Attachment as a predictor of dropout in mentalization-based treatment.

Abstract

Although treatments of patients with borderline personality disorder (BPD) were historically associated with relatively high dropout rates, dropout rates in contemporary evidence-based treatments for BPD are typically substantially lower. However, only a few studies have investigated dropout rates in mentalization-based treatment (MBT), and even fewer have investigated predictors of dropout in this type of treatment. In this study, we investigated dropout rates in two types of MBT (day hospital MBT [MBT-DH] and intensive outpatient MBT [MBT-IOP]) using data from a recent multicenter randomized clinical trial. Given the central importance of attachment considerations in MBT, we also investigated the relationship between dropout in these two treatments and attachment dimensions. Within a multicenter randomized clinical trial, 114 BPD patients were randomized to MBT-DH (n = 70) or MBT-IOP (n = 44). Dropout in both types of MBT was investigated using descriptive analyses, and its association with attachment anxiety and attachment avoidance, as measured by the Experiences in Close Relationships questionnaire at baseline, was investigated using regression analyses. Dropout rates were relatively low (10.5% across both types of MBT) and did not significantly differ between groups (11.4% in MBT-DH, 9.1% in MBT-IOP). Attachment avoidance and attachment anxiety did not impact dropout, nor did their interaction or the interaction with the type of MBT. Low dropout rates in both types of MBT indicate a high level of engagement of patients in both programmes. Attachment dimensions were not associated with dropout, consistent with the principle that MBT is tailored to each individual's needs. More research is needed, however, to investigate to what extent attachment is a dynamic context-bound adaptive process rather than a static personality feature.