It is largely assumed that adolescence is a period of change and turmoil. This might be the reason that it is confusing for clinicians to consider diagnosing a personality disorder during a time of identity questioning and consolidation. This review aims to clarify the question in order to work more efficiently with those patients in whom the affective instability and the identity disturbance surpass normal adolescent levels, and might lead to increased morbidity and mortality if not treated or treated inadequately.

Does borderline personality disorder exist in adolescents?

Many studies suggest that we can reliably diagnose borderline personality disorder (BPD) in adolescents1 and the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) agrees with this; it states that:

Personality disorder categories may be applied to children or 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 an episode of an Axis I disorder... To diagnose a personality disorder in an individual under 18 years of age, the features must have been present for at least one year.2

Hence, according to DSM-IV-TR, when personality traits are inflexible, maladaptive, and chronic, and cause significant functional impairment or subjective distress, they constitute a personality disorder, regardless of age.1 The DSM also mentions that the onset is often traced to adolescence, which is corroborated by the literature.2-5 The same criteria as for adults are used. It is being more and more demonstrated that the diagnostic criteria for BPD are as reliable, valid, and stable in adolescence as they are in adulthood.6-9 BPD is estimated to affect between 0.9% and 3% of teenagers in the community,6 which is equivalent to the prevalence in adults.10

Miller et al point out that studies indicate that, while there is a legitimate subgroup of severely affected adolescents for whom the diagnosis remains stable over time, there appears to be a less severe subgroup that moves in and out of the diagnosis.1

The literature suggests that individual symptom presentation is likely to vary over time, but that one can make an accurate diagnosis by considering core dysfunctional areas of BPD (identity disturbance, affective instability, relationship difficulties, impulsivity).1 In the same vein, Chanen et al demonstrated that the stability of the categorical BPD diagnosis was rather low, but that its stability measured dimensionally was considerably higher.9 Indeed, we can understand that a dichotomous diagnosis might make it easy for an adolescent to switch from being just above the threshold to a subclinical level of symptoms, while a dimensional approach allows variations in the level of symptoms. Miller et al also mention that a dimensional approach may better account for the developmental variability and heterogeneity found in adolescents.1

Clinicians tend to be reluctant to diagnose BPD in adolescents, saying that adolescence is a period of transition that can be marked by turmoil, and that this should not be called a personality disorder. Also, as these disorders are chronic, clinicians prefer to wait before making such a conclusion. It is true that moodiness and some degree of impulsive behavior and risk-taking are common in adolescents, but most of them are not seriously troubled. Some clinicians also fear that labeling the teenager could be prejudicial.

Though we should avoid pathologizing a normal behavior, diagnosing BPD in adolescents when clinically appropriate has important advantages. Less emphasis could be put on psychopharmacology, and the use of psychotherapy could be enhanced, as there is stronger evidence for its efficacy.11 Making the diagnosis earlier also suggests an early intervention and thus prevention of crystallization of behaviors that can have severe consequences on functioning. As BPD traits are malleable and flexible in young people,12 it means this is a good period to try an intervention. Indeed, the evidence supports the use of early intervention programs for BPD in youth.6

Also, although BPD traits in adolescents tend to attenuate over time, this does not mean they recover. According to the CIC Study,13 high symptom levels of any personality disorder in adolescence have negative repercussions on functioning over the subsequent 10 to 20 years, and these repercussions are often more serious or pervasive than those associated with Axis I disorders. The same study also found that symptoms of BPD were 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 and less productive adult role functioning, a lower educational attainment and occupational status in middle adulthood; an adverse effect on relationship quality, and a lower adult life satisfaction.14 Elevated BPD symptoms in adolescence have been shown to be an independent risk factor for substance-use disorders during early adulthood.15 These are all further arguments to advocate for the development of accessible intervention programs for youth with BPD symptoms. Besides, the symptoms have been shown to peak around ages 14 to 17, making it a critical risk period and a good point in time to intervene and modify the trajectory of the disorder towards a better functioning.16

Appropriate management of BPD symptoms in the right settings would also alleviate the burden on the health system. Patients with BPD symptoms and no treatment plan may consult at the ER repeatedly, at every crisis. In the absence of a treatment team to be directed to, they might also be hospitalized, with all the possible iatrogenic effects that could be envisioned, and a deleterious effect on functioning caused by suicidal threat or acting-out behaviors.


