5 - (IOP 17) Exploring Changes in Insecure Attachment and Borderline Personality Disorder Symptoms During Treatment with BPD Compass: A Network Analysis

Background: Attachment theory posits that significant others (e.g., caregivers, friends, partners) influence our emotions, thoughts, and interpersonal behaviors (Mikulincer et al., 2003). Insecure attachment develops when close others respond inconsistently and is defined by two dimensions: attachment anxiety (i.e., excessive fear of rejection and need for approval), and attachment avoidance (i.e., excessive fear of and reluctance to intimacy), both of which are strongly associated with borderline personality disorder (BPD; Smith & South, 2020). BPD Compass is a cognitive-behavioral therapy (CBT) which targets attachment insecurity as a factor maintaining BPD symptoms by fostering perspective-taking skills and challenging negative beliefs about trust; however, its effectiveness at modifying insecure attachment is unclear. The present study applied network analysis to explore how insecure attachment changed during treatment with BPD Compass, and how changes in attachment related to changes in BPD symptoms.

Methods: Participants (N = 100; Mage = 28.13, 74% female, 78% White) were treatment-seeking adults diagnosed with BPD who received BPD Compass (Sauer-Zavala et al., 2023), an 18-session CBT, as part of a randomized controlled trial. Before each session, participants self-reported past week BPD symptoms (ZAN-BPD) and attachment (ECR-S). To investigate how changes in attachment were associated with changes in BPD symptoms during treatment, we estimated a multilevel vector autoregression model (Bringmann et al., 2013), including time, symptoms, and attachment dimensions to generate networks of within-person contemporaneous and lag-1 associations, and between-person contemporaneous associations.

Results: Within-person reductions in attachment anxiety were associated with reduced fear of abandonment, suspiciousness, and moodiness, while reductions in avoidance were associated with reduced emptiness and identity disturbance (rₚ(78): .07-.11). Increased within-person suspiciousness was associated with session-to-session reductions in attachment avoidance (B = -.05, p = .02). Between-persons, participants with greater attachment anxiety reported less fear of abandonment and suspiciousness, participants with greater attachment avoidance reported less emptiness and greater fear of abandonment (rₚ(78): -.28-.44).

Discussion: On average, insecure attachment improved during BPD Compass and reductions in both domains were associated with reduced BPD symptomology. Attachment anxiety exhibited associations with mostly interpersonal symptoms, whereas attachment avoidance was associated more with internalizing symptoms. Results provide preliminary evidence of insecure attachment as a factor maintaining BPD symptoms and suggest it is a viable CBT treatment target.