| dc.description.abstract | Background: Borderline personality disorder is thought to be a serious mental disorder with a series of challenges in everyday life, and is characterized by severe psychosocial impairment and a high mortality rate due to suicide (Lieb et al., 2004). Multiple studies have found a high prevalence of traumatic experiences in childhood among BPD patients, and BPD is the most frequently studied out of the PDs due to the association with early traumatic experiences. We addressed the following problems: One, by applying a regularization regression method for model sparsity, what types of stressful life events are associated with borderline personality disorder traits? Two, to what extent is the association between the combined stressful life events and BPD accounted for by familial genetic and environmental background factors? Method: In total 2801 twins between the age of 19 and 36 participated, recruited from The Norwegian National Medical Birth Registry. The sample was obtained from the Norwegian Institute of Public Health (NIPH) Twin Panel. The data was first collected from 1992-1998 (Q1-Q2), and later from 1999 (wave 1) to 2011 (wave 2). In wave 1, each twin in a pair was interviewed by a different interviewer. The interview study in wave 1 consisted of two parts: a Norwegian computerized version of the Munich-Composite International Diagnostic Interview (M-CIDI) assessing ICD-10 lifetime diagnoses, and The Norwegian version of the Structured Interview for DSM-IV Personality (SIDP-IV) assessing the DSM-IV Axis II personality disorders. Further, a total of 2284 twins were re-interviewed by telephone between 2010 and 2011 in wave 2, approximately 10 years later. In this wave, the participants were also mailed a questionnaire that consisted of questions about normative personality, maladaptive personality, and substance abuse. In addition, the experience of stressful life events was assessed. An overview of prevalence of BPD traits and experienced SLEs was presented in a descriptive manner. Negative binomial Poisson regression was used to explore which SLEs could predict BPD traits. An Elastic net analysis was then performed to develop 2 the final logistic regression model for prediction. At last, discordance within twin pairs was examined and a co-twin control was performed to differentiate between environmental and genetic factors in our sample. The data in this study is based on already collected data material from the Axis I / Axis II study. Conclusion: We found that in childhood, life-threatening experiences and an unpredictable and unsafe early life environment is associated with the development of BPD traits. In adulthood, life-threatening experiences, economic issues, and relationship conflicts seem to have the same association. Our results indicate that these experiences can predict 22% of the development of BPD traits. In the debate about how stressful life events are associated with the development of BPD traits, it does not appear to be a causal factor. Our results point to there being something in the shared family environment that causes both the stressful life events and the development of BPD traits. It is therefore of importance to limit the effect of, or prevent, familial environmental background factors causing both SLEs and BPD traits. | eng |