Having an idea of the origin of BPD aids in considering it when an adolescent consults with suggestive symptoms. It is believed that BPD results from the interaction between temperament and parenting failures. Fonagy and Bateman postulated17 that constitutional vulnerabilities coupled with parental underinvolvement or neglect result in deficits in the child's ability to regulate emotions through mentalization. The invalidating environment described by Linehan18 may also interfere with attachment and the learning of emotion regulation strategies. The temperamental factors might be emotional reactivity or difficulty being soothed, which are challenging for any parent, and especially for those who share these genetic predispositions.

Studies investigating the type of attachment of BPD patients largely conclude that there is a strong association between BPD and insecure (mainly preoccupied) attachment.19,20 Preoccupation is characterized by affective instability and unsteady representations of attachment figures. As a result, patients expect that they can not trust others to be available to support them.

Factors identified as predictors or risk factors for BPD in adolescents include history of disrupted attachment, maternal neglect, maternal rejection, grossly inappropriate parental behavior, number of mother and father surrogates, physical abuse, sexual abuse, and parental loss.21,22 These are all supportive of an insecure attachment etiological model. In their review, Chanen and Kaess add low socioeconomic status to childhood abuse and neglect, and problematic family environment, as significant risk factors for personality pathology, especially BPD.22

The results of a large prospective study in UK suggest that inherited and environmental risk factors make independent and interactive contributions to borderline etiology, supporting the current models of diathesis-stress theories, pointing to an interaction between genetic vulnerability and harsh treatment in the family.23 Borderline characteristics at age f 2 were more frequent in children who had exhibited poor cognitive function, impulsivity, and more behavioral and emotional problems at age 5 years, but also in those who were exposed to harsh treatment. These all become higher risk factors in the presence of each other and also when there is a family history of psychiatric illness.23

Clinical manifestations

The disorder's first manifestations typically arise during adolescence or young adulthood.13

As noted earlier, the DSM-IV-TR criteria2 are the same as for adults. It is a “pervasive pattern of instability of interpersonal relationships, self-image, and affects, and marked impulsivity beginning by early adulthood and present in a variety of contexts.” It is indicated by five (or more) of the criteria. The first criterion describes frantic efforts to avoid real or imagined abandonment. As they tend to have insecure attachment—mainly unresolved, preoccupied, or fearful19—patients with BPD expect other people cannot be trusted and will not be available for support. In a study about diagnostic efficiency of BPD criteria in adolescents compared with adults, the abandonment fears were found to be the best inclusion criteria for adolescents.24 In this study, patients had an 85% chance of meeting the full diagnostic criteria when they endorsed the abandonment fears. On a regular basis, we hear these patients tell us that the worst thing that could happen to them would be to be left alone. At its extreme, this symptom can lead young girls to do such things as undressing in front of a Web cam, or agreeing to prostitution in order not to lose their boyfriends. On the other hand, their fear of being abandoned is so great that in some circumstances, they break bonds or ruin their relationships in anticipation that they might be rejected.

The second criterion describes the intense and unstable relationships characterized by alternating between extremes of idealization and devaluation. Anyone likely to take care of an adolescent with BPD—like a teacher, schoolmate, or a therapist—is very soon of great importance and he or she is being idealized for his or her virtues and capacities. However when the patient unfortunately becomes disappointed, which happens at some point given the great expectations and the extreme sensitivity to feeling of rejection, there is a rapid shift to a devalued position.

