Self-reported patterns of impairments in mentalization, attachment, and psychopathology among clinically referred adolescents with and without borderline personality pathology

In adult populations, personality disorders (PDs) in general and borderline personality disorder (BPD) in particular are related to significant impairments in general functioning when compared to subjects without PD diagnoses and those with other mental disorders [1]. Specifically, evidence suggests that adults with PD exhibit poorer social and interpersonal functioning, are less likely to prospectively maintain an occupation, and report less life-satisfaction compared to people without PD [2–4]. Regarding adolescents, longitudinal studies show that early maladaptive and pathological personality features predict later social and functional impairments (i.e., failure to complete school, alcohol and drug dependence, and hazardous and antisocial behaviors) [5–10].

Additionally, studies report a high prevalence of PDs in both the general and clinical populations [11] and that these disorders are associated with excessive societal costs [12, 13]. The increasing attention given to and research conducted in the field of PDs in adult populations has encouraged the development of new and specialized treatments for adults with PDs, notably BPD, in the last two decades [14].

Historically, however, less attention has been given to PDs in childhood and adolescence [15]. Until recently, many clinicians and researchers did not acknowledge the existence of personality pathologies in adolescents [16–18]. Indeed, they did so despite the fact that, according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) and its predecessors, PD diagnoses may be applied to adolescents when the individual’s particular maladaptive personality traits appear to be pervasive and persistent, are unlikely to be limited to a particular developmental state or to another mental disorder, and are present for 1 year or more [19].

Indeed, the available research suggests that PDs in adolescents younger than 18 years can be diagnosed as reliably and with as much validity as in adulthood and that the prevalence of PDs in adolescents in both the general population and clinical settings are comparable to those reported for adults [15, 17, 20, 21]. Developmental research suggests that PDs are moderately stable during adolescence [22] and are strongly related to childhood emotional difficulties and problematic behavior [23–25]. Furthermore, studies have indicated that delays in the diagnosis of PDs and the provision of interventions in adolescence can potentially result in devastating consequences and poor long-term prognoses [26–28].

Most theoretical and empirical developmental models of BPD either implicitly or explicitly assume that attachment problems or interpersonal trauma and difficulties are related to the later development of BPD. According to the mentalization-based model of BPD, the core pathology underlying BPD is associated with dysfunction in mentalization and insecure attachment patterns [29]. Mentalization refers to the ability to understand the self and others as intentional agents with minds [30]. Mentalizing is considered important for interpersonal functioning because it enables people to understand behavior in terms of mental states in regard to both the self and others [31]. Research has demonstrated that dysfunctions in mentalization are a core feature in patients with BPD [32], and based on many studies that link BPD and mentalizing dysfunctions, promising theories have been proposed that apply the mentalization-based model to explain the emergence of BPD in adolescents [33]. The mentalizing theory suggests that the capacity to mentalize is developed via the close relationship between a child and his or her primary caregiver and is dependent upon a secure attachment relationship [31] in which the primary caregiver adequately mirrors the child’s mental state. The mirroring process must be both contingent (e.g., fear is mirrored with fear and not joy) and marked (e.g., the mental state being mirrored must be similar but clearly different from that of the caregiver). Thus, a secure attachment relationship in which the caregiver benignly and accurately represents the child as an intentional agent with intentions, thoughts, and emotions underpins the development of the capacity to mentalize and secure the normative development of the child’s personality [34].

In contrast, the pathological trajectory leading to BPD is characterized by a caregiver who is unable to provide a secure attachment relationship, specifically defined by inadequate mirroring (i.e., un-marked and non-contingent; see [34] for details). In this case, because the caregiver is unable to mirror and represent the mind of the child, the child will display difficulties in understanding how actions and mental states are linked in the self and others [31]. In the mentalizing theory, the difficulties pertaining to insecure attachment relations and dysfunctional mentalizing, as explained above, are specifically believed to underlie BPD. This does not mean that BPD is the only psychiatric disorder characterized by dysfunctional mentalization and insecure attachment [31]. However, the mentalizing theory emphasizes these characteristics in particular as underlying BPD. Problematic attachment relations and dysfunctional mentalization have also been found in empirical studies in adolescents with BPD [35–38]. Recent empirical findings showed that problematic family functioning and low maternal care were predictors of BPD in adolescents [39], underscoring the role of attachment relationships between parents and children in the development of BPD in adolescents. Another recent study, displayed how BPD patients compared to non-BPD psychiatric controls and healthy controls, showed more dysfunctional emotion regulation, even when controlling for important sociodemographic and clinical variables [40]. In a community-dwelling study with Italian adolescents, findings showed that non-suicidal self-injury (NSSI) and emotional dysregulation are moderately related to BPD features in adolescents [41]. This finding was replicated by Kaees and colleagues [42] in adolescent inpatients with NSSI and suicide attempts (SA) and showed that dimensional borderline pathology was associated with NSSI and SA. In line with recent developmental theories explaining BPD (i.e., mentalisation-based theory), Sharp and colleagues [43] found that specifically hypermentalizing (i.e., ascribing intentions and beliefs to people where non is) mediated the relationship between attachment coherence and borderline pathology. In another study Ramos and colleagues [44] found, in a sample of 60 adolescents BPD patients, that attachment anxiety was positively related to internalizing psychopathology but negatively related to externalizing pathology. Furthermore, in a study examining the trajectories of borderline pathology and psychosocial functioning, results indicated that the development of BPD was significantly related to worsening in academic, social and mental health outcomes [45]. Finally, in a recent systematic review and meta-analysis, Winsper and colleagues [46] found that BPD in adolescents is related to the same aetiological and psychopathological issues as those found in adults with BPD.

Despite emerging theories on BPD in adolescents and research findings pointing to psychological dysfunctions in BPD, there still exist gabs in the research literature on BPD in adolescents. First, a variety of different clinical variables have been identified as pertaining to BPD, but rarely have they been investigated in the same study. Second, many studies have compared BPD groups to healthy controls but few have included a clinical non-BPD comparison group. Third, and specifically related to attachment, no studies have explored the quality of self-reported attachment in relation to both parents and peers in patients with and without BPD. Finally, and to the authors’ knowledge, no studies have been conducted as a naturalistic clinical comparison study in an ordinary child and adolescents’ psychiatric clinic, adding ecological validity to the findings. Hence, to the best of our knowledge, no studies have explored the differences between patients with and without borderline pathology in terms of attachment, mentalizing and psychopathology in a sample of adolescent psychiatric patients.

Thus, the aim of this study was to explore the patterns of impairment in an outpatient adolescent clinical sample diagnosed with BPD compared to those of a clinical group without PD but with other mental disorders. Specifically, we wanted to determine whether there was a difference between BPD and clinical comparison subjects with respect to attachment to peers and parents and mentalization. We also examined differences regarding the severity of psychopathology, self-harm and risk-taking behaviors, and depression. We hypothesized that the BPD group would display more problematic attachment relations, more mentalizing dysfunctions, a significantly higher level of psychopathology, more depressive features and more self-harm and risk-taking behaviours than the group without BPD. We also predicted that significant differences would be apparent from both the dimensional (number of borderline features) and categorical (meeting the criteria for a BPD diagnosis) perspectives.