Phenomenology and Diagnosis of Personality Disorders at Adolescence
Already by the late 1990s Paulina
Kernberg (1997;
Kernberg et al. 2000) had identified a constellation of severe symptoms indicative of PDs in adolescents. This led her to advocate for awareness of adolescent BPD. She argued that it is important to diagnose BPD in young people and that early intervention is essential to help adolescents address the severe disruptions associated with BPD and reduce the negative impacts during this critical period so as to help adolescents resume a more adaptive course of development. Current guidelines published by the U.K. National Institute for Health and Care Excellence (NICE) (
National Institute for Clinical Excellence 2009; see also
Kendall et al. 2009) and the Australian
National Health and Medical Research Council (2012) also provide clear support for diagnosing BPD in young people, and they propose a less restrictive age range than DSM-5 and ICD-10. The World Health Organization (WHO) Guideline Development Group (GDG;
World Health Organization 2009) also indicates that their proposals about BPD apply to young people postpubertally. Furthermore, they stipulate that in some circumstances the diagnosis of BPD may be warranted even before age 13. The GDG recommends that clinical features such as suicidal/self-harming behaviors, significant emotional instability, increasing intensity of symptoms, multiple comorbidities, poor response to ongoing treatment, and high level of functional impairment should alert clinicians to assess for the possibility of BPD as part of a comprehensive clinical evaluation.
Despite these practice guidelines clearly indicating that adult criteria can be reliably used for diagnosing and treating young people from age 14 on, clinicians remain hesitant to diagnose BPD in adolescents or young adults. Surveys show that less than 40% of clinicians who work with adolescents diagnose BPD in patients younger than 18 years (
Griffiths 2011;
Laurenssen et al. 2013). Common concerns clinicians give for not wanting to make the diagnosis are that the problems may be transient (41%); that DSM-IV-TR does not allow PD diagnoses in adolescents (26%); and that the diagnosis could be stigmatizing (9%). In sum, these findings suggest that scientific findings and practice guidelines are not currently being applied and integrated into clinical practice (
Coghill 2014).
BPD in young people is also frequently overlooked because of its comorbidity with mood disorders (i.e., anxiety, depression, and bipolar disorders) and behavior difficulties (e.g., oppositional defiant disorder [ODD], conduct disorder [CD]). When mood and behavior difficulties are present, there is a tendency to overlook the possibility of PD and focus only on Axis I disorders. However, considering that untreated BPD in adolescence predicts Axis I and II problems in adulthood (
Cohen et al. 2007), it is important for clinicians to be alert to BPD features in adolescents and preadolescents. Identifying and treating BPD is essential to lessen the likelihood of undiagnosed and untreated BPD disrupting the course of development, undermining successful engagement with key developmental challenges, and increasing the risk of maladjustment and psychopathology in adulthood.
DSM-5 Criteria and Relevant Research
DSM-5 currently includes a mixed categorical and dimensional approach to the diagnosis of BPD (
Oldham 2018;
Skodol 2018;
Skodol et al. 2014), with a categorical approach outlined in Section II and a dimensional approach described in Section III. This hybrid model is the outcome of an attempt to replace the categorical approach with a dimensional one more consistent with empirical evidence suggesting that BPD is best considered as levels of disturbance across a range of different functional domains. However, concerns that this dimensional approach may have disadvantages in clinical settings, where clinicians tend to rely on a categorical approach to diagnose and organize interventions, led to the retention of the categorical diagnostic criteria. At present, Section II responds to the needs of clinicians who are used to thinking categorically, while Section III provides an alternative dimensional approach. Although the Section III approach may be more compatible with research and thinking about the processes underlying the pathology, its practical utility in real-life clinical settings remains to be seen. Section II specifies that the diagnosis of PD can be applied to children and adolescents “in those relatively unusual instances in which the individual’s particular maladaptive personality traits appear to be pervasive, persistent, and unlikely to be limited to a particular developmental stage or another mental disorder” (
American Psychiatric Association 2013, p. 647). Furthermore, a duration of only 1 year is necessary for the diagnosis of child and adolescent PD, in contrast to the 2 years required before adult BPD can be diagnosed.
The criteria specified in Section II of DSM-5 for the diagnosis of BPD include abandonment fears, unstable and intense interpersonal relationships, identity disturbance, impulsivity, suicidal behaviors, affective instability, chronic feelings of emptiness, inappropriate intense anger, and transient stress-related paranoid ideation or severe dissociative symptoms. BPD can be clearly and systematically distinguished from externalizing and internalizing pathology by the presence of a mix of both internalizing and externalizing symptoms. Furthermore, the symptoms of BPD in terms of their pervasiveness and severity are clearly distinguishable from the range of reactions and “ups and downs” that adolescents may have over the course of engaging with the characteristic challenges, inevitable frustrations, and possible failures of this period.
As a complement to Section II, Section III of DSM-5 provides a dimensional approach. A standard template is provided in Section III that requires clinicians to consider functioning in terms of two criteria. Criterion A requires evaluation of difficulties involving identity, self-direction, empathy, and intimacy. Criterion B requires the presence of at least four of the following seven traits of personality pathology: emotional lability, anxiousness, separation insecurity, depressivity, impulsivity, risk taking, and hostility. Any PD must meet two primary defining criteria: Criterion A (moderate or greater impairment in personality functioning in the realms of self [identity and self-direction] and interpersonal relationships [empathy and intimacy]) and Criterion B (presence of one or more pathological personality traits in any of five trait domains [negative affectivity, detachment, antagonism, disinhibition, and psychoticism]). Criteria C through G must also be satisfied, as in DSM-5 Section II: Criterion C (inflexibility and pervasiveness), Criterion D (relatively stable impairment in functioning over time), Criterion E (impairment not better explained by another mental disorder), Criterion F (impairment not solely attributable to substance use or a medical condition), and Criterion G (impairment not better understood as a normal developmental stage or a normal aspect of a sociocultural environment).
