Evaluation of inpatient Dialectical-Behavioral Therapy for Borderline Personality Disorder — a prospective study

Abstract

Dialectical-Behavioral Therapy for Borderline Personality Disorder (DBT) developed by M. Linehan is specifically designed for the outpatient treatment of chronically suicidal patients with borderline personality disorder. Research on DBT therapy, its course and its results has focused to date on treatments in an outpatient setting.

Hypothesizing that the course of therapy could be accelerated and improved by an inpatient setting at the beginning of outpatient DBT, we developed a treatment program of inpatient therapy for this patient group according to the guidelines of DBT. It consists of a three-month inpatient treatment prior to long-term outpatient therapy. In this pilot study 24 female patients were compared at admission to the hospital, and at one month after discharge with respect to psychopathology and frequency of self-injuries. Significant improvements in ratings of depression, dissociation, anxiety and global stress were found. A highly significant decrease in the number of parasuicidal acts was also reported. Analysis of the average effect sizes shows a strong effect which prompts the development of a randomized controlled design.

Introduction

DBT is a cognitive-behavioral therapy, developed by M. Linehan originally for the outpatient treatment of chronically suicidal patients with borderline personality disorder. As with standard behavior therapies, DBT presumes that attention to both skills acquisition and behavioral motivation is essential for change. Taking into account the characteristic features of patients with borderline personality disorder, several modifications to standard behavioral therapy were made (Linehan, 1993). First, a number of treatment strategies that reflect acceptance and validation of the patients' current capacities and behavioral functioning were gathered and added to the treatment. The dialectical emphasis of the treatment ensures the balance of acceptance and change within the treatment as a whole and within each individual interaction. Second, treatment of the patient was split into three components: one that focuses primarily on skill acquisition, one that focuses primarily on motivational issues and skills strengthening, and one designed explicitly to foster generalization of skills to everyday life outside the treatment context. Third, a consultation-team-meeting with specific guidelines for keeping the therapist within the treatment frame was added. In standard outpatient DBT, treatment consists of structured psychosocial individual or group therapy (for skills training), individual psychotherapy (addressing motivational and skills strengthening), telephone contact with the individual therapist (addressing generalization), and peer supervision meetings (to monitor the therapist). DBT is further characterized by a clear hierarchy of treatment targets (the behavior identified for change), and a set of treatment strategy groups (tactics and procedures of the therapist used to achieve change). In contrast to many behavioral approaches, DBT also places great emphasis on the therapeutic relationship.

Reliable data are available for an outpatient treatment period of one year. During this period and in the framework of a controlled randomized study, DBT proved to be superior with regard to several factors compared to experienced therapists following an unspecified ‘treatment as usual‘ approach. Frequency and severity of parasuicidal acts were significantly reduced in the group of patients treated according to DBT; the same is true for the frequency of premature treatment termination, as well as for the frequency and length of stays in psychiatric hospitals (Linehan et al., 1991, Linehan et al., 1993). Meanwhile, data from replication studies are available (Koons, 1998).

As discussed above, DBT was originally developed as a form of outpatient therapy and emphasizes the potential risk of nonspecific inpatient treatment. One of the main risk factors seems to be the (unintended) reinforcement of dysfunctional patterns of behavior such as self-injury, suicide attempts, and/or suicide communications by the therapeutic milieu. This notion is similar to the views of numerous depth psychology-oriented authors who particularly emphasize the distinct tendency towards ‘regression’ on the part of borderline inpatients, as well as ‘manipulatory behavior’ and difficult transference and counter-transference phenomena. A deterioration of the symptoms and a tendency towards long-term hospitalization are the most frequent results (Nurnberg and Suh, 1978, Rosenbluth et al., 1992).

