Article Text
PDF
Statistics from Altmetric.com
QUESTION
Question:
How effective is long term psychodynamic psychotherapy (LTPP) in people with complex mental disorders?
Outcomes:
Overall outcome, target problems, general psychiatric symptoms, personality and social functioning (see notes).
METHODS
Design:
Systematic review with meta-analysis.
Data sources:
MEDLINE, PsycINFO and Current Contents were searched from 1960 to May 2008 for randomised controlled trials (RCTs) and observational studies. Reference lists of identified articles and textbooks were hand searched, and authors and experts contacted for further information.
Study selection and analysis:
Two reviewers appraised the studies and selected RCTs and observational studies of LTPP. Inclusion criteria: studies in adults; ⩾50 sessions or 1 year of LTTP; reliable outcome measures used; mental disorders of participants clearly described; and sufficient data to determine effect sizes. Studies with concomitant treatments were included but analysed separately to compare results of combined treatments versus LTTP alone. Two reviewers independently rated study quality, extracted data and assessed effect sizes. Point biserial correlation coefficients (rp) were used as measures of between and within group effect sizes, and to test for differences in effect sizes between RCTs and observational studies. Subgroup analyses were carried out for personality disorders, multiple mental disorders, chronic mental disorders and complex depression and anxiety disorders.
MAIN RESULTS
Eleven RCTs and 12 observational studies (1053 participants) met the inclusion criteria. Data from RCTs and observational studies were combined as the point biserial correlations revealed that the observational studies did not produce within group effect sizes for LTPP significantly different from RCTs. LTTP produced significantly larger pretreatment–post-treatment effect sizes for overall effectiveness, personality functioning and target problems compared with other psychotherapy methods (CBT, cognitive analytic therapy, family therapy, supportive therapy, short term psychodynamic therapy, dialectical–behavioural therapy and psychiatric treatment as usual; eight studies; see online table). There was no significant difference in effect size between groups for social functioning or symptoms. In people with complex mental disorders (personality disorders, multiple mental disorders or chronic mental disorders; seven studies), LTTP produced significantly larger pretreatment–post-treatment effect sizes compared with other psychotherapy methods in overall effectiveness (equivalent to Cohen’s d of 1.8, 95% CI 0.7 to 3.4), personality functioning (Cohen’s d of 6.9, 95% CI 3.0 to 14.6) and target problems (Cohen’s d of 1.9, 95% CI 0.7 to 3.5). Converting these results into percentiles indicated that LTTP treated participants had, on average, a better overall outcome than 96% of participants in the comparison groups.
CONCLUSIONS
LTTP improves overall outcome, target problems and personality functioning, but not social functioning or general psychiatric symptoms, in people with complex mental disorders compared with other psychotherapies.
ABSTRACTED FROM
Leichsenring F, Rabung S. Effectiveness of long-term psychodynamic psychotherapy—A meta-analysis. JAMA 2008;300:1551–65.
Commentary
Long term psychodynamic psychotherapy (LTPP) has been used widely in the treatment of mental disorders. Evidence of its effectiveness is, however, limited. This meta-analysis is thus both a valuable contribution to and a reminder of the urgent lack of information on the significance of LTPP from a scientific, clinical and economic perspective. The study suggested that LTPP is an effective treatment for complex mental disorders in comparison with less intensive or less structured therapies. This effect difference was stronger than that between LTPP and two short term therapies on patients with lesser disorder complexity,1 thus underlining the apparent superiority of LTPP in the treatment of complex mental disorders.
The observational study showed an improvement during and after treatment but the inherent design dependent restrictions allow no firm conclusions about causality. The meta-analysis based on seven trials showed LTPP to be more effective than other lengthy treatments. Despite a carefully conducted meta-analysis and clinical justifications for using LTPP in patient populations with complex disorders, the findings cannot be generalised, especially in the case of long term versus short term comparisons. The paucity of the original studies generally based on small number of patients, ignorance of the different precisions of the effect size estimates in them, lack of short term therapy comparisons and heterogeneous diagnoses and types of therapy restrict reliable conclusions.
The present study definitely suggests that LTPP is an effective therapy but we are still far from evidence based selection criteria for short versus long term psychotherapies. Additional evidence, based on large amounts of data covering different psychiatric disorders and therapies of different modalities and length, are needed. The design of such studies should allow for an estimation of the effect of the therapy form, intensity and duration simultaneously with the effect modification caused by patient, therapist and alliance factors. The follow-up times should enable the recognition of enduring benefits or relapses and calculation of cost effectiveness in the long run. Joint analyses simultaneously carried out in different populations based on identical protocols would be one path along which to proceed.
Supplementary materials
Web only appendices 12;2:56
Files in this Data Supplement:
Footnotes
Source of funding: Not reported.
▸ Additional notes, a table and a reference list are published online only at http://ebmh.bmj.com/content/vol12/issue2