Borderline Personality Disorder: A Disorder Encompassing Heterogeneous Symptoms
Considering the fact that BPD criteria can be met by individuals with very different symptom combinations, questions have been raised regarding the validity and phenomenology of BPD and whether the DSM criteria describe a single disorder. Current evidence based on studies of BPD in adolescents (
Michonski 2014) suggests that while multidimensional models best describe the phenomena of BPD, a unidimensional model is also supported, and the dimensions or factors hang together and are strongly related, consistent with the notion of BPD as a unidimensional construct and disorder (
Michonski 2014;
Sharp and Fonagy 2015). From a scientific/research perspective, further studies are required to examine whether BPD in adolescents can be demonstrated to be unidimensional and differentiated from other PDs, or whether, as in adults, it is difficult to demonstrate undimensionality for BPD when PDs are analyzed simultaneously.
Another important line of investigation has examined whether particular symptoms can be identified that are most predictive of adolescent BPD. Earlier studies identified chronic feelings of emptiness and inappropriate intense anger (
Garnet et al. 1994), and later studies pointed to identity disturbance, affective liability, and inappropriate intense anger (
Becker et al. 2002;
McManus et al. 1984;
Meijer et al. 1998;
Westen et al. 2011) as most predictive of adolescent BPD. Furthermore, there appears to be gender-related differences in predictors of BPD, with paranoid ideation in boys and identity disturbance in girls appearing most predictive (
Michonski et al. 2013). BPD frequently is comorbid with internalizing and externalizing disorders, with comorbidity rates ranging from 70.6% to 86% in adolescent clinical samples (
Ha et al. 2014;
Speranza et al. 2011). This suggests that BPD represents a confluence of internalizing and externalizing disorders that is best understood in terms of a separate disorder (
Sharp and Fonagy 2015).
Table 1–1 presents core diagnostic features identified for BPD in adolescents (from
Fossati 2014), early childhood markers of vulnerability to BPD, and positive outcome (“resilience”) factors. In the case that follows, we present a typical manifestation of an adolescent with BPD.
Developmental Precursors and Etiological Risk Factors
The development of BPD is likely a multidetermined process with roots in early childhood that tends to more fully manifest around puberty (Chanen and Kaess 2012; Shiner 2009; Zanarini et al. 2001) and involves complex and interactive biological and psychosocial risk factors (
Carlson et al. 2009;
Crowell et al. 2009;
Lenzenweger and Cicchetti 2005;
Paris 2003a,
2003b,
2007). BPD is marked by disturbance in mood (e.g., wide mood swings), impulsivity (e.g., nonsuicidal self-injury, overdoses, substance abuse), and interpersonal dysfunction (e.g., stormy relationship with attachment figures). There is no single explanation for its cause; risk factors vary among patients and reflect the concept of equifinality (
Cicchetti and Rogosch 2002), in which different pathways may lead to similar outcomes. Risk factors include
genetic contributors (
Distel et al. 2008), and more specifically inherited biological plasticity genes that interact with positive and negative events (
Amad et al. 2014);
temperamental contributors, which may be noticeable in early infancy and childhood and expressed as negative affectivity, stress reactivity, and impulsivity, which are linked to emotional vulnerability (
Goodman et al. 2004;
Posner et al. 2003) and which can have a negative impact on others and contribute to the development of poor attachment relations; and
environmental and
experiential contributors, such as the manner in which parents interact with, are influenced by, and maintain or modify their child’s behaviors, which can also influence gene expression.
Genetic Risk Factors
There is growing evidence that genetic polymorphisms underlie specific personality traits and that secondary traits (e.g., emotional dysregulation, constraint/conscientiousness) may be sufficient to represent personality pathology (
Livesley 2005) as described by some theoretical models. There appears to be a strong heritability component to BPD (
Torgersen et al. 2000), with evidence of familial patterns in BPD in which first-degree relatives are more likely to have BPD and BPD symptoms (
Zanarini et al. 2005). More specifically, research suggests that genetic factors substantially increase the risk for developing BPD as well as antisocial personality traits, as indexed by examination of the genetics of the serotonin neurotransmitter system (
Lyons-Ruth et al. 2007). This genetic effect appears to be independent of the quality of early care. For instance, in a longitudinal study of borderline personality–related characteristics (BPRC;
Belsky et al. 2012), in which twins were examined at ages 7, 10, and 12 years, genetics were found to account for 66% of the variance in BPRC. Variables examined included family history of psychopathology, physical maltreatment, and maternal negative expressed emotion. Furthermore, harsh parental treatment had a more powerful impact on children originating from families with a positive history of the disorder, thereby supporting a diathesis-stress model.
