In this study we present a clinical application of the STEPPS model in an Italian sample of severely affected patients with borderline personality disorder (BPD) or personality disorder (PD) with prominent borderline features in comorbidity with a mood disorder. The aims of this work are: 1) to confirm our preliminary results in a larger sample and at a 12-month follow-up, and 2) to identify predictors of drop-out vs completion of STEPPS in order to understand which characteristics of patients make them suitable or not for this treatment. The sample is composed of 32 subjects recruited from a population of inpatients of the Mood Disorders Center, Department of Clinical Neurosciences, Hospital San Raffaele-Turro, Milan. To confirm STEPPS efficacy at 12-month follow-up, we selected the following outcome criteria: reduction in the number of hospitalizations related to self-harm acts; reduction in the number of suicidal attempts; reduction of perceived emotional intensity levels; changes in cognitive filter scores; changes in the scores on self-report questionnaires. To identify predictors of drop-out vs completion, we analysed the following variables: demographic features (sex, marital status, school level achieved, and job status); Axis-I diagnosis; Axis-II categorical and dimensional diagnosis; and personality features. Seventeen (53%) subjects completed the treatment successfully. The drop-out rate was 47%. Patients who completed the program show a significant decrease in the number of hospitalizations, both at the end of the treatment and at 12-month follow-up. Friedman ANOVA test shows a significant decrease in suicidal attempts during and after STEPPS, and at 12-month follow-up. Analysis of drop-outs showed no significant differences with regard to sex, marital status, school level and job status between the two groups. Axis-I and Axis-II categorical diagnoses did not discriminate between the two groups. Those patients who dropped differ significantly from completers in histrionic personality traits and magical thinking index, given by the interaction between low scores in Self-Directedness and high scores in Self-Transcendence.
Borderline personality disorder (BPD) is a debilitating disorder that occurs in approximately 1%–3% of the general population [1] and [2]. It is characterized by emotional instability, identity disturbance, interpersonal dysfunction and impulsivity [3]. Borderline patients often engage in self-injurious and suicidal behavior, substance or alcohol abuse, gambling, compulsive shopping, binge eating and reckless driving. So they often have high rates of co-occurring disorders, like mood and anxiety disorders, substance abuse and eating disorders [4], [5] and [6]. All these aspects may lead to psychiatric hospitalization and/or incarceration: the rate of BPD in psychiatric settings is approximately 20% and the rate in incarceration settings is even higher [1] and [3].
Despite a current debate about how and how much BPD can be treated effectively and which treatment is optimal [7], in the last few years several psychotherapeutic models have been developed: Dialectical Behavior Therapy (DBT) [8], Mentalization-Based Treatment (MBT) [9], Transference-Focused Psychotherapy (TFP) [10], Schema-Focused Therapy (SFT) [11] and Systems Training for Emotional Predictability and Problem Solving (STEPPS) [12].
Most of the treatments are manualized and their efficacy is supported by clinical and empirical data. All of these treatments share some common features: a stable setting, a clear and well-defined therapeutic focus, promoting patient compliance, focusing on the therapeutic relationship, and assigning an active role to the therapist. There are no empirical data to support the primacy of one of these models [13] and [14], but it is possible that in the future, different treatments may prove to be effective for specific aspects of BPD. It may be important to understand the specific therapeutic action underlying each treatment, which could help clinicians determine the best approach for a specific patient.
Despite the development of these evidence-based treatments, the drop-out rate for borderline patients in psychotherapy is still high [15], [16], [17] and [18]. Random effects meta-analyses yielded an overall completion rate of 75% for interventions of <12 months duration and 71% for longer treatments [19]. There are little data on predictors of which patients are likely to end treatment prematurely. Therefore it may be important to evaluate what factors are associated with completion vs drop-out in order to understand which patients are more suitable for each treatment. Previous studies of clinical and demographic variables have shown that factors such as high baseline psychopathology, length of therapy [20] and [21], young age [22] and [23], high levels of hostility and anxiety [24], and anger and impulsivity [25] and [26] may be important predictors of drop-out.
In this study we present a clinical application of the STEPPS model in an Italian sample of severely affected patients with BPD or PD with prominent borderline features in comorbidity with a mood disorder. In a previous study [27] we showed preliminary efficacy data for the Italian STEPPS program; however, the drop-out rate was high. The objectives of the current study were to: 1) confirm our results in a larger sample and at a 12-month follow-up, and 2) identify predictors of drop-out vs completion in order to understand which patient characteristics make them suitable or not for this treatmen
