Attachment style and schema as predictors of social functioning in youth with borderline features

Theorized associations between adverse attachment experience, maladaptive core schema (EMS) and the interpersonal dysfunction seen as characteristic of borderline personality disorder were investigated in two groups of outpatient youth aged between 15 and 25 years. Patients with sub‐syndromal (meeting at least 3 DSM‐IV criteria) and full syndrome BPD (n = 30) and patients with major depressive disorder completed self‐report measures of adult attachment style, EMS and interpersonal functioning. Results for both groups indicated a level of disadvantage for each of the domains assessed. Contrary to prediction, mean scores for the two groups did not differ for anxious or avoidant attachment style or for 13 or the 15 EMS assessed. BPD participants scored at a significantly higher level for the Mistrust/abuse and Entitlement schema. Consistent with results reported for adult samples, the BDP group acknowledged a predominantly anxious attachment style, a wide range of EMS and poor interpersonal functioning. Hierarchical multiple regression analyses undertaken for each EMS domain in turn indicated that an anxious attachment style and endorsement of the Social Isolation EMS were the strongest predictors of social functioning. Against, prediction, there was limited evidence supporting the specificity of theorized relationships between BPD diagnosis and attachment style and prominence of Disconnection and Rejection domain EMS. Copyright © 2011 John Wiley & Sons, Ltd. Borderline personality disorder (BPD) is an etiologically complex disorder characterized by emotional instability, impulsivity and self and interpersonal impairment (Livesley, 2008; Skodol et al., 2002). A high level of social dysfunction is a prominent and enduring feature in adults with BPD (Skodol et al., 2005; Winograd, Cohen, & Chen, 2008) and in young people assessed early in the course of the disorder (Chanen, Jovev, & Jackson, 2007). The latter study found that young outpatients with three or more BPD features had higher levels of social dysfunction than those with other personality disorders and that their social impairment was not attributable to Axis 1 diagnoses. Poor social adjustment has been found to be an enduring feature of BPD, even when symptomatic Copyright © 2011 John Wiley & Sons, Ltd. 5: 209–221 (2011) DOI: 10.1002/pmh Personality and Mental Health 5: 209–221 (2011) Published online 2 June 2011 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pmh.169 improvement has occurred (Skodol et al., 2005), and also a prospective predictor of the risk of suicide attempts (Brodsky, Groves, Oquendo, Mann, & Stanley, 2006; Soloff & Fabio, 2008). Awareness of the high personal and social costs from derailments in social and occupational trajectories for young people with BPD provides impetus for research that goes beyond description to examine factors contributing to social dysfunction in BPD. Theoretical accounts suggest that the social dysfunction seen in BPD has its origins in early psychosocial adversity, including in dysfunctional primary relationships that foster the development of maladaptive attachment styles and self‐ invalidating modes of thought (Fonagy&Bateman, 2008; Gunderson & Lyons‐Ruth, 2008; Young, Klosko, & Weishaar, 2003). BPD diagnosis is reliably associated with the report of early separation and loss and disturbed (conflicted, distant or uninvolved) relationships with one or both parents, childhood sexual abuse and emotional neglect (Zanarini & Frankenburg, 2007). There is reliable evidence, also, of associations between insecure attachment styles and BPD diagnosis (for a review, see Agrawal, Gunderson, Holmes, & Lyons‐Ruth, 2004), and adults and adolescents with BPD have been found to endorse a range of maladaptive beliefs regarding self and others (e.g. Ball & Cecero, 2001; Jovev & Jackson, 2004; Lawrence, Allen, & Chanen, 2010; Nordahl, Holthe, & Haugum, 2005). However, evidence for the specific contribution of attachment style and maladaptive beliefs about self and others to social dysfunction in BPD is limited. This paper reports on an investigation of these associations in two youth outpatient samples, including a sample with sub‐threshold or full threshold BPD and a sample with major depressive disorder (MDD). Research grounded in Bowlby’s (1969, 1980) attachment model has found adults with BPD report a fearful attachment style characterized by high levels of both anxiety, related to expectations of rejection and abandonment and perceptions that one is unlovable, and of avoidance of intimacy, based on expectations that attachment needs will not be met (Agrawal et al., 2004). However, associations reported between self‐ reported anxious attachment style and a diagnosis of depression, and a fearful (anxious and avoidant) attachment style and the experience of recurrent depression raise questions about the specificity of the relationships between attachment style and diagnosis (e.g. Conradi & de Jonge, 2009). Fossati et al. (2001) found that attachment style discriminated BPD patients from non‐clinical participants and patients without personality disorder but not from those with other personality disorders. Indeed, evidence indicating limited specificity as regards attachment style and diagnosis is consistent with Bowlby’s (1969, 1980) formulation of insecure attachment styles as manifestations of broadly based internal working models. Internal working models are seen as multi‐faceted, comprising dysfunctional core schema, autobiographical memories, goals and motives and behavioural strategies that function to shape unhelpful self perceptions, behaviours and expectations of others in relationships (Collins & Read, 1994; Main, Kaplan, & Cassidy, 1985). For the most part, research into the association between attachment style and core schema relating