Temperamental Risk Factors
Extreme temperamental vulnerability may well be one of the most important risk factors for the development of BPD. In a demonstration of the intricate interaction of temperamental and experiential factors,
Zanarini and Frankenburg (2007) offered a “complex” model based on more than a decade of empirical research on the etiology of BPD. They suggested that a vulnerable “hyperbolic” temperament is central to the development of BPD. Individuals with this temperament are easily offended, and they work hard to get others to pay attention to their inner pain, often by utilizing indirect means to obtain comfort or support, and will covertly reproach others for their “insensitivity, stupidity, or malevolence.” BPD symptoms emerge when these individuals encounter a “kindling” event, which may occur in latency or in early, middle, or late adolescence.
Zanarini and Frankenburg (2007) propose that even relatively normative events, such as starting one’s first job or beginning one’s first intimate relationship, may be stressful enough to be a trigger for these extremely vulnerable individuals.
Rogosch and Cicchetti (2005), while studying the impact of maltreatment in childhood, constructed a BPD precursor composite based on personality, interpersonal relationships, and representational models of the parents, self, and peers, as well as suicidal/self-harming behavior. They found that both maltreatment and attentional control (a particular cognitive dimension of temperament) independently predicted the BPD precursor composite. This finding is consistent with
Zanarini and Frankenburg’s (2007) model with regard to the interaction of temperament and experience, as well as with
Belsky et al.’s (2012) findings, but in addition underscores the importance of considering the contribution of maltreatment experiences.
Environmental and Experiential Risk Factors
There are a range of findings associating different features of parenting with the development of BPD in children. These features include aspects of early rearing and the quality of caretaking to later interactions and parents’ capacity to foster autonomy and promote healthy identity formation.
There is converging evidence that BPD is more likely to develop in children who have experienced neglect and abuse associated with family dysfunction, parental psychopathology, and family interactions that are invalidating, conflictual, negatively critical (
Fruzzetti et al. 2005), and unempathic (
Guttman and Laporte 2000). Most forms of adversity are more frequently reported by BPD patients, and many of these forms of adversity co-occur (
Zanarini et al. 1997,
2000).
Zanarini et al. (2005) reviewed the evidence regarding the link between BPD and maltreatment, including early separation and loss, disturbed parental involvement, experiences of verbal and emotional abuse, experiences of physical and sexual abuse, and experiences of physical and emotional neglect. Prolonged early separations and losses were significantly more common among the childhood histories of BPD patients than among the childhood histories of other diagnostic groups. BPD patients also reported significantly more highly conflictual, distant, or uninvolved relationships with maternal figures. The failure of fathers to be present was another discriminating factor, as were disturbed relationships with both parents. Verbal and emotional abuse was also significantly more present in the histories of patients with BPD than in those of depressed patients or patients with other Axis II pathology. Physical and sexual abuse was also relatively common in the histories of patients with BPD; sexual abuse was significantly more common in patients with BPD than in other psychiatric patient comparison groups, although physical abuse was not. In addition, severity of sexual abuse was related to severity of symptoms. The findings with regard to physical neglect were inconclusive. However, emotional neglect (emotional withdrawal, inconsistent treatment, denial of feelings, lack of a real relationship, parentification of the child, and failure to provide needed protection) was very common among BPD patients and highly discriminating.
As
Rogosch and Cicchetti (2005) observed, the similarities in the dysfunctional developmental processes displayed by maltreated children and adult patients with BPD are suggestive of a prospective pathway from childhood maltreatment to BPD. Maltreated children show sequelae across multiple domains of functioning that are similar to the key features of BPD, including affect dysregulation; relationship disturbances with parents and peers; disturbances in representations of self, parents, and peers; cognitive and affective processing anomalies; and maladaptive personality organization (
Cicchetti and Valentino 2006). Furthermore, maltreatment is related to suicidal ideation and behavior and psychopathology (
Cicchetti and Toth 2016). However, it is important to take into account that although maltreated children are more likely to show precursors of BPD, not all do, and there are also children who have not been abused who display BPD precursors (
Rogosch and Cicchetti 2005). Child temperament is likely to interact with and exacerbate parenting vulnerabilities. For example, the home environment is more chaotic for children who are high in extraversion-urgency, but less chaotic for childen with high effortful control (
Lemery-Chalfant et al. 2013a,
2013b), likely because children with greater effortful control make fewer demands and provoke less negative caretaking behaviors from more vulnerable parents.
Maltreatment that evokes fear and aggression or involves children being left for long periods without help to regulate affect is known to undermine the development of regulatory processes (
Fonagy et al. 2002). Maltreatment may have permanent negative implications for affect regulation through the negative impact of prolonged exposure to high levels of cortisol on the developing hypothalamic-pituitary-adrenal (HPA) axis, with long-term consequences for the physiological processes associated with stress regulation (
Cross et al. 2017). Childhood maltreatment has been identified as a risk factor for BPD in adulthood in community samples (
Afifi et al. 2011;
Widom et al. 2009), as well as in clinical samples (
Battle et al. 2004;
Chiesa and Fonagy 2014;
Sansone et al. 2011). Research on the relationship between maltreatment and BPD in adolescence is more limited, but consistent with the adult literature, there is some evidence that childhood abuse and neglect are risk factors for the development of borderline traits in school-age children and adolescents (
Bounoua et al. 2015;
Jovev et al. 2013;
Zelkowitz et al. 2001). For example, children who experienced sexual abuse were found to be at a fourfold risk of developing borderline personality traits (
Zelkowitz et al. 2001). In addition, adolescents who experienced sexual abuse and parental antipathy in combination with others forms of maltreatment were found to manifest significantly more borderline personality features (
Bégin et al. 2017).