The third criterion introduces the identity disturbance, a markedly and persistently unstable self-image or sense of self. Their perceptions of themselves, their values, their friends, and even their sexual identity can change dramatically. Questioning about one's identity is of course normal in adolescence, but it is the marked and persistent character of the instability that distinguishes normal from pathological. In BPD, confusion and changes are out of proportion. Westen et al assessed the potential manifestations of identity disturbance in adolescence, and they concluded that the items most distinctively associated with BPD describe feelings of emptiness, fluctuations in self-perception, and dependency on specific relationships to maintain a sense of identity.25

Criterion 4 concerns self -damaging impulsivity in at least two areas. We can often see these youngsters either abuse drugs, drive recklessly, engage in dangerous sexual practices, or have bulimic episodes, for example, but beyond the level of normal experimentation in adolescence.

These patients recurrently demonstrate suicidal behavior, gestures, or threats, or self-mutilating behavior (criterion 5). This is often what first brings them to clinical attention, as they are taken to the emergency department for these threats or gestures. Miller26 mentions studies stating that interpersonal conflicts and separations are the most common precipitants of adolescent suicide. He points out examples derived from different studies: breakups of romantic relationships, disciplinary crisis or legal problems, humiliation and arguments, which are stressors identified in attempted and completed suicides of youth.

Self-mutilation must be distinguished from suicidal attempts, as there is no intent to die in the former. Indeed, in the literature, it is widely called “non-suicidal self-injury” (NSSI). It generally begins in early adolescence.27 Zanarini et al reported that 32.8% of BPD self-injurers began before age 12, as 30.2% began as adolescents and 37% began as adults.28 Jacobson et al29 point out that the explanations of NSSI remain mostly theoretical, including psychodynamic, behavioral, and emotion-regulation models. They state that the emotion-regulation model has received the most empirical support. Indeed, the patients do feel relieved after the act. They might say it distracts them from their suffering, it allows them to vent their anger, it stops derealization, it makes them regain a sense of control, or it is self -punishment. The precipitant is most often abandonment, real or perceived, or a separation.

When assessing for NSSI with an adolescent, one needs to inquire about what is going on in the peer group, as cutting is susceptible to social contagion. It can be learned from friends (or social networks and other media) and it can be normalized or even valued among them; the teen becomes part of a “community of suffering.”

While being different from a suicide attempt, self-injury is still a risk factor for suicide, as are substance use; childhood sexual and physical abuse, neglect, losses (particularly interpersonal), psychiatric comorbidity, struggling with sexual orientation issues, and parental mental disorders.30,31 Adolescents being susceptible to suggestion and contagion, media coverage of suicides or a suicide in their community also increases the risk, specifically for adolescents.

Criterion 6 describes affective instability due to a marked reactivity of mood (eg, intense episodic dysphoria, irritability, or anxiety usually lasting a few hours and only rarely more than a few days). We are normally able to find the precipitant, which can appear minor from an external point of view but is experienced intensely. Indeed, the family or friends will often not share the patient's perception of the circumstances. The shifts seem exaggerated and unpredictable.

We find in criterion 7 chronic feelings of emptiness. The study by Becker et al24 did not support the observation by Pinto et al32 that emptiness or boredom was among the best discriminators of BPD in adolescents.

The inappropriate, intense, and hard-to-control anger of criterion 8 is regularly expressed when the patient feels neglected or abandoned; hence the therapist might be targeted at some point, or at least witness it. In the study by Becker et al24 anger was found to be the best exclusion criterion for BPD in adolescents.

The transient paranoia or dissociative symptoms of criterion 9 arise in highly stressful situations. The patient might describe feeling detached from his or her body or “like in a dream” (depersonalization, derealization). In the extreme, this can take the form of brief psychotic-like episodes.

Even if absent from diagnostic criteria, splitting deserves to be mentioned, as it is widely used by BPD patients, especially teenagers, who tend not to tolerate ambiguity or grey zones. They-or their parents-often note that they think and act in “all black or all white” way.

