As psychoanalytical treatment methods are coming under increasing pressure to verify their effectiveness by means of empirical studies and systematic quality assurance measures, particular attention has been accorded recently to the measurement of constructs which are founded on psychoanalytical theory (Benecke, 2014). The construct of the structural impairment, as well as the concept of the mentalisation ability provide important indicators for the understanding and the treatment of structurally severely disturbed patients, for instance with borderline personality disorders. Despite the sparse evidence overall on the effectiveness of stationary psychodynamic therapy (Liebherz & Rabung, 2010), it already proved possible in two previous studies on the stationary therapy now being examined in the present study to verify its effectiveness in the symptomatic and interpersonal areas (Huber et al., 2009; Herrmann & Huber, 2013).
The intention is to examine the influence of stationary analytically oriented psychotherapy on the change in the symptomatology, mentalisation ability and the degree of the structural impairment. The mentalisation here includes constructs such as affect regulation, affect perception, self-reflection and persistence in an equivalence mode. The intention is to also identify potential differences with various diagnosis groups here.
The Kernberg psychoanalytical model for the personality organisation assesses the structural level in the three functional areas of identity diffusion, maturity of the defence mechanisms and reality testing. Here too, the intention is to examine whether differing impairments occur in various diagnosis groups in the three functional areas, and how these change.
The data basis for the study consisted of 898 patients who were in stationary psychoanalytically oriented treatment between March 2013 and March 2016 in the Clinic for Psychosomatic Medicine and Psychotherapy, Klinikum München-Harlaching Clinic, Munich. The primary diagnoses consisted of depressive disorders, angst, trauma-consequence, eating, somatoform and personality disorders. More than three-quarters of the patients exhibited comorbidities.
In order to measure the symptom severity, both prior to and after the stationary treatment the patients filled out the following subscales of the medical questionnaire for patients (PHQ-D; Gräfe et al., 2004; Löwe et a., 2002): PHQ9 (depressive symptomatology), PHQ15 (somatic symptomatology), GAD-7 (angst symptomatology; Löwe et al., 2008). In order to measure the impairment of basal mentalisation abilities, the MZQ mentalisation questionnaire (Hausberg et al., 2012) was used. The degree of the structural impairment as per Kernberg was measured with the IPO-16 questionnaire (Zimmermann et al., 2013).
In a first step, the statistical significance of the change was assessed by means of a T-Test for dependent samples. Differences between the diagnosis groups were tested with univariate ANOVA. The degree of the respective change was intended to show the effect sizes for all results. Any potential influencing factors were assessed by linear regression. The intention in a second step is to assess the rates of change by means of a latent-change-score model and/or a structural equation model. In this way, the aim is to reciprocally identify potential moderator or mediator influences on the change in the mentalisation ability and structural impairment, as well as on the change to the symptomatology.
With regard to the symptomatology, a highly significant therapy success and with major effect sizes was ascertained in all the criteria assessed. It proved possible to calculate the improvement in the depressive symptomatology at an effect size (Cohen, 1988) of ES = 1.51, the angst symptomatology at ES = 1.16 and the somatic symptomatology at ES = 0.70. An improvement in the mentalisation ability was determined at ES = 0.43 and an improvement in the structural impairment at ES = 0.25. The effect sizes at which the mentalisation abilities improved are to be classified as per Gignac and Szodorai (2016) as high, and the improvement in the structural impairment as medium to high.
As expected, patients with an emotionally instable personality disorder evidenced clearly higher values in structural impairments prior to the treatment compared to patients without this personality disorder, and that especially in the primitive defence subscale. With these patients, the mentalisation ability was also more severely impaired when compared to the remaining patients; the difference was especially great in the impaired affect regulation subscale.
The influencing factors on the measured therapy success included comorbidity and the treatment motivation, among others. With the therapy result deteriorating with an increasing number of diagnoses and/or comorbidities. Those patients assessed by the therapist as being barely motivated benefit less from the treatment.
Prof. Dr. Dr. Dorothea Huber
International Psychanalytic University
Stromstraße 3b, 10555 Berlin und
Clinic for Psychosomatic Medicine and Psychotherapy
Klinikum München-Harlaching Clinic, Munich
Joachim Frank, freelance employee at IPU