Early Onset

Zanarini et al⁹ have demonstrated that patients with BPD begin their first treatment, typically individual psychotherapy, at the age of 18, although symptoms are likely to start earlier. Her team has also demonstrated that over 30% of patients with BPD began self-harming when they were 12 years of age or less, with another 30% initiating self-harm between the ages of 13 and 17.¹⁰ As previously mentioned, DSM-5 permits the diagnosis of BPD in patients younger than 18 if symptoms persist for at least 1 year. Symptoms of BPD usually start prior to adulthood and the diagnosis can be made reliably.

There have been many different precursor syndromes proposed for the diagnosis of BPD,¹¹ but none have examined how often these patients will meet criteria in adulthood. Paris and colleagues (see Paris et al¹² and Guzder et al¹³) examined a syndrome called borderline pathology of childhood, which combines symptoms in affective, impulsive, and cognitive domains, modelling the criteria and diagnostic measure after the Diagnostic Interview for Borderlines, a semi-structured interview that identifies a more homogenous population than the DSM-5 criteria and is one of the most reliable diagnostic measures for BPD in all age groups.^14,15 These authors found that children from ages 7 to 12 who meet criteria for borderline pathology of childhood were more likely than a clinical comparison group to have experienced abuse and neglect, as well as to have neuropsychological findings suggesting executive dysfunction,^12,13 similar to what is found in adults with BPD.^16,17 When this group was reassessed 5 years later at a mean age of 15 years old, only a handful of patients met criteria for BPD, although the group that met criteria for borderline personality of childhood did have poorer functioning as adolescents.¹⁸

Another study¹⁹ looked at a large community sample of high-risk girls between the ages of 6 and 12. This study reported several important findings, including that both parent and teacher ratings of characteristic features of BPD, such as lack of control and aggression, are quite stable during 6 years of follow-up.

These results suggest that, while BPD symptoms may begin in childhood in some patients, it is not yet possible to identify a precursor syndrome specific to BPD. Thus there is an absence of evidence regarding the course and outcomes of patients who do meet the full criteria for BPD in childhood.

Adolescence

While BPD in childhood is a relatively understudied topic, there is growing evidence that BPD can be reliably and validly diagnosed in adolescence. The course of adolescent-onset BPD is similar to what is seen in adult populations.²⁰

There are an increasing number of studies that examine the outcomes of patients diagnosed with BPD in adolescence. In general, the course of BPD during adolescence is not very stable. In particular, a prospective study²⁰ of a clinical population between the ages of 15 and 18 found that only 40% of patients with BPD met criteria at 2-year follow-up. However, this sample only included a smaller number of adolescents with BPD, as this study looked at the course of all PDs. A community study of self-reported symptoms in adolescent twins also found a decrease in rates of BPD diagnosis from 14 to 24 years of age, with significant reductions in symptoms at each 2- to 3-year interval throughout the 10-year follow-up.²¹ A 2-year follow-up of adolescent inpatients also found a change of diagnosis, with an over 50% reduction in the number of patients with BPD at follow-up.²² However, a large longitudinal community study of adolescents with 2- and 8-year follow-up^23–25 found that BPD symptoms tended to persist, even when formal diagnostic criteria were no longer met. When a sample of adolescents previously diagnosed with BPD were followed up about 4 years later, 65% remitted from the diagnosis, consistent with other studies.²⁶ Another research group attempted to identify risk factors for the development of BPD in a community sample of adolescents, and found that maternal–child discord, maternal BPD, paternal SUD, as well as depression, SUD, and suicidality in the adolescent, predict later development of BPD symptoms.²⁷ These risks may moderate the improvement of patients during this phase of development.

Adulthood

The course of BPD in adults has been a focus of research during the past 2 decades, with several longitudinal studies providing insights into its course. The 2 most important prospective studies are the MSAD and the CLPS. The MSAD, which is still ongoing after 20 years, is a prospective, longitudinal study that has been following a group of 290 patients who were initially hospitalized at McLean Hospital with a diagnosis of BPD and a comparison group of 72 patients initially hospitalized with other PD diagnoses. Reassessments occur every 2 years and include various measures to assess symptomatology, functioning, and comorbid psychiatric and medical conditions.²⁸ The CLPS recruited 668 patients from several different sites, mostly as outpatients, who met criteria for either BPD, avoidant PD, obsessive–compulsive PD, schizotypal PD, or MDD without a comorbid PD. These patients were reassessed 6 months after baseline and then annually for 10 years. Measures also focused on symptomatology, comorbidity, and functioning, but also included several assessments of dimensional personality traits.²⁹

Both studies found that most patients with BPD improve with time. The CLPS provides evidence that, even when followed up 2 years after the initial assessment, about one-quarter of patients experience a remission of the diagnosis (defined here as meeting less than 2 symptoms for a period of 2 months or longer) during the prior 2 years. During a 10-year period of follow-up, 91% achieve at least a 2-month remission, with 85% achieving remission for 12 months or longer.³⁰ The MSAD has found similar results extended out to 16 years using a slightly different definition of remission (no longer meeting diagnostic criteria for a period of 2 years or longer) and found that by 16 years, 99% of patients have at least a 2-year period of remission and 78% have a remission lasting 8 years.³¹ Both of these studies also demonstrated that BPD is slower to remit than other PDs and MDD. Finally, 1 study followed patients after 27 years and found that 92% of them no longer met criteria for BPD.³²

However, rates of recurrence were not quite consistent between studies. In the CLPS, recurrence rates for patients with BPD were 11% at 10-year follow-up in those who had achieved at least a 12-month remission. This was significantly lower than for the other PDs and also significantly lower than the recurrence rate for patients with MDD.³⁰ The MSAD found that recurrence rates decreased the longer the remission lasted; 36% of patients experienced a recurrence if their remission lasted only 2 years, but this declined to 10% if their remission lasted 8 years. In general, the BPD group was faster to remit than the comparison group of patients with other PDs.³¹ Taken together, these results suggest that patients with BPD are able to achieve remission of symptoms, and that the longer the remission lasts, the lower the risk of relapse.

When the course of individual symptoms of BPD are examined, studies have generally demonstrated an overall decrease in all symptoms, but with symptoms relating to impulsivity and