Analyzing Effectiveness of Long-term Psychodynamic Psychotherapy—Reply
- Falk Leichsenring, DSc falk.leichsenring@psycho.med.uni-giessen.deDepartment of Psychosomatics and PsychotherapyUniversity of GiessenGiessen, Germany;
- Sven Rabung, PhDDepartment of Medical PsychologyUniversity Medical Center Hamburg-EppendorfHamburg, Germany
Since this article does not have an abstract, we have provided the first 150 words of the full text.
- KEYWORDS:
- DATA INTERPRETATION, STATISTICAL
- META-ANALYSIS AS TOPIC
- PSYCHOTHERAPY
- PUBLICATION BIAS
- RANDOMIZED CONTROLLED TRIALS AS TOPIC
- TIME FACTORS
In Reply: In response to Dr Thombs and colleagues, because the number of controlled trials of LTPP was relatively small, we assessed within-group effect sizes in all studies. In the controlled studies, we found significantly larger within-group effect sizes in the LTPP conditions than in the control conditions. To assess the extent of this difference, we transformed the point biserial correlations used for tests of significance into between-group effect sizes in the form of a d statistic.1
These between-group effect sizes are not identical to between-group effect sizes as commonly assessed by calculating the difference between 2 treatments for each study. The between-group effect size we assessed may be larger than the underlying within-group effect sizes because they refer to different units. Our comparison of LTPP with other treatments considers treatment group rather than patient as the unit of analysis. A between-group effect size of 1.8 indicates that in …
