[A favourite adversary: "classic long-term psychoanalysis". Commentary on Rief and Hofmann's "Psychoanalysis should be rescued. By all means?"].

Rief and Hofmann (2009, Nervenarzt 80:593-597) harshly criticise the meta-analysis on the effectiveness of long-term psychodynamic psychotherapy (LTPP) by Leichsenring and Rabung (2008, JAMA 300(13):1551-1565). They find fault with the inclusion of naturalistic studies in addition to randomised clinical trials. Furthermore, they criticise the heterogeneity of the treatments included and the disorders studied. They suspect that a number of RCTs of LTPP with negative results for LTPP have been done and not been published. This paper comments on the following issues: the strict determination of RCTs is scientifically outdated and in order to investigate the effectiveness of psychotherapy naturalistic studies have to be included; the heterogeneity of studies included in meta-analysis as well as the heterogeneity of the patients studied reflect clinical reality, which is the purpose of effectiveness studies. The accusation of repressing results of LTPP RCTs is unsustainable. All in all, the meta-analysis by Leichsenring and Rabung was done accurately, and the results were controlled for by separate analyses of single subgroups. Therefore, their study does provide evidence of the effectiveness of long-term psychodynamic psychotherapy for patients with complex mental disorders.