Many patients (and their parents) also describe themselves as moody and sensitive children.33

A study demonstrated that the antecedents of adolescent personality disorder could be traced to 10 years earlier in the form of childhood emotional and behavioral problems. Conduct problems were predictive of all three clusters, as depressive symptoms were associated with cluster B.34 The early temperament differences and early-onset mental state or behavioral problems are confirmed in a later review.22 Oppositional defiant disorder and attention deficit-hyperactivity disorder are also pointed out as possible predictive factors of BPD in adolescents.35

Management of BPD in adolescents

First, one should establish the aims of the intervention, to avoid wearing it down with unrealistic expectations. An important goal should be to improve the psychosocial functioning, and decrease the BPD symptoms, suicide and self-harm being primary targets. Discussing the management of BPD in adolescents implies addressing psychotherapies, pharmacology, hospitalization, and family implication.


Dialectical behavior therapy (DBT)

DBT18,36,37 is an adaptation of cognitive behavioral therapy by Marsha Linehan, who uses the dialectical philosophy in her therapeutic interventions, by flexibly balancing and synthesizing acceptance and change. It refers to the fact that opposite constructs can both be true at the same time; “you are doing the best you can at the moment and you need to do better.” The core dialectic in DBT is accepting patients where they are in the moment and working to help them change.

The therapy, as initially used with adults, includes weekly individual sessions, skills-training group sessions, phone consultation available at all times with the therapist, and team consultation meetings. The skills taught are mindfulness, interpersonal effectiveness, emotion regulation, and distress tolerance. There is an important hierarchy of the treatment targets, with life-threatening behaviors being addressed in priority. They are followed by therapy-interfering behaviors and quality-of-life interfering behaviors. This constitutes the first stage of treatment; subsequent stages are described but have not yet been the focus of studies.

Miller and colleagues first adapted DBT for use with suicidal adolescents. The treatment length was decreased from 12 months to 16 weeks to increase the likelihood of commitment and also assuming that teens might not need the same length of therapy as adults. The number of skills taught was reduced to make learning easier given the new length of treatment. Family members were included in the weekly skills sessions and were offered intersession skills coaching to enhance generalization of skills. Family sessions can also be added to address specific family issues. The terminology of the handouts was adapted to teenagers. Finally, a fifth skills module was added; walking the middle path, to help patients and families with polarized ways of thinking, feeling, and interacting.

Clinical research suggests that DBT may be an effective treatment for adolescents with BPD features, as it has been associated with reductions in suicidal and non-suicidal self-injury, psychiatric hospitalizations, and other problems associated with BPD.

Mentalization-based treatment

Mentalization-based treatment (MBT) was the second psychotherapy technique developed specifically for BPD. Mentalization is the imaginative mental capacity to perceive and interpret human behavior in terms of intentional mental states (feelings, needs, desires, beliefs, and goals.38 It is believed that this understanding of others in terms of their thoughts and feelings is a developmental achievement dependant on the quality of attachment relationships (particularly early ones).38 The capacity to mentalize varies in relation to emotional and interpersonal context. According to Fonagy, the failure of mentalizing, in combination with profound disorganization of self-structure, may account for the core features of BPD.38 In fact, adolescents with BPD features were found to hypermentalize,39 defined as “over-interpretative mental state reasoning.”40 MBT aims to improve the patient's ability to understand his own and other's mental states (mentalizing) in the context of attachment relationships, which is demonstrated as helpful in both affective and behavioral aspects of BPD. Concretely, this is done using weekly individual sessions and group sessions. MBT has proven more effective than usual treatment in reducing self-harm and depression in adolescents. It reflected improvement in emergent BPD symptoms and traits.41

