In the last decennia, the reluctance and ambivalence about assessing borderline personality disorder (BPD) in young people, has shifted to an increasing focus on adolescence and early adulthood as the developmental phases when BPD commonly has its onset . BPD is defined by high comorbidity and poor outcomes . BPD has been associated, early in the course of the disorder, with high levels of social impairment , such as poorer general psychosocial functioning, poorer peer relationships and problems with family relationships , and impairments in theory of mind and mentalizing . Moreover, research has shown that BPD in young people has a unique predictive value for poor psychosocial functioning, above and beyond Axis I disorders and other personality disorder diagnoses [3, 4]. Even at a subthreshold level, the criteria of BPD are associated with poorer social and occupational functioning in adolescence and young adulthood , as well as in adulthood [7, 8].
The diagnosis BPD is not a fixed diagnosis and the symptoms wax and wane during development. In adolescents, individual changes in psychosocial functioning appear to be related to changes in BPD symptoms, as an increase of BPD symptoms was related to an increase in psychosocial dysfunction, while a decline of BPD symptoms was related to improvement of psychosocial functioning . This is an important finding given that adolescence is considered a key developmental period concerning the onset of BPD. During adolescence and young adulthood, (subclinical) BPD may interfere with the process of gradually assuming more adult roles and responsibilities, necessary for adequate interpersonal functioning. Problems in interpersonal functioning are considered a central problem in BPD as well as in personality pathology in general  and in contrast to the relatively unstable nature of the diagnosis BPD, both in adolescents and in adults, problems in social functioning are relatively stable and may have long-lasting consequences for the individual’s functioning . Social and interpersonal functioning develop in the context of social relationships, which could be seen as the key element of understanding the course of BPD . This suggests that a better concept of relationships in relation to BPD in adolescence and young adulthood is needed in order to grasp the context of development of possible impairments in social functioning better.
Social relationships in young people with BPD
During adolescence and young adulthood the development of social autonomy, establishing intimate relationships, and finding a new balance in the relationship with parents are important developmental tasks . Bowlby's theory of attachment  states that in infancy and childhood, children are primarily attached to their parents because they can offer them security and protection. As children enter adolescence, the importance of physical protection and security decreases and adolescents become more independent and start to explore their environment solely . Throughout adolescence and young adulthood, internal representations of relationships which shape mental concepts of interpersonal behaviour in close relationships across the life span, are considered to become resistant to change and generalized to other close relationships [14, 16].
Given the interpersonal nature of BPD, the relationships of adolescents and young adults with (subthreshold) BPD can be challenged. Relatives, partners and friends of adults with BPD report considerable elevated objective and subjective burden and mental health problems, including depression and anxiety . Similarly, families and friends of young people with BPD features report elevated levels of distress, negative caregiving experiences, and family environments high in expressed emotion such as criticism and emotional overinvolvement, also when compared to adults in the general population or families and friends of young people with other serious illnesses . In addition, reciprocal relations between BPD symptoms and parenting factors in adolescent girls in the community have been found, indicating that parenting may affect subsequent BPD symptoms and vice versa .
Research on BPD features in relation to aspects of the relationships with mothers focuses mostly on maladaptive parenting, such as maternal abuse or neglect in the intergenerational transmission of BPD or abuse of neglect [20, 21], or affective behaviours during a discussion task [19, 22]. As both maladaptive parenting by the mother and maternal affective behaviours seem related to BPD in adolescents, this raises the question which features of the mother-child relationship exactly are related to BPD in young people and how these features are experienced by respectively the young person and the mother. As far as we know, associations between quality of relationships and BPD have not been studied yet from both the perspectives of both the young person and the mother.
Considering the current social relationships with parents, there is found that adolescents who are developing personality disorders are more likely to experience conflicts with family members throughout the transition to adulthood . In turn, persistent conflict with family members may have an adverse impact on psychosocial development throughout this important transitional period. Different explanations have been proposed for the findings that personality disorder traits were associated with both elevated contact and elevated conflict with family members . One of the hypotheses the authors stated was that due to social skills deficits and interpersonal conflict adolescents with personality disorders may find it difficult to maintain satisfying relationships with others outside the family circle. In addition, they may tend to maintain frequent contact with family members during the transition to adulthood because they need sustained support from the family. This raises the question whether BPD, in addition to challenges in relationships with parents, also is associated with challenges in the relationships with friends during adolescence and young adulthood.
As the importance of parental physical presence decreases, relationships with peers are increasingly taking over functions of the relationship with parents, including intimacy, advice on behaviours and feelings, and social influence [24, 25]. Friendships are particularly important for socialization towards more mature roles in late adolescence and early adulthood and play an important part in adaptive social development. Indeed, in a review Brechwald and Prinstein  have shown that in comparison to risky peer influence, healthy peer socialization processes can provide potential protection from maladaptive outcomes. Especially in adolescence, beyond friendships, the closeness elicited in best friends provides the context in which intimacy, trust and emotional support are established and tested .
BPD features are consistently associated with problematic functioning across facets of peer functioning, such as friendship quality, peer victimization and bullying and peer aggression . Friendships during adolescence can be characterized by intense intimacy, and conflict is likely to occur faster and have a bigger impact when contact in relationship is very frequent . Close friendships appear to place special burdens during early adolescence, and the expectations for exclusivity of best friendship in particular may be a marker of BPD . Although different findings in associations between parental and best-friend relationship quality and BPD were reported [11, 31, 28], research is needed to consider the dynamics of exclusive “best-friend” roles in relation to BPD . The literature paints a developmental pictures on how friendship relationships may be seen as an extension of the attachment relationship developed in parent-child relationship in childhood. However, few studies have integrated the findings on quality of parental and best-friend relationships, leaving a gap in how quality of both parental and best-friend relationships might be associated with BPD and whether these associations are related to age during adolescence and young adulthood.
The current study
Taken together, gaining insight into the associations between BPD features and social relationships with mothers and a best friend in adolescents and young adults, would provide a window into the context for development of psychosocial functioning during a crucial phase in the onset of BPD. The current study aims to add to the literature by examining relations between quality of relationships with both mothers and a best friend and BPD in young people, investigate the role of age in these relationships, and compare the results on the quality of relationships with the mother as reported by the young person with mother reported quality of the relationship.
We expect that less supportive interactions and more negative interactions with mothers are related to more BPD symptoms. Similarly, we expect that less supportive interactions and more negative interactions with a best friend are related to more BPD symptoms [26, 32]. With regard to the role of age, it is hypothesized that the link between maternal supportive and negative interactions, and BPD becomes weaker when adolescents grow older. Thus, associations between the maternal factors (maternal supportive and maternal negative interactions) and BPD might be smaller for older individuals and larger for younger individuals (i.e., moderation). For relationships with best friends, the opposite pattern is hypothesized, that is, we expect that associations between the best friend factors (best friend supportive and negative interactions) and BPD might be larger for older individuals and smaller for younger individuals.
Finally, in addition to study the quality of parental relationships from a young persons’ perspective, a parents perspective on this relationship can add to a more informative view by understanding individual differences between adolescents’ and parents’ reports of family relationships. Therefore, we will test robustness of our findings by comparing the results based on self-report with results from a subsample in which supportive and negative interactions with mothers were rated by the mother.