Statistics from Altmetric.com
Systematic review with meta-analysis.
MEDLINE, PsycINFO, and Current Contents between January 1970 and September 2004.
Study selection and analysis:
The review included randomised controlled trials (RCTs) of short term psychotherapy, delivered by trained therapists, targeted at the treatment of specific psychiatric disorders. The review did not include studies of interpersonal therapy. Within-group effect sizes were determined for each study using Cohen d statistic (baseline standard deviations were pooled where there was more than one patient group). Separate effect sizes were determined for targeted symptoms, general symptoms, and social functioning. Between-group effect sizes were determined by measuring the difference in magnitude of change between treatment and comparison groups in units of standard deviation. Comparisons between groups used t tests (for independent samples) and MANOVA (for differences between short and other forms of psychotherapy).
General symptoms (Symptom Checklist-90 and Beck Depression Inventory); targeted symptoms of disorders (tools specific to particular psychiatric disorders); and social functioning (Social Adjustment Scale).
Seventeen studies met criteria for inclusion in the review. Delivery of short term psychotherapy ranged from 7–40 sessions and mean follow up was about one year. Measures of effect size from pretreatment levels showed that short term psychotherapy improved targeted symptoms (17 studies, p<0.01 for post-treatment effect size v zero), general symptoms (15 studies, p<0.01 for post-treatment effect size v zero), and social functioning (11 studies, p<0.01 for post-treatment effect size v zero). Between-group comparisons showed that short term psychotherapy was more effective than no treatment or treatment as usual (p<0.01 for comparison of post-treatment effect sizes for target and general symptoms and social functioning) but there was no difference between short term and other forms of psychotherapy (15 studies, p = 0.69 for comparison of post-treatment effect sizes for target and general and social functioning). Comparisons between pretreatment effect sizes and those at study follow up did not differ significantly from pre- and post-treatment scores.
Short term psychodynamic psychotherapy is an effective treatment for psychiatric disorders though there is a need for further research into effectiveness for specific disorders.
Studies included in the review used different models of short term psychotherapy, of varying duration and follow up. The authors noted, however, that effect sizes of short term psychotherapy were not correlated with therapy duration. Due to the small number of studies assessing outcomes in people who had no treatment, and also in those who were treated as usual, results for these studies were combined.
This paper addresses a controversial issue: is short term psychodynamic psychotherapy (STPP) effective? The controversy arises from the fact that cognitive behaviour therapy (CBT) and related psychological therapies are seen to have eclipsed psychodynamic therapies. It is a rigorous systematic review using established criteria for the identification and assessment of papers. However, the analysis of the data is not as strong. The major problem is the reliance on within-group comparisons to calculate effects sizes, on which rests the major conclusion that STPP is “an effective treatment in psychiatric disorders”. Baseline and end point or follow up scores are used, rather than the more robust relative comparisons with either treatment as usual/waiting list (TAU/WL) or other active interventions. To describe such effect sizes (which are large—1.57 for target problems) as providing “strong evidence” for short term psychotherapy seems inappropriate. Effect sizes comparing STTP with either TAU/WL (0.94) or other active interventions (0.23) suggest more modest conclusions. There are two reasons for this. First, the effect sizes are smaller both in absolute terms and in comparison to effect sizes reported elsewhere—for example, in the UK, the National Institute for Health and Clinical Excellence reported effects of 1.7 for waiting list versus trauma focused psychological therapies and 1.18 for other active interventions for PTSD.1 Second, there are problems with the small number of trials in key areas (for example, TAU/WL n = 5), with the combining of different outcomes from the same study in the meta-analysis and the failure to consider clinical as opposed to statistical significance.
There are also problems with the span of the review—of the 17 trials covered only three relate to anxiety and depression, the most common disorders. Other disorders covered included personality disorder, eating disorders, and substance misuse. This approach goes against much contemporary thinking about the efficacy of psychological therapies and the general trend which has suggested that specifically targeted psychotherapies are more effective.2
Will this alter my clinical practice such that STTP takes an equal place, for example, to CBT and interpersonal psychotherapy (IPT) in the treatment of depression? No. There is a substantial body of research in depression which supports the use of CBT and IPT. That is not the case for STPP but Leichsenring et al do suggest that STPP may be useful as a second line treatment. However, more research is required before STPP is adopted as a first line treatment for the disorders covered in their review.
For correspondence: Falk Leichsenring, DSc, Department of Psychosomatics and Psychotherapy, University of Goettingen, von Siebold Str 5, 37075 Goettingen, Germany;
Sources of funding: not reported.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.