Patterns of inner change and their relation with patient characteristics and outcome in a psychoanalytic hospitalization-based treatment for personality disordered patients

Vermote, R., Lowyck, B., Luyten, P., Verhaest, Y., Vertommen, H., Vandeneede, B., Corveleyn, J., & Peuskens, J. (2011). Patterns of inner change and their relation with patient characteristics and outcome in a psychoanalytic hospitalization-based treatment for personality disordered patients. Clinical Psychology and Psychotherapy, 18 (4), 303-13.
Blatt, S. J. (2004). Experience of Depression: Theoretical, Clinical and Research Perspectives. Washington, DC: American Psychological Association.
Blatt, S. J. (2008). Polarities of Experiences: Relatedness and Self-Definition in Personality Development, Psychopathology and the Therapeutic Process. Washington, DC: American Psychological Association.

Brief Summary

The efficacy and effectiveness of psychodynamic treatments for personality disorders (PDs), and for borderline patients in particular, have been demonstrated in a number of randomized controlled trials as well as in naturalistic pre–post studies. In particular, these treatments have been shown to lead to clinically significant improvements in symptom levels, interpersonal functioning and global adjustment. Yet, the fundamental premise of psychoanalytic theories of PD is that besides improvement in symptoms, long-term psychodynamic treatment also results in changes in personality structure or organization. The concept of personality organization (PO) refers to the underlying organization of structural–dynamic components of personality and has been operationalized from different theoretical perspectives in terms of (a) the developmental level of representations of self and others; (b) the capacity for reality testing and maturity of defenses; (c) the background of safety or ‘felt security`; as well as, more recently, (d) the capacity for mentalization operationalized both in terms of reflective functioning and the Bion-Grid Scale. Although changes in PO are supposed to constitute a core mechanism of change in psychodynamic treatment for PD, today only a handful of studies have empirically investigated this assumption.

Based on this, the first aim of this study was to replicate and extend existing research in this area by simultaneously assessing changes in PO using four different measures reflecting different aspects of PO, namely (a) the developmental level of representations of self and others as measured with the DR-S; (b) mentalization as assessed by the Reflective Functioning Scale (RFS) and the GRID; and finally, (c) levels of felt safety as measured with the Felt Safety Scale (FSS). All scales are scored on the Object Relations Inventory (ORI). The second aim of this study was to investigate whether the 44 patients in a psychoanalytic hospitalization-based treatment for PD show different trajectories of change in PO and whether these different trajectories were associated with different pre-treatment characteristics. In particular, several studies have provided considerable evidence suggesting that anaclitic and introjective personality features (Blatt, 2004) are associated not only with different responses to treatments but also with different changes in terms of PO. In this context, Blatt (2004, 2008) has proposed that anaclitic patients are characterized by a distorted preoccupation with relationship issues, such as trust and intimacy, at the expense of self-development, as for instance expressed in dependent, histrionic and borderline personality features. Introjective patients are primarily preoccupied with intense and distorted attempts at establishing and maintaining a sense of self, including feelings of autonomy, self-control and self-worth, at the expense of developing interpersonal relationships, as in schizoid, schizotypic, paranoid, narcissistic, antisocial, avoidant, self-defeating and obsessive–compulsive PDs and features.

The last aim of this study was to investigate whether different clusters of patients were differentially related to outcome. Indeed, some studies have suggested that patients with PDs not only show different trajectories of change during treatment, but that these different trajectories are also associated with different outcomes. For instance, that introjective patients changed primarily in higher levels of thought disorder, whereas anaclitic patients changed more in the more pathological forms of thought disorder. To the best of our knowledge, this is the first study to directly investigate possible associations between different trajectories of change and outcome in the psychoanalytic treatment of PDs.

In line with findings of earlier studies, the results of this study showed that the psychoanalytic hospitalization-based treatment for PDs is indeed associated with changes in different aspects of PO, including representations of self and others, mentalization and felt safety. In addition, this study extends previous studies in that two different clusters of patients could be identified, which showed a distinct pattern of change in PO during and after treatment. On one hand, one cluster consisted of patients who showed a more fluctuating and thus less stable pattern of changes in PO during treatment and at follow-up. The other cluster consisted of patients showing more consistent improvement in PO both during and after treatment. Importantly, further analyses showed that these two clusters showed substantial differences in terms of pre-treatment personality features as assessed by the SCID-II. More specifically, results showed that the more fluctuating cluster seemed to be mainly characterized by anaclitic features as described by Blatt (2004, 2008), as was shown in higher scores on dependent, borderline, avoidant and depressive characteristics compared with patients from the stable cluster. Patients from the more stable cluster, in turn, tended to have higher scores on SCID-II narcissistic PD features, which is in line with Blatt’s suggestion that higher level personality disordered patients with introjective features are mainly characterized by narcissistic characteristics (Blatt, 2004, 2008).


This study including a longitudinal, multi-wave design as well as multi-trait, multi-method approach lends further support to the assumption that psychoanalytic hospitalization-based treatment of PD is associated with changes in PO. Finally, results of this study also suggest that different types of patients may differentially respond to different treatment factors, adding to the growing evidence for the importance of considering patient–treatment interactions in psycho-therapy research towards PO.


Prof. Rudi Vermote

University Psychiatric Centre, University of Leuven, Campus Kortenberg

Leuvense-steenweg 517, 3070 Kortenberg, Belgium