Theoretical and empirical models suggest an interaction between genetic and environmental factors in the etiology of BPD (
Bornovalova et al. 2009;
Crowell et al. 2014;
Joyce et al. 2003). Consistent with this, maltreatment (physical and psychological abuse) before the age of 10 was found to predict borderline personality traits in 12-year-olds, but genetic vulnerability (measured by psychiatric antecedents in the family) greatly increased the chances that maltreated children would develop borderline personality traits in adolescence (
Belsky et al. 2012). In line with this, temperament, which is assumed to be genetically based, has been shown to moderate the expression of borderline pathology in adolescents who experienced abuse and neglect (
Jovev et al. 2013).
Lyons-Ruth and colleagues (
Khoury et al. 2019;
Lyons-Ruth et al. 2013) identified maternal withdrawal and lack of responsiveness as being particularly important in predicting personality disorder and suggested that parental withdrawal may be even more devastating than parental hostility. While this may be a surprising conclusion given previous research highlighting the links between abuse and BPD, it may be that by withdrawing, the parent is failing to comfort the infant when he is distressed, resulting in an escalation rather than regulation of distress and a permanent hyperactivation of the attachment system. This finding may be particularly relevant for understanding the difficulties in affect dysregulation, dependency, and abandonment anxiety commonly experienced by individuals with BPD.
Some studies relevant to understanding etiological factors involve mothers diagnosed with BPD. For instance, mothers who showed deficits in validating their teenager’s opinions, were hostile and overpersonalized disagreements, and pushed their teens to agree without attempting to offer satisfying justifications had difficulty supporting their adolescents’ autonomy and minimized their children’s relatedness (
Frankel-Waldheter et al. 2015). They also overpersonalized situations and thereby reduced their child’s independent thought and action and attempted to keep them close by using hostile means to inhibit the child’s development of relatedness. Mothers with BPD were similar to control mothers in several of their approaches to interacting with their child but showed an increase in role reversal. If, for example, they were expecting their child to satisfy their needs, they could not take on the role of the nurturing parent, especially one who could foster secure attachment and subsequent autonomy. The object relations model that is incorporated into the Transference-Focused Psychotherapy for Adolescents (TFP-A) approach (to be described in
Chapter 2) offers the potential to utilize these self-other interaction patterns in clinically useful ways when appropriate to the individual’s treatment. The various features just reviewed can contribute to young persons’ developing hypersensitivity to others’ reactions and deficits in awareness and appreciation of others’ thoughts, and when the situation overwhelms their adaptive ability, they develop primitive defenses. These narrow and rigid coping skills can limit their capacity to adapt in a mature fashion.
Maltreatment and Pathological Narcissism
From the point of view of psychological structure, narcissistic PD and BPD are closely related. However, there are still important gaps in our knowledge regarding the developmental precursors of grandiose narcissism and vulnerable narcissism (
Campbell and Miller 2011;
Dickinson and Pincus 2003). Some studies report positive relationships between maltreatment and both grandiose and vulnerable narcissism (
Khoury et al. 2019;
Lyons-Ruth et al. 2013), while others suggest that parental coldness or intrusive behaviors contribute to the development of vulnerable narcissism, but not grandiose narcissism (
Crowe et al. 2016). In another study, negative parenting such as inconsistent discipline and lack of supervision was found to be linked to vulnerable narcissism in 16- to 17-year-olds, while grandiose narcissism was associated more with parental investment and positive reinforcement (
Mechanic and Barry 2015). Consistent with this, overvaluation, and not lack of warmth and affection, was found to predict grandiose narcissism in children ages 7–12 (
Brummelman et al. 2015). Furthermore, role reversal, or
parentification, in which the child is forced to adopt parental emotional roles and responsibilities (
Haxhe 2016), has also been linked to narcissism (
Jones and Wells 1996).
Attachment
Attachment styles are considered to be the outcome of multiple repeated affective interactions between mother and infant. In this way, attachment reflects the repeated balance of positive and negative interactions in which the mother either responded adequately to the infant’s affective communication and distress and acted as a secure base, helping the infant reestablish self-regulation, or responded in an intrusive, hostile, helpless, passive, or other noncontingent way, contributing to an escalation of distress or leaving the child in an intolerable state of heightened stress. Furthermore, studies of mother-infant affective communication, such as those of Daniel
Stern (1995) and others (
Beebe 1986;
Lyons-Ruth 1999), demonstrate how repeated patterns of dyadic affective communication form the foundation for later constellations of expectancies of how another is likely to respond to particular affects. Stern argued that these expectancies will have an impact on how the infant (and later the child and adult) expresses his or her affects in dyadic contexts. For example, the infant who repeatedly experiences his mother as becoming overly intrusive when he signals that he is distressed may come to express his distress with a mixture of frustration in anticipation of a mother who will try and engage him in another activity before he is regulated. His mother, and later his partner or significant others, seeing his look of frustration or anger when he feels distressed, may feel confused about whether he wants to be comforted, and may be hesitant to offer help for fear of being rejected.
Attachment organization has long-range predictive value and is especially relevant for understanding BPD and the development of emotional regulation, self and other representations, and accompanying interpersonal relations. For example, in a longitudinal study ranging from infancy to early adulthood, the relationship between attachment disorganization and BPD symptoms in early adulthood was explained (mediated) through the impact of attachment disorganization on self representation and the implications of this in turn for BPD symptoms (
Carlson et al. 2009).