The HYPE clinic: an early intervention service for BPD

HYPE42 stands for “helping young people early.” The clinic is based in Melbourne and uses a team-based intervention model comprising time-limited cognitive analytic therapy (CAT) by Ryle, case management, and general psychiatric care. The goal is to offer treatment as early as possible in the course of BPD (in contrast to services working only with individuals with a severe disorder) with an intervention appropriate to the phase of the disorder and the developmental stage of the patient and his or her family. Meeting three DSM-IV-TR BPD criteria is enough to be included. CAT is the core of the therapeutic model, and has been demonstrated to be effective in reducing externalizing psychopathology in teenagers with subsyndromal or full-syndrome BPD.43 It integrates elements of psychoanalytic object relations theory and cognitive psychology by focusing on understanding the individual's problematic relationships patterns and the resulting thoughts, feelings, and behavioral responses. Routinely, 24 CAT sessions are offered with four post-therapy follow-ups. The patient also benefits from general psychiatric care for assessment and treatment of comorbidity and use of eventual pharmacotherapy, plus crisis team and occasional brief and goal-directed inpatient care. The HYPE program also engages families with psychoeducation and up to four sessions of family intervention. The HYPE intervention is supported by effectiveness data and can be adapted to existing services in other settings.42


There is very little empirical evidence supporting the use of pharmacotherapy with adolescents struggling with BPD. This discussion will be derived from what is suggested in adults and from our clinical experience with adolescents (the reader may refer to the article by Luis H. Ripoll [p 213] in this issue for a review of the pharmacologic treatment of BPD). In BPD, medication should only be used as an adjunct to a multidimensional psychosocial approach and its limitations should be made clear for the patient. If two different persons are involved as the psychotherapist and the prescribing doctor, communication is very important. The pharmacological treatment will be symptom-oriented and will address impulsivity, affective instability, suicidal behaviors, and non-suicidal self-injury. No medication has received an official indication in the treatment of BPD, and long-term use of pharmacotherapy has not been studied in BPD. A good strategy could be to maintain a medication that works until psychotherapy has led to the development of new strategies.

Selective serotonin reuptake inhibitors

In BPD, most studies suggest that selective serotonin reuptake inhibitors (SSRIs) are most effective in reducing anger and impulsive symptoms; a reduction in mood swings is also mentioned.44,45 Other antidepressants are also studied (tricyclics and MAO inhibitors) but SSRIs are preferred, since they are better tolerated in regard to side effects and also they appear safer in case of overdose, which is a particular concern with BPD patients. Bulimia nervosa, a form of behavioral dyscontrol that usually develops in adolescents, is frequently associated with BPD and tends to respond to SSRIs.46 Regarding antidepressants, which are widely prescribed to patients with BPD, one has to keep in mind that they do not treat the disorder and do not produce remission.44


The literature concerning antipsychotics in BPD is sparse and the samples are small.44,45 Cognitive-perceptual symptoms (reference and paranoid ideas, illusions and hallucinations, derealization) arise mainly in periods of intense emotional stress. Because they have a rapid effect, antipsychotics can be used on a short-term basis for crisis periods. We tend to prefer atypical neuroleptics over typical ones because of their side-effect profile; however, even if they produce much fewer extrapyramidal symptoms, we still have to consider their potential to induce a metabolic syndrome and weight gain. Longer-term low-dose antipsychotics can be used as an adjunct to anger management, but only if an alternative with a better side-effect profile, like an antidepressant, has failed.

Mood stabilizers

Adult meta-analyses have shown that mood stabilizers as a class reduce anger and impulsivity somewhat, and may have some effect on affective instability and depression.47 However, evidence for individual medications comes from only one or two studies each47 and the risk of overdose may be great.


A 2004 article stated:

Hospitalization is of unproven value for suicide prevention and can often produce negative effects. Day treatment is an evidence-based alternative to full admission. Chronic suicidality can best be managed in an outpatient setting.48

Specialists criticized the American Psychiatric Association guidelines49,50 when they were published, as they recommended hospitalization whenever patients were suicidal. When facing self-destructive behaviors, clinicians can be tempted to use hospitalization but it may prove useless, and even damaging. First, the behavior will very likely have relieved the crisis and the message given to the patient that he or she is not able to get through this crisis without the hospital would be invalidating. Paris states that “hospitalizations make the therapy almost impossible as you cannot help people learn to cope with life or get a life if they are living on a psychiatric ward.“50 Repeated hospitalizations seriously hinder the adolescent's normal functioning. Things go quickly in young patients' lives, and being away can rapidly degrade their social network, just as not attending school will likely delay them academically, which may increase pressure and stress. Being in hospital will prevent dealing with interpersonal conflicts or misunderstandings, which are often the trigger of the gesture, and then create an overrating of the problem by the youngster. Hospitalization may also reinforce pathological behaviors and make the patient worse.

