Mentalization is the capacity to understand behavior as expressions of various mental states. It is assumed to be important for understanding the underlying psychopathology, the therapeutic process, and the outcome of therapy associated with patients with personality disorders (PDs). However, to date, empirical findings are scarce and inconsistent. This study aimed to examine whether the pre-treatment level of mentalization, operationalized as Reflective Functioning (RF), was associated with differential responses to two different treatment modalities and might predict clinical improvement. We analyzed data from a randomized clinical trial (Ullevål Personality Project). Seventy-eight patients with borderline and/or avoidant PD had been randomly assigned to either a step-down treatment program or outpatient individual psychotherapy. The step-down treatment comprised short-term day hospital treatment, followed by long-term, combined group and individual psychotherapy. RF was rated before treatment and after 36 months. Outcome measures were administered at baseline and after 8, 18, and 36 months. The moderator analyses indicated that patients with low RF levels at baseline responded better to outpatient individual psychotherapy than to the step-down treatment in terms of improvements in psychosocial functioning. Patients with medium-high RF levels responded equally well to both therapy formats. Determining which therapy format is appropriate for specific groups of patients can improve treatment efficiency. Therefore, our findings may have important clinical implications. Future research should address RF as a mediator of change.

Despite increasing interest in the development of effective treatments for patients with PDs, there is still no consensus about the optimal treatment setting for this group of patients. This study reports the 36 months follow-up of the Ullevål Personality Project (UPP) (n=113), a randomized clinical trial comparing two treatment modalities for patients with PDs: an intensive long-term step-down treatment program, consisting of short-term day hospital treatment followed by combined group and individual psychotherapy organized in a hospital setting, with "ordinary" outpatient individual psychotherapy in private practice for patients with moderate to severe PDs. Patients in both treatment groups showed improvements in several clinical measures after 36 months. However, contrary to our expectations, patients in the outpatient treatment setting improved significantly more. Possible explanations for this surprising finding are discussed. The study cannot exclude the possibility that treatment aspects other than differences in modalities could explain some of the differential effectiveness (e.g. differences between therapists).

It has been suggested that deficits in mentalization are a core problem of borderline personality disorder and that enhancement of mentalization represents a central mechanism of change in psychotherapy with this group of patients. The present single-case study investigates changes in mentalization measured by the Reflective Functioning Scale on the Adult Attachment Interview and its relationship with clinical measures of psychopathology during a long-term psychotherapy. The usefulness of evaluating RF along three subdimensions, the object, content and process of RF, is underlined. Furthermore, it is argued that there is a complex interaction between RF and symptomatology that becomes evident when evaluating RF along the different subdimensions. Areas for future research clarifying the relationship between mentalization, psychotherapy and psychopathology are suggested.