Developmental challenges during adolescence involving investment in a social network—reliance on friends and partners rather than on family and parent—may be positively or negatively influenced by early attachment styles and self and other representations that are activated. For example, secure attachment styles and trust will facilitate the transitions associated with separation from the family and individuation when adolescents increasingly rely on their own friends and social networks for support. Adolescence may also provide ample opportunities to reexamine and reorganize representations of self and other and to address identity issues, as adolescents are confronted by the difficulties that insecure and disorganized attachment styles may contribute to in their interpersonal relationships.
In the therapeutic relationship, this pattern translates into a transference
1 in which the adolescent is reluctant to express a range of emotions, negative as well as good feelings, about a positive event that might have happened. Within our object relations model, this attachment pattern with the therapist can and should be jointly examined. Such an intervention leads the adolescent to reexamine his or her self-other constructions and the defenses that maintained them, and to use more mature cognitive strategies for self-reflection so as to consider alternative constructions of the adolescent’s self and of his or her reality.
Multiple Pathways to BPD in Adolescence
As is evident from the preceding section, adolescents and young people who are diagnosed with BPD are not a unitary group. Adolescents who have quite different symptom profiles according to DSM criteria may meet criteria for the diagnosis of BPD, as is the case with adults. Research instruments also reflect the diversity among these young people. For example, the BPRC scale utilized by
Belsky et al. (2012), derived from the Shedler-Westen Assessment Procedure–200 for Adolescents (SWAP-200-A) (
Westen et al. 2003), reflects three fundamental features of adult BPD (affective instability/dysregulation, impulsivity/behavioral dysregulation, disturbed relatedness/interpersonal dysfunction). Specific items of the BPRC scale include “easily jealous”; “falls for new friends intensely”; “changes friends constantly”; “fears he/she will be rejected or abandoned”; “feels others are out to get him/her”; “acts overly seductive or sexy”; “emotions spiral out of control, has extremes of rage, despair, excitement”; “cannot think when upset, becomes irrational”; “unable to sooth or comfort self”; “lacks stable image of self, changes goals/values”; “expresses emotions in an exaggerated dramatic ways”; “irritable, touchy, or quick to fly off the handle”; “angry and hostile”; and “engages in self-harm behavior.” It is not difficult to imagine that different adolescents with BPD can have different feature patterns and different developmental histories as well. Clinical experience also suggests the existence of qualitatively different groups of adolescents for whom the diagnosis of BPD is useful and appropriate.
Those adolescents who successfully navigate through the demands of adolescence, despite their ups and downs with parents and peers, demonstrate continuity of personality and convey the sense that they are building on their childhood selves rather than reinventing themselves. Even as they become more independent and develop views and values that are differentiated from those of their parents, they maintain their relationships with parents and family members. In addition, even when they develop views and values that differ and in some cases are in conflict with those of their parents, they maintain a certain capacity to see past these differences and appreciate positive aspects of the parents and are able to turn to their parents for help in times of need.
In contrast, the vast majority of adolescents who are diagnosed with BPD have childhood histories of long-standing and marked difficulties in affect and behavior regulation. One group whose symptoms frequently meet criteria for BPD during adolescence includes those who have shown relational disturbances involving inappropriate aggression directed toward others during childhood, such as ODD and CD.
Another group of adolescents with BPD may have a consistent pattern of inflexible and maladaptive reactions but do not display the problems of affect dysregulation and aggression associated with ODD or CD. These maladaptive qualities might be difficult to observe in situations that are structured, nonchallenging, or predictable. They are more likely to appear in periods of change and stress—for example, in the transition between middle school and high school; during activities that make greater interpersonal demands, such as making new friends and establishing a level of intimacy in a relationship; in situations involving challenges, competition, and the risk of failure and humiliation, such as taking tests, team sports, or performing publicly at school; or in situations that make new demands for autonomy, such as attending a sleepover, finding a job, and performing a job in the absence of supervision. Consistent maladaptive reactions to these activities and situations may be indicative of disturbances in characteristic defenses and coping mechanisms, and these underlying difficulties will become more and more evident at each developmental period so that there may be definite but less flamboyant evolution toward a PD when the individual is faced with the inevitable challenges of adolescence to separate, become more independent, and establish social and intimate relationships outside of the security of the family.
Adolescents who develop eating disorders or engage in self-harm frequently have such apparently unremarkable childhood histories. At most, they may have been somewhat sensitive, dependent, submissive, and obsessive as children. It is fair to conclude, though, that these adolescents, like those with ODD and CD, have long-standing difficulties that are entrenched in their personalities but that become increasingly evident in the context of demands to become autonomous and take on increasing responsibilities and make decisions, while separating from parents and developing new intimate relationships.