There are exceptions we can make to this rule of not hospitalizing. We should consider it for very brief periods of intense distress that could lead to a suicidal gesture. Paris also points that micropsychotic episodes might be treated with medications in a hospital setting, and near-lethal suicide attempts can be briefly admitted in order to re-evaluate the treatment plan.50

Not hospitalizing does not mean that we should ignore suicidal behaviors—which tend to provoke a ”boy who cried wolf“ scenario in families and doctors—as suicide rate is estimated at 10% in BPD,49,51 and suicidal ideas are a sign of distress. The therapist can acknowledge the patient's suffering and his or her need for relief of dysphoria by working with him or her to develop alternative strategies to self-harm.

Family involvement

BPD symptoms in an adolescent have a tremendous impact on his or her family; the greatest effect is suggested to be on their emotional health.52 The same study also found that a majority of parents reported physical health problems and marital difficulties. In the same study of 233 female offspring meeting strict criteria for BPD, symptoms correlated with intensity of parental burden were acting-out behavior, property destruction, delusional symptoms, and hallucinatory symptoms.52 This suffering of the family has to be validated. Parents need to be told that their anger, guilt, or anxiety are normal and can be controlled to avoid an exacerbation of their child's pathological behaviors. The therapist has to build on their strengths and avoid blaming them.

Not only is the family a valued ally as a source of information and the primary support of the adolescent, it is essential in the management of a teenager with a BPD.

Indeed, an interview with the family enlightens the therapist on the relational mode of the patient and allows targeted interventions.

Family work is important because the home environment often plays a major role in the adolescent's behavior.53 Parents can help their child to use the skills learned in therapy and even use the same skills themselves. They may also learn to modify the way they respond to the patient's pathological behaviors.53 Miller suggests they be partners rather than targets in treatment.53 Also, assuming that the environment influences the genetic vulnerability in the expression of the disorder, an intervention at the family level might be protective.16

Psychoeducation is the basis of the necessary intervention with the family. They need information about BPD; its symptoms, what we know about its etiology, recommended treatments. Parents shall be taught about effective communication, behavior management, and problem-resolution strategies. While being validated regarding how much the situation is worrisome and frustrating, they can also be told that they can remain optimistic since something can be done.

The therapist also has to be clear from the beginning with the patient and his or her family about confidentiality issues. Confidentiality shall be broken if the patient's safety is at stake, if there is a suicidal plan with an intent to act it out, a plan to seriously hurt oneself or someone else, or if there is a situation of physical or sexual abuse or neglect. Regarding self-mutilation, the DBT model proposes that we validate the parents' worry while telling them that we won't disclose every gesture unless it threatens life, or there is an uncontrollable escalation of the behavior. This will allow the adolescent to feel more comfortable to discuss his or her behaviors. When it becomes necessary to break confidentiality, the patient should be involved as much as possible in the process.


In the Harry Potter novels, Professor Dumbledore told Harry Potter that he could call the evil Voldemort by his real name instead of “He-who-must-not-be-named,” because not calling things by their real name just makes us more afraid of them. Avoiding stating that an adolescent has features of BPD when it is the case is burying one's head in the sand, and this can result in being inefficient in addressing the problem. It can result in the patient receiving inappropriate treatment, or no treatment, with the imaginable consequences on his or her functioning, even on his or her life, and also on the health system. By contributing to detecting BPD and becoming skilled in addressing it properly, we, as clinicians, can contribute to the improvement of these patients' quality of life and both short and long-term prognosis.