Logo: to the web site of Uppsala University

Background: Borderline personality disorder (BPD) is a heterogenous disorder with indistinct developmental trajectories. Diagnosing BPD is challenging, partly due to overlapping features with disorders such as attention deficit hyperactivity disorder (ADHD) and bipolar disorder (BD). The main aims of this thesis were to explore models of BPD development and examine the diagnostic boundaries between BPD, ADHD, and BD. 

Methods: In Study I we investigated the psychometric properties of the Swedish version of the Wender Utah Rating Scale (WURS) self-report questionnaire in 121 patients with BPD and/or AHD and/or BD.

In Study II we evaluated the temperament profiles and childhood trauma of 19 patients with BPD and compared them to 95 patients with ADHD, BD, and subclinical cases. We also explored a theoretical model of BPD development based on the interaction between temperament traits and childhood trauma.   

In Study III we conducted a cluster analysis in 150 patients with BPD, and/or ADHD, and/or BD based on the Attachment Style Questionnaire (ASQ).

In Study IV we investigated the prevalence of autistic traits in patients with BPD. Autistic traits were measured by the Autism Spectrum Quotient (AQ). We compared autistic traits between three groups: 1) patients with BPD (n=20), 2) patients with BPD with comorbidity (n=37), and 3) patients without BPD but with BD, and/or ADHD, and subclinical cases (n=98). Additionally, we investigated the association between autistic traits, childhood trauma, gender and BPD.     

Results: The Swedish version of the WURS displayed good psychometric properties and the proposed three-factor structure. Analyses suggested a cut-off score of 39.

The TCI subscales for Harm Avoidance (HA) and Novelty Seeking (NS) differed significantly between the groups. The interaction model between temperament (HA and NS) and trauma (TT) explained a small part of the variance of BPD.

Three clusters with adult attachment were identified. The three clusters differed in attachment characteristics, temperament, and percentage of patients with BPD with comorbidity and ADHD, but did not differ regarding frequency of childhood trauma.

Autistic traits were significantly higher in females with BPD with comorbidity compared to females without BPD. There was a correlation between autistic traits and the number of fulfilled BPD criteria in females. Autistic traits, childhood trauma, and gender individually predicted a BPD diagnosis.

Conclusions: Patients with BPD differed in temperament, attachment, and autistic traits compared to patients with ADHD and BD. A multifactorial model offers the best explanation for BPD development.