There is another small group of adolescents without obvious childhood psychological problems or personality difficulties who may display an identity crisis that is difficult to distinguish from BPD in adolescence. These young people have difficulty preserving a sense of continuity and coherence when their capacities for adaptation are overwhelmed by the challenges that often accompany adolescence. The longitudinal trajectory study of self-esteem by
Birkeland et al. (2012) included some individuals that may be representative of this group of adolescents. The self-esteem of a group representing approximately 7% of adolescents in the study was described by a U-shaped trajectory in which initially good self-esteem decreased markedly between ages 14 and 18 years and reached its lowest level in late adolescence, before improving during the next 5 years. The adolescents in this group may have had preexisting fragilities that compromised their capacity to adapt to change and their difficulties during adolescence, despite the apparently improved self-esteem, may have led to some kind of scarring, because their global self-esteem at age 30 was significantly lower than that of individuals with consistently high self-esteem during adolescence. They also presented with significantly higher levels of depression. This would suggest that adolescents who experience identity crisis or sharp decreases in self-esteem may also warrant intervention and potentially derive significant long-term benefits from therapy. The therapeutic intervention needs to be sensitive to the presence of a PD and not just view the difficulties as manifestations of a depressive disorder.
Another group comprises those who are victims of sexual abuse. Sexual abuse just before or at the beginning of adolescence may be particularly destabilizing. It can be the final blow for girls who had shown resilient personality characteristics and who were able to continue to function well at school and invest in friendships despite parental neglect, substance abuse, psychological problems, and immaturity. Although many PTSD symptoms can be expected to resolve in due course, sexual abuse may interfere with the capacity of adolescent girls to form intimate relationships and to develop trust in partners.
Adolescents who engage in increased risk taking, especially when drugs, alcohol, and sex are involved, may also be at higher risk of presenting with identity crisis and lowered self-esteem when they develop addictions, experience trauma, or become overwhelmed when their behaviors take them down paths for which they are unprepared.
Another group that may be particularly at risk are those adolescents who are confronted with the task of assuming a sexual identity that is not culturally desirable and involves the possibility of being alienated from peers and family.
Breakdown and identity diffusion that presents like BPD can be seen in hypersensitive adolescents when there is parental separation, especially when the separation is accompanied by conflict and geographical moves to distant cities that make parents less available. These circumstances can also be associated with a loss of friends, and the adolescent is challenged to integrate into a new social circle and adapt to a new academic environment. This may be associated with the onset of suicidal/self-harming behavior.
Parental mental illness and chaotic family environments, in which parents respond with inappropriate physical aggression to adolescent self-assertion and bids for separation, can also lead to breakdowns that result in identity diffusion in sensitive or vulnerable adolescents.
Adolescence as a Critical Period for Development of Personality Disorders
BPD symptoms typically appear in adolescence and peak from ages 14 to 17 years and then gradually decline (
Arens et al. 2013;
Bornovalova et al. 2009;
Chanen and Kaess 2012). Even when symptoms like impulsivity tend to decline, underlying negative affect and feelings of emptiness are more likely to persist (
Meares et al. 2011). Also, according to the Children in the Community (CIC) Study (
Cohen et al. 2005), high symptom levels of any personality disorder in adolescence have negative repercussions on functioning over the subsequent 10–20 years, and these repercussions are often more serious or pervasive than those associated with Axis I disorders. The same study also found that some youth manifest an increase in PD symptoms from mid-adolescence to early adulthood and that symptoms of BPD are the strongest predictors of later PD. Data from the CIC Study were used to investigate the relationship between early BPD symptoms and subsequent psychosocial functioning. They demonstrated an association of early BPD symptoms with less productive adult role functioning, lower educational attainment and occupational status in middle adulthood, an adverse effect on relationship quality, and lower adult life satisfaction (
Winograd et al. 2008). Elevated BPD symptoms in adolescence have been shown to be an independent risk factor for substance use disorders during early adulthood (
Cohen et al. 2007).
There are many reasons why BPD symptoms may first become apparent in adolescence. Most apparent is that this pattern is attributable to developmental changes in brain structures, with the decline in BPD symptoms associated with maturation of control mechanisms (
Powers and Casey 2015). Also physical, cognitive, and social changes may contribute to its onset during that period.
Neurobiological Changes in Adolescence
Transformations occur across adolescence and extend into early adulthood. They include brain changes in the frontal lobe regions that contribute to the marked growth in higher-order abstract reasoning, problem solving, decision making, and mentalization. Yet, despite the rapid development of these abilities, adolescents seem more emotionally reactive and vulnerable to making bad decisions, and go against their better judgment and engage in risky behaviors, especially when under the influence of emotions and peers.
Casey and Jones (2010) have proposed an “imbalance model” to describe adolescent brain development, whereby the limbic system is functionally mature at a time when the prefrontal systems are still developing, leaving the adolescent more susceptible to the influence of the reward-sensitive limbic system. Adults too, as described by a dual processing model, use reflexive or automatic, intuitive, affect-driven heuristic processes, mediated by subcortical systems, although they are capable of more reflective, controlled rational processes subserved by the prefrontal cortex (PFC) (
Evans et al. 2002;
Galvan 2012;
Reyna 2004;
Reyna and Farley 2006;
Romer et al. 2017). In the dual systems model, decisions result from an interaction between more thoughtful processes and more experienced-based, affective, heuristic, and motivational processes (
Damasio 1994a,
1994b;
Epstein 1994;
Evans 2008;
Lerner and Keltner 2000; M.D.
Lieberman 2000;
Loewenstein et al. 2001;
Schneider and Caffray 2012,
Stanovich and West 2000). BPD pathology in adults and adolescents, including dysregulated negative affect, impulsive and aggressive behavior, and interpersonal difficulties, can be seen as derived from deficits in these reflective and executive control processes, coupled with a biased reflexive process in which there is an automatic hypersensitivity to negative social cues (
Koenigsberg et al. 2009), an expectation of untrustworthiness (
King-Casas et al. 2008), and increased negative affect (
Sadikaj et al. 2010).
