Summary of the main findings
Chapter 2 described the study protocol of our comparative multiple case study investigating the feasibility and preliminary results of a CCP for patients with severe personality disorders.

Chapter 3 elucidated the content of CCP in more detail. The CCP was elaborated in a manual for professionals and patients. It consists of several aligned structured interventions divided in three stages: preparation, treatment and evaluation. The preparation stage includes seven activities in order to provide a treatment frame. These activities are: introduction of CCP, forming of a Collaborative Care team, making a time line concerning the treatment history of the patient, explication of collaboration agreements, drafting a crisis card, need assessment and establishing the treatment plan. The treatment stage consisted of four main components: early recognition and intervention of problem behaviours, problem solving treatment, life orientation and psycho-education. The goals, as described in the treatment plan, are evaluated every three months within the Collaborative Care team.

In Chapter 4 we presented the preliminary results of the CCP compared to Care as Usual and investigated the factors within the CCP which were indicative for positive or negative results. We found a significant decrease in borderline symptoms in the CCP-group compared to Care as Usual. Several other process indicators showed clinically relevant effect sizes in favour of CCP: ‘satisfaction with care’ and ‘quality of the therapeutic relationship’ among patients; ‘satisfaction with care’ and ‘perceived burden’ among informal carers; and ‘quality of the therapeutic relationship’ among nurses. Three core elements of the CCP contributed to the positive results: 1) improved goal orientation in the treatment process; 2) a stronger appeal to patients’ self-management skills; and 3) improved skills in establishing and maintaining effective therapeutic relationships. Although far from 100% successfully implemented, our data suggest that not only patients, but also their informal carers and the nurses involved in the treatment benefited from CCP.

In Chapter 5 we analyzed the process of execution of CCP, and identified hampering and fostering factors in this process. In 57% of the cases CCP was moderately to well applied as opposed to 43% of the cases in which CCP was poorly carried out. The execution of CCP was most successful in the preparation stage. Four interdependent factors were identified explaining the process of application. Factors were related to: 1) the context in which CCP was executed; 2) the patient population; 3) the CCP itself and 4) the individual application of CCP by the nurses. The key to successful execution appeared to be the individualized application of CCP by the nurses. At the same time, this step proved to be the most complex due to a more general unfamiliarity with working according to a protocol and problems in adjusting this specific protocol to the individual patient, poor agenda setting and the avoidance of core problems, e.g. discussing suicidal behaviour or lack of progress in treatment. In conclusion, although challenging, effective execution of CCP was achieved in a part of the nurses and preliminary results of CCP are encouraging. This indicates that CCP is feasible and might be beneficial to patients, their informal carers and nurses.

Chapter 6 described the perceived needs of care and health care utilization of persons with suicidal ideation compared to those without suicidal ideation in a large sample of patients with current depressive or anxiety disorders, derived from the NESDA study. We found that persons with suicidal ideation had higher odds for perceived unmet and met needs than persons without suicidal ideation. We also found that persons with suicidal ideation had more intensive contact with mental health care providers than persons without suicidal ideation. Our data also clearly showed that differences in perceived needs and health care utilization were largely explained by severity of the axis-I symptomatology.

In Chapter 7 we studied the role of comorbid BPD traits in relation to recurrent suicide attempts in a large sample of patients with depression and/or anxiety disorders. In line with our expectations, borderline personality traits were strongly associated with recurrent suicide attempts within this large cohort of persons with lifetime depressive and anxiety disorders. Borderline traits become progressively more important in patients with increasing numbers of suicide attempts. Of the different BPD traits, particularly internalizing anger problems were significantly and independently associated with recurrent suicide attempts. The other characteristic that showed a significant and independent association with recurrent suicide attempts was a lifetime diagnosis of dysthymia.