychological Association, and Lawrence Erlbaum. He has served as a consultant for Prophase, Inc., and CroNos, Inc., and has served on the speakers’ bureau for the International Obsessive Compulsive Disorder Foundation. Mr. De Nadai reports no biomedical financial interests or potential conflicts of interest. 0890-8567/$36.00/ª2013 American Academy of Child and Adolescent Psychiatry http://dx.doi.org/10.1016/j.jaac.2013.08.012
REFERENCES 1. Abbass AA, Rabung S, Leichsenring F, Refseth JS, Midgley N. Psychodynamic psychotherapy for children and adolescents: a meta-analysis of short-term psychodynamic models. J Am Acad Child Adolesc Psychiatry. 2013;52:863-875. 2. Deacon BJ, Farrell NR. Therapist barriers to the dissemination of exposure therapy. In: Storch EA, McKay D, editors. Handbook of Treating Variants and Complications in Anxiety Disorders. New York: Springer; 2013:363-373. € L, editors. Intensive One-Session 3. Davis TE III, Ollendick TH, Ost Treatment of Specific Phobias. New York: Springer; 2012.
Dr. Abbass et al. reply: e received the comments submitted by Mr. De Nadai and Dr. Storch and provide the following responses and further information about our study.1 First, it is important to note that we measured within- and between-group effects. We reported the absence of between-group effects first in the Results section of the abstract and then elsewhere. The finding of large and increasing within-group effects in a reasonably well-performed set of trials is noteworthy and we also studied and reported this. This runs counter to the ongoing perception in some circles that psychodynamic treatments are not effective, despite broad evidence to the contrary published in recent years. Also, the comparators included bona fide interventions: only 3 of the studies included what would be considered control conditions, whereas the remainder offered conditions as robust as prolonged exposure, group therapy, or family therapy. These represent bona fide treatment approaches to these sets of problems in children and youth.
W
1214
www.jaacap.org
Second, Mr. De Nadai and Dr. Storch appear to be confusing 2 separate things: treatment process research versus outcome research. They are correct in saying that the features we referred to are “not unique to any psychosocial treatment,” but in fact a review of empirical studies by Blagys and Hilsenroth2 and others since then have shown that these elements are consistently found to be more characteristic of psychodynamic therapy than of cognitive-behavioral therapy (CBT): sessions of the 2 therapies can be reliably distinguished based on ratings of these features. Blagys and Hilsenroth3 in turn reported on those features that are more characteristic of CBT than of short-term psychodynamic psychotherapy (STPP), although that does not mean that those features are unique to CBT. In fact, we never made this claim. However, this was a meta-analysis of treatment outcomes from STPP and not a study of treatment processes and correlates of outcome. So the same argument and criticism can be leveled against any outcome-based meta-analysis of psychotherapy approaches. We concur with Mr. De Nadai and Dr. Storch that specificity of interventions in specific populations would allow more generation of specific data and treatment recommendations. Unfortunately, research into this is sorely lacking across any psychotherapy method or population in childhood and adolescent mental illness. With regard to treatment efficiency and cost, 8 of the studies averaged 25 or fewer sessions and the (unweighted) mean was 22.1 sessions. It is noteworthy that the samples included borderline personality disorder, anorexia nervosa (2 studies), and posttraumatic stress disorder, conditions in which any approach is challenged to bring robust within-group effects in such short treatments. So, overall these are cost-effective approaches that enable more access in economically strained health systems. Although we disagree with some of Mr. De Nadai and Dr. Storch’s points as noted, we also were critical of this body of research and expressed hope and expectation for improved quality and quantity of research in STPP. We are pleased that future research looks to improve on existing methods in several directions. For example, the Improving Mood with Psychoanalytic and Cognitive Therapies (IMPACT) study currently underway, which we referred to at the end of the article, is testing precisely this kind of more focused hypothesis.4 Specifically, these researchers are examining the capacity for STPP to decrease
JOURNAL
OF THE
AMERICAN ACADEMY OF C HILD & ADOLESCENT PSYCHIATRY VOLUME 52 NUMBER 11 NOVEMBER 2013
LETTERS TO THE EDITOR
relapse in the medium term for adolescents with moderate/severe depression, because this is a group that has been identified as less responsive to other forms of intervention. This also will allow for testing of delayed accrual of gains after treatment, referred to as the “sleeper effect.” The IMPACT study also will evaluate cost-effectiveness. Further regarding the second point above, these researchers also are testing whether the process of STPP can be empirically distinguished from CBT and standard clinical care, using the rating scale developed by Blagys and Hilsenroth. Likewise, Ulberg et al.5 are testing a specific mechanism, by comparing STPP for depressed adolescents with or without transference interpretations: this would address a specific question about specific procedures and the effects on outcome. Allan Abbass,
M.D.
Dalhousie University Halifax, Nova Scotia, Canada
[email protected]
Sven Rabung,
Ph.D.
University Medical Center Hamburg Eppendorf Hamburg, Germany Alpen-Adria-Universit€ at Klagenfurt Klagenfurt, Austria
Falk Leichsenring,
D.Sc.
Nick Midgley,
The authors thank Mark Hilsenroth, Ph.D., of Adelphi University, for his comments on this letter. Disclosure: Please see the disclosure statement in the original article published in August 2013. 0890-8567/$36.00/ª2013 American Academy of Child and Adolescent Psychiatry http://dx.doi.org/10.1016/j.jaac.2013.08.010
REFERENCES 1. Abbass A, Rabung S, Leichsenring F, Refseth J, Midgley N. Psychodynamic psychotherapy for children and adolescents: a meta-analysis of short-term psychodynamic models. J Am Acad Child Adolesc Psychiatry. 2013;52:863-875. 2. Blagys M, Hilsenroth M. Distinctive activities of short-term psychodynamic-interpersonal psychotherapy: a review of the comparative psychotherapy process literature. Clin Psychol. 2000;7:167-188. 3. Blagys MD, Hilsenroth MJ. Distinctive features of short-term cognitive-behavioral psychotherapy: a review of the comparative psychotherapy process literature. Clin Psychol Rev. 2002;22:671-706. 4. Goodyer IM, Tsancheva S, Byford S, et al. Improving Mood with Psychoanalytic and Cognitive Therapies (IMPACT): a pragmatic effectiveness superiority trial to investigate whether specialized psychological treatment reduces the risk for relapse in adolescents with moderate to severe unipolar depression: study protocol for a randomized controlled trial. Trials. 2011;12:175. 5. Ulberg R, Hersoug AG, Høglend P. Treatment of adolescents with depression: the effect of transference interventions in a randomized controlled study of dynamic psychotherapy. Trials. 2012;13:159.
University of Giessen Giessen, Germany
Johanne S. Refseth,
M.Sc., Psych.D.
Anna Freud Centre University College London London, UK
M.Sc. Psych
Alternative to Violence Oslo, Norway
JOURNAL OF THE AMERICAN ACADEMY OF C HILD & ADOLESCENT PSYCHIATRY VOLUME 52 NUMBER 11 NOVEMBER 2013
All statements expressed in this column are those of the authors and do not reflect the opinions of the Journal or the American Academy of Child and Adolescent Psychiatry. See the Instructions for Authors for information about the preparation and submission of Letters to the Editor.
www.jaacap.org
1215