Nonetheless, several reasons speak for the development of a specific module of inpatient treatment according to the guidelines of DBT. First, the number of patients who meet the criteria for borderline personality disorder is estimated at 30% of all inpatients worldwide who are treated for personality disorders, thus ranking in first place (Loranger et al., 1994). Second, the probability of requiring psychiatric or psychosomatic inpatient treatment at some point in one's lifetime is unusually great for persons with a borderline disorder. We studied a representative population of 40 female patients in Germany with borderline personality disorder according to DSM-IV and DIB-R (Gunderson, Kolb & Austin, 1981) and discovered this to be the case for 84% of these patients. Following initial hospitalization, 80% of them returned annually for an average annual inpatient stay of 70 days (Jerschke, Meixner, Richter & Bohus, 1998). On the basis of these findings, inpatient stays seem to increase the probability of readmission or at least they do not contribute substantially to outpatient stabilization. A third argument does not rely on data but on the clinical experience that in an unstructured inpatient setting both therapists as well as other caregivers are often unable to cope with borderline patients and that the nonprofessional handling of parasuicidality or states of dissociation often contributes to an aggravation of the dysfunctional behaviors. This frequently demands a high level of energy and time in conjunction with team meetings, consultations or supervision — to the disadvantage of other patients. And finally, despite the empirical superiority of DBT to nonspecific types of treatment, its therapy results are not optimal. Vital psychopathological parameters such as depression and anxiety, for example, were still high at the end of one year of treatment. It should be possible then to harness the synergy potential of a multidisciplinary team of inpatient therapists so as to increase the efficacy of DBT.

In developing our concept we referred to the experiences at New York Hospital, White Plains, NY, where Ch. Swenson and S. Sanderson in collaboration with M. Linehan had developed an inpatient treatment program according to the guidelines of DBT. But it was Barley and colleagues who have meanwhile presented initial findings (Barley et al., 1993). They compared the average frequency of self-injuries and overdoses occurring monthly on the ward by borderline patients while receiving treatment according to one of two treatment conditions: a treatment approach based on depth psychology versus a treatment approach on the same ward during a subsequent time period following a restructuring of the ward concept according to DBT. With the introduction of DBT, parasuicidal acts decreased significantly. By comparison, on a general psychotherapy ward, where during the comparable time period therapy was not carried out along these guidelines, no changes occurred. K. Silk, who developed a short-term inpatient therapy module according to DBT, reported above all a high level of acceptance of the program on the part of caregivers, who felt more competent and better enabled to cope (Silk et al., 1994).

The basic concepts of treatment, as we have established them at the University Hospital for Psychiatry and Psychomatics in Freiburg, are described elsewhere (Bohus, Swenson, Sender, Kern & Berger, 1996). In summary, the three-month period of treatment can be divided into three stages. Diagnostics, including information as to the nature of the disorder according to DBT, as well as the treatment goals and strategies derived therefrom are clarified at the start.

The first stage of therapy covers approximately 3 weeks. An analysis of the targeted behavior, its antecedents and consequences takes place (at the highest possible resolution). The focus is particularly on the problem behavior responsible for the current hospitalization as an inpatient, as well on behavior that prevents any current outpatient treatment. Thereafter the therapy planning takes place together with the patient and the entire therapeutic team. Specifically oriented to the individual problem and the resources of the patient, members of the multiprofessional team develop the treatment targets that will guide the rest of the treatment, ordinarily the acquisition and strengthening of the patient's capacity to regulate tension as well as emotion in the face of real psychosocial conflicts. The second stage of therapy encompasses the following goals:

• Theoretical training of the patient targets the greatest possible understanding of the disorder. This implies both the acquisition of knowledge concerning the consequences of possible traumatization as well as the basics of emotion regulation, learning theory, and the effective mechanisms of psychotherapy (the patient should become a specialist in regard to her disorder).

• Acquisition of specific skills for addressing those problems which led to hospital admission. E.g. stress tolerance, emotion modulation or development of self-management of dissociation or flashbacks.

• Contingency management of reinforcers following self-injurous behavior or suicidal communication.

Focus of the third stage of therapy is discharge planning. This means not only the preparation for release from hospital but also the actual establishment of contact with the therapist responsible for continuing treatment, exposure to job stress or being alone. Support from social workers is important during this stage.

Approximately one year after beginning of restructuring the ward, the developmental phase was complete insofar as articulation of the basic treatment targets and the specific strategies and techniques to be used by members of the multiprofessional team were concerned. The treatment had been rated positive by both the therapeutic staff and the patients. At this point we do not have data addressing whether this inpatient module provided as a preparatory element prior to commencement of outpatient dialectic-behavior therapy actually has any positive effect on patient course and outcome. This can only be resolved within the framework of a controlled ‘add-on design’ (Bergin & Garfield, 1994). This would mean randomization into two groups, one starting immediately with outpatient DBT, while the other group starting first with inpatient therapy, followed by an outpatient therapy. In order to justify the expense of such a study, it is necessary to first test the feasibility, safety, and effect of the inpatient treatment module in question. We conducted a pre–post comparison of treatment outcomes to address these questions.