self and others has drawn on Young’s (1999; Young, Klosko, & Weishaar, 2003) model. Young et al. (2003) specifically identified internal working models that have become rigid and resistant to change as the precursors of early maladaptive schema (EMS). Young et al. (2003) defined EMS as enduring coping styles reflecting broad, pervasive self‐ defeating themes regarding self and relationships with others whose development is influenced by genetic predispositions as well as aversive family experiences. EMS are organized into five domains reflecting a core emotional need (disconnection and rejection, impaired autonomy, impaired limits, other‐directedness and overvigilance). Although results from a Schema Therapy treatment outcome study for BPD were encouraging 210 Hulbert et al. Copyright © 2011 John Wiley & Sons, Ltd. 5: 209–221 (2011) DOI: 10.1002/pmh (Giesen‐Bloo et al., 2006), conclusive evidence for the theorized associations of Young et al. between harsh or aversive family experiences, the development of EMS and later personality disorder diagnosis is lacking. Findings regarding the prediction of Young et al. (2003) of an association between the diagnosis of BPD and the endorsement of disconnection and rejection domain EMS are quite inconsistent, with a number of studies reporting significant relationships for abandonment and/or mistrust/abuse EMS (Ball & Cecero, 2001; Jovev & Jackson, 2004; Lawrence et al., 2010; Nordahl et al., 2005). Results across studies indicate that adult and adolescent BPD patients acknowledge a range of EMS, including some theoretically related to other personality disorders (Ball & Cecero, 2001; Jovev & Jackson, 2004; Lawrence et al., 2010; Nordahl et al., 2005; Petrocelli, Brain, Glaser, Calhoun, & Campbell, 2001; Reeves & Taylor, 2007). No consistent findings have emerged from this research. For example, significant associations have been reported between BPD diagnosis and dependence/incompetence and vulnerability to harm EMS from the impaired autonomy and performance domain (Nordahl et al., 2005) and, also, entitlement EMS from the impaired limits domain (Reeves & Taylor, 2007). To our knowledge, investigation of theorized links between attachment style, disconnection and rejection EMS and the social dysfunction seen in BPD is limited. Findings from a clinical study using a mixed diagnosis outpatient sample showing that adults with a fearful (high anxious and avoidant) attachment style acknowledged more EMS and relationship dysfunction than those with a dismissive (low anxiety and high avoidant) or preoccupied (low anxious and avoidant) attachment style (Mason, Platts, & Tyson, 2005) are broadly consistent with the model of Young et al. (2003). Investigations of associations between self‐reported attachment style in adult romantic relationships and maladaptive beliefs in non‐clinical undergraduate samples have returned findings relevant to the study of social functioning in young adults with BPD features. For example, Wilkinson’s (2006) finding that securely attached university students endorsed fewer EMS than their less securely attached peers is consistent with reported associations between BPD diagnosis and higher levels of endorsement of EMS across schema domains. Further, students’ self‐reported rejection sensitivity and negative self beliefs were found to mediate the relationships between endorsement of anxious and avoidant attachment styles in current romantic relationships and the number of BPD features (Boldero et al., 2009). Social psychology research indicates that a higher level of rejection sensitivity is associated with poorer social functioning, including increased interpersonal conflict (Downey & Feldman, 1996). Using a youth sample, Chanen et al. (2007) noted that BPD diagnosis predicted poor functioning in peer relationships, a finding not attributable to any Axis 1 disorder. However, the same study reported a predictive relationship between mood disorder and family and relationship problems. Hence, this study aimed to extend the literature by investigating the specificity of relationships between attachment style, EMS and current social functioning in outpatient youth with sub‐syndromal or full‐syndrome BPD, compared with outpatient youth with a DSM‐IV (APA, 2000) diagnosis of MDD and no personality disorder diagnosis. The present study builds on very reliable findings of associations between BPD diagnosis and highly anxious and avoidant attachment styles and, also, endorsement of abandonment and mistrust/abuse EMS. The use of a sample of youth early in the course of BPD allowed testing of theorized links while minimizing potentially confounding duration of illness effects, such as neurobiological changes, prolonged mental illness, continuing psychosocial adversity, medication and substance use (Chanen et al., 2009). Given findings of high levels of depressive symptoms in youth with BPD features and evidence of associations between mood disorder and social functioning (Chanen et al., 2007), the inclusion of a sample of depressed 211 Predictors of social functioning in youth with BPD features Copyright © 2011 John Wiley & Sons, Ltd. 5: 209–221 (2011) DOI: 10.1002/pmh young outpatients allowed a more rigorous investigation of the contribution of BPD features to the variables of interest. The study hypothesizes were that, compared with participants with MDD and no personality disorder diagnosis, young people with BPD features would acknowledge higher levels of anxious and avoidant attachment, disconnection and rejection domain EMS, abandonment and mistrust/abuse and social dysfunction. Further, it was hypothesized that, with the level of current depression controlled for, attachment style (anxious and avoidant), disconnection and rejection domain EMS of abandonment and mistrust/ abuse and a diagnosis of BPD would contribute significant unique variance to the prediction of social functioning.