Casey (2015) provides a more detailed developmental analysis of the changing interrelationship during adolescence among the PFC, ventral striatum, and the amygdala that offers a framework for understanding several of the behavioral features that seem to characterize adolescents. The PFC, as a mediator of reasoning and behavioral regulation processes, can suppress output for both the ventral striatum and amygdala. As the functioning of these areas becomes more coordinated over time, there is a greater integration of cognitive and emotional processes, which results in increased motivated and goal-oriented actions. However, the development of the PFC extends through adolescence, whereas the ventral striatum and amygdala reach more mature functioning at an earlier time; therefore, behaviors that require greater self-control (e.g., focusing attention on relevant information and withdrawing attention from irrelevant but possibly interesting information) suffer, compared with later adolescence and early adulthood. As an illustration of a function that would have clinical importance, the ventral striatum is more responsive to larger rewards than to smaller ones in mid-adolescence when compared with childhood and adulthood, and as a result 15-year-olds are more likely than adults to make risky gambles for immediate reward feedback (
Casey 2015). In general, adolescents are more impulsive responders to positive cues than are children or adults.
The presence of peers is associated with even riskier decision making (
Steinberg 2008). It is not that adolescents lack the knowledge to make the wiser choice; rather, adolescents appear to be especially sensitive to incentives (e.g., money, peer acceptance) and contexts, such as the presence of peers, that seem to heighten motivational states. Therefore, the adolescent will be less likely to suppress inappropriate actions and desires when the context includes salient cues because their capacity for response regulation is still incompletely developed.
Casey (2015) also noted that drug use may interact with the dopamine system and heighten responsivity in the ventral striatum, thereby strengthening the reward properties of the drug during a phase when the PFC and its associated control mechanisms are still immature.
Casey (2015) indicated that the changes she described in adolescence are found across species, and so it may be that the movement from dependence to autonomy is greatly aided by the increase in novelty seeking and peer interactions found among adolescents. The increased value of incentives and the wish to obtain more resources and new sexual experiences may enhance the movement out of the home and support the separation-individuation process. She noted that adolescence may be thought of as “a period of thrills and fears,” but many of our patients have not found the balance to master the fear by using judgment in a manner that would also promote safe exploration and autonomy. Of clinical interest, during this period of neurological imbalance, adolescents show diminished fear extinction, which suggests that exposure procedures might be less effective, and, as
Casey (2015) pointed out, there is some evidence for reduced treatment efficacy of cognitive-behavioral therapy in adolescence compared with children and adults. One possible mechanism in TFP-A interventions that could contribute to their effectiveness is support of ego functioning, providing a form of scaffolding that helps the adolescent compensate for cognitive immaturities.
Puberty
Puberty takes several years from onset to completion, and hormonal changes precede the observable physical changes. The process typically begins earlier in girls than in boys. The physical changes include the development of secondary sexual characteristics (e.g., pubic hair growth; breast and penile development) and a growth spurt; changes in body shape, size, and composition; and menarche/spermarche. These observable changes signal that the developing young person is entering a new phase in life, and this transition brings with it new expectations and reactions from parents, siblings, and peers, aside from reactions of the young person whenever he or she looks in a mirror. What is often clinically relevant is the degree of synchrony between the adolescent’s experience of physical and personal maturity and the granting of social maturity and autonomy by the family and society (
Rudolph 2014).
Pubertal status can be associated with features of psychopathology such as an increase in anxiety, depression, antisocial behavior, and problematic substance use during the pubertal transition (Rudolph 2014). There is also a heightened reactivity to emotions and information with emotional connotations during this period, and cognitive control is more readily undermined by salient emotional information and incentives (
Silk et al. 2009). Early maturation may serve to exacerbate already existing individual differences in vulnerability to psychopathology, and of value to the clinician’s understanding is that assessment in early adolescence (11–13 years) can be a better predictor of adult functioning than assessment of similar behaviors made in younger children or in middle adolescence (
Livson and Peskin 1967). Perhaps individual differences become magnified across periods of change, such as during pubertal transition, as these periods of biological transformation also evoke demands for social and intrapsychic change, and the greater inter-individual variability during such periods of change allows for greater predictive utility.
Many studies have focused on the timing of pubertal changes—are they early or late occurring, and does the timing have a differential effect on boys and girls? It appears that there are advantages and disadvantages to early and late maturing for boys and girls. Although it is usually thought that early maturation is advantageous for boys, conferring heightened social status and respect, it is also the case that early-maturing boys may feel out of sync with their age and grade mates and are thus more likely to enter into social relations with older boys who are more likely to engage in risky and norm-breaking behavior at a time when the early maturers are not emotionally or cognitively prepared to handle it. For early-maturing girls, risky, pseudo-mature sexual behavior may also occur. They, and later-maturing boys, may be more likely to experience internalizing problems and depressive symptoms. For girls, high-for-age hormonal levels are associated with depressive features, including sadness.
