Borderline Personality Disorder Traits in Early Adulthood: Development, Causes and Predictors

Abstract

Borderline personality disorder (BPD) was classified as a diagnosis in the Diagnostic and Statistical Manual of Mental Disorders, Third edition (DSM-III) in 1980. Since then, the causes of BPD have been extensively theorized and studied. For a long time, its etiology was assumed to be exclusively environmental. Especially traumatic experiences in childhood and young age have long been considered important environmental risk factors for BPD. From the early 2000s, genetically informative studies have revealed that also genes are important in the development of BPD. A complication has been that what we have considered and measured as environmental factors, such as life events, are in fact partly influenced by genes. This finding has important implications for our understanding of the causes of BPD, as almost everything we know about its development comes from pure association studies that do not control for the potential confounding effects of shared genetic influences between potential risk factors and BPD. By using data from a Norwegian population-based twin sample, the overreaching aim of this thesis was to enhance the knowledge about the development of BPD traits in early adulthood. More specifically, we examined whether childhood trauma and reported life events throughout adolescence have direct effects on levels of BPD traits in early adulthood, after controlling for shared environmental and genetic influences. Beyond studying the causes of BPD, we examined whether levels of sense of coherence (SOC) and feelings of loneliness in adolescence predicted levels of BPD traits in early adulthood. Not much is known about possible precursors of BPD in adolescence before personality disorders usually are diagnosed. To enhance the knowledge about the causal architecture behind these possible predictors of BPD traits (i.e., SOC and loneliness), we also examined their heritability and stability, and examined whether life events influenced levels of these characteristics in adolescence. The findings presented in this thesis suggest that childhood trauma (i.e., emotional abuse, physical abuse, sexual abuse, and witnessing violence) and negative life events in adolescence are associated with BPD traits mainly due to common genetic influences. That is, these measured environments during childhood and adolescence do not seem to have any causal effects on levels of BPD traits in early adulthood. With respect to SOC and loneliness, we found these constructs to be moderately heritable. In addition, both SOC and loneliness showed trait-like stability in adolescence similar to what is found for personality traits in general. Reported life events did not predict levels of SOC or loneliness throughout adolescence. Rather, life events were correlated with SOC and loneliness mainly for genetic reasons. Furthermore, lower levels of SOC and feelings of loneliness already at the age of 12 years were associated with increased levels of BPD traits in early adulthood. The prediction of SOC and loneliness on BPD traits increased in strength later in adolescence with also shorter time before the assessment of BPD traits. Findings from genetically informative analyses showed that the associations were mainly attributable to shared genetic influences. Together, these findings indicate that low levels of SOC and feelings of loneliness may be important indicators of later development of BPD due to their common genetic influences.