Optimierung der Effekte von Psychotherapie: Wirksamkeit, Prädiktoren und Nebenwirkungen in der stationären Routineversorgung

In den letzten Jahrzehnten hat sich durch klinische Studien mit methodisch fundierten Designs eine breite Evidenzbasis von Psychotherapie für verschiedene psychische Störungen gebildet. Durch verschiedene randomisiert-kontrollierte Studien (engl.: „randomized controlled trial“; RCT), die in Meta-Ana...

Full description

Saved in:
Bibliographic Details
Main Author: Herzog, Philipp
Contributors: Brakemeier, Eva-Lotta (Prof. Dr.) (Thesis advisor)
Format: Doctoral Thesis
Published: Philipps-Universität Marburg 2020
Online Access:PDF Full Text
Tags: Add Tag
No Tags, Be the first to tag this record!

Over the last decades, clinical studies with methodologically sound designs have created a broad evidence base of psychotherapy for various mental disorders. Numerous randomized controlled trials (RCT), which are subsequently summarized in meta-analyses, have demonstrated the efficacy of psychotherapy with mostly large effects. According to the phase model of psychotherapy research, the last phase - phase IV - investigates psychotherapy in a naturalistic context, i.e. under everyday routine conditions. Within the framework of a practice-oriented research paradigm, such practice-based studies can complement RCTs to address existing problems in psychotherapy research (e.g. research-practice gap, stagnation of effects). In Germany in particular, patients with severe mental disorders are often treated in an inpatient setting (e.g. in psychosomatic or psychiatric clinics). In this naturalistic context, there is a lack of studies that investigate the clinical effectiveness with lower inclusion and exclusion criteria. Furthermore, the findings on variables that predict treatment outcome are often heterogeneous. Therefore, the objectives of this dissertation are (1) to examine the effectiveness of inpatient psychotherapy for various mental disorders (borderline personality disorder, post-traumatic stress disorder, obsessive-compulsive disorder, depression) in German psychosomatic clinics, and (2) to identify starting points for optimizing these effects by analyzing predictors of outcome. Overall, studies 1-4 indicate that specific evidence-based psychotherapy programs have been effectively implemented in inpatient psychosomatic care according to the respective national treatment guidelines, but indicate differences in their effectiveness expressed by different effect sizes. While the effect sizes for the treatment of obsessive-compulsive disorders (OCD; N = 1,595 and N = 514, respectively) and depression (N = 22,681 and N = 6,377, respectively) can be classified as large, there are medium to large effects for the treatment of post-traumatic stress disorders (PTSD; N = 612) and small to medium effects for the treatment of borderline personality disorders (BPD; N = 878 and N = 703, respectively). Due to large sample sizes, the effects might serve as a benchmark and therefore can be used as a comparison in future studies. Moreover, the present dissertation aims to identify general predictors for treatment outcome of psychotherapy considering the above-mentioned mental disorders. In the sense of prognostic variables, patient characteristics were determined - separately for the various mental disorders - within the framework of clinical prediction models that indicate which patient benefit more vs. less from the treatment. In patients suffering from BPD, the most important predictors for treatment completion are a higher education and a comorbid recurrent depressive disorder, and low affect regulation skills and no previous outpatient psychotherapy for symptom-specific changes. In patients with PTSD, a higher degree of psychoticism, total number of diagnoses, and a diagnosis of bronchial asthma consistently show a stable negative predictive relationship to treatment outcome. In patients with OCD, symptom severity at admission, general psychopathological distress, higher social support and more washing compulsions are important predictors of symptom changes. In patients suffering from depression, almost all sociodemographic and clinical pretreatment factors significantly influence the overall psychopathological burden on various baseline factors, but this direct influence weakens over the course of treatment. Several factors are resistant to change and indicate stable psychopathological factors during treatment, such as suicidality, agoraphobia, life dissatisfaction, physical disability and pain. The strongest impact is exerted by suicidality at the beginning of treatment on negative cognitions at the end of treatment, by agoraphobia at the beginning of treatment on anxiety at the end of treatment and by physical disability at the beginning of treatment on perceived disability at the end of treatment. Ultimately, the other side of the coin is also considered: in addition to positive effects, psychotherapy can also produce negative effects. Within the framework of a systematic review, study 5 evaluates instruments for assessing negative effects in the light of psychometric properties, proposes a consual definition and conceptualization of negative effects and derives recommendations for improving the assessment. This systematic review shows that although the existing instruments cover many relevant domains, their psychometric properties are usually unsatisfactory. Subsequently, the implications of this review are used to investigate the relationship between the occurrence of negative effects and treatment outcome during an inpatient CBASP program for chronic depression as described in study 6. It is shown that although almost all patients report having experienced at least one negative effect, only a perceived dependence on the therapist is negatively related to treatment outcome. Overall, it can be stated that the results from phase III studies of efficacy research can be transferred to an applied inpatient setting. With regard to studies 1-4, some predictors (e.g. symptom severity at admission) could be replicated to previous findings of RCTs; however, other variables also play an important role in predicting treatment outcome in these samples of inpatients, which can be attributed to different reasons, such as the specifics of the setting (e.g. higher treatment dose in a shorter period of time) and the heterogeneity of the samples. In studies 5-6, both definitional aspects of negative effects could be addressed in order to further improve their detection, and the relevance of specific negative effects could be determined in more detail by establishing the relationship to treatment outcome. In summary, all studies can provide starting points for optimizing the effects of inpatient psychotherapy in Germany. For example, studies 1-4 in particular provide benchmarks for the absolute effectiveness of psychotherapy in routine inpatient care as well as practice-based evidence through the predictor analyses, which in turn can be taken into account in future RCTs, e.g. in randomisation procedures. By feedback loops of the results of studies 1-6 to clinical practitioners, an evidence-based outcome monitoring can take place in the short term and, in the long term, the practice-research network in Germany can be strengthened.