Sexuality, Identity, and Early Experience
Changes in sexuality and aggression associated with puberty demand integration into personal identity and necessitate learning how to express these drives in interpersonal and intimate contexts. Sexual attraction and desire for someone pulls and pushes adolescents to develop intimate experiences and relationships outside of the family, forcing them to face an ensuing sense of elation combined with loss. The change from being physically childlike to accepting a new physical self that is at once experiencing new sensations and functional capacities and also capable of feeling and eliciting sexual desire may be challenging for any adolescent to integrate at the level of identity. The continuous body changes will, alternatively and in waves, either confuse or confirm the adolescent’s identity feelings. At the same time the relationship with family and peers may become sexualized. Sexual desire for another on the one hand and being desired on the other hand propel adolescents into exploring new experiences involving physical and emotional intimacy with others outside the family. In an entirely novel setting they have to negotiate the experience of being close to someone else when there are the threats of rejection as well as regression and fusion. Sexual experiences also challenge the adolescent to develop control of aggression and impulsivity; consider the needs, desires, and feelings of another in an intimate physical context; and discover new ways of communication to manage the vulnerability of each of the adolescent partners. Masturbatory activity continues to offer an outlet for self-soothing and also provides a means for learning and practicing the expression and control of sexual action.
Preadolescent experiences impact the young person’s integration of sexual and aggressive demands into a developing sense of self. For adolescents who have experienced sexual abuse or sexualization, or for adolescents who have conflicts or confusion regarding their sexual identity, the unavoidable confrontation with sexuality during adolescence may strain and threaten the cohesion and continuity of their sense of self.
For adolescents with extremely difficult and painful attachment histories whose early experiences of intimacy have been marked by fear, hostility, frustration, intrusiveness, rejection, abandonment, and helplessness, sexual intimacy is likely to re-evoke these deep anxieties with associated fears around dependency, regressive fusion, aggression, shame, and abandonment. So, as they experience a novel, heretofore unique, extrafamilial relationship, the desire for closeness and their first sexual experiences may, in vulnerable adolescents, trigger deep anxieties and crises that challenge their fragile sense of self and bring to the forefront weaknesses in the self that may not have been so evident during childhood.
Adolescents with histories of sexual abuse may also feel particularly vulnerable. The traumatic experiences and feelings related to the abuse intrude into their first sexual relationship, and this makes it more complex and challenging for them to integrate this experience into their identities. In addition, vulnerable adolescents with personality pathology may fail to assert and protect themselves sufficiently and are less likely to anticipate and avoid potentially dangerous interpersonal situations; they also may consume drugs and alcohol so that they are more likely to have nonconsensual and traumatic sexual experiences in adolescence that may destabilize them further.
Adolescents who have been sexualized prematurely, whether through sexualized mother-infant or toddler contact or through lack of parental boundaries or controls around witnessing sexual intercourse, may also be particularly vulnerable and may require additional help to understand these experiences and their consequences in order to better integrate into their identity their engagement in sexual behavior and promiscuity.
The issues of sexual orientation and gender identity can provoke anxiety and be a major source of deep conflict in adolescence while straining the capacity to integrate aspects of the self that may be experienced as alien and dangerous, and may lead to breakdown and serious suicide attempts. While adolescence is a period when the boundaries between friendship and sexual attraction may blur, confronting the adolescent with their potential for bisexual or same-gender sexual attraction, for the vast majority of adolescents this is a passing phase that does not result in undue anxiety. However, for adolescents who are confronted by the homosexual nature of their attractions, it is frequently challenging to integrate this awareness into their identities. This is further complicated when it is their first sexual experience, potentially testing their personality’s cohesion and capacity for flexibility. It also tests their sense of independence, to see if they will experience and then can survive parental, familial, peer, and societal rejection or disappointment. For adolescents without a strong sense of self, the anxieties and conflicts stirred up by discovering a sexual orientation not accepted by their family or community could lead to personality crisis.
Cognitive Changes
It is also interesting to consider the possible contribution of the development of abstract thinking during adolescence, which of course is also subject to neurodevelopmental and genetic contributions. Therefore, we might expect that for those who show a continuance of BPD symptoms over the adolescence period—namely, those for whom it is not a stage-specific phenomenon (cf.
Moffitt 1993a,
1993b)—more is involved than the contribution of advanced abstract thinking, and the crisis is not normative. In the normative condition, it is less likely that PD features such as true splitting would be seen.
Summary
BPD, as well as narcissistic, antisocial, and avoidant PDs, can be diagnosed in adolescence. A reluctance to diagnose BPD in young people may increases the risk of more severe outcomes because it reduces access to psychotherapy. Psychotherapy that addresses personality disturbance is essential to help adolescents resume engagement with developmental challenges in a way that can facilitate personality development. There are likely multiple pathways to BPD, consistent with the concept of equifinality. In addition, there may be several “subtypes” of adolescent BPD as a function of varied developmental histories, and there may be many experiences before and during adolescence that can compromise the process of identity formation.
In a TFP-A model, difficulties at the level of identity formation constitute a core feature of BPD. Contributions to the development of BPD come from genetic, temperamental, and experiential forces in interaction with one another. So, although these pathways are somewhat nonspecific, perhaps the common factor is that they impact parent-child interactions and the nature of the attachment relationship and other ensuing relationships. This, in turn, impacts the development of the growing child’s sense of self, which together with the developmental demands of adolescence, shapes identity formation, the bedrock of personality and PDs.
In the next chapter, we present a contemporary psychodynamic conceptualization of PDs in adolescence as well as structural changes and developmental challenges that are hypothesized to be disrupted and sidetracked because of the core problem of PDs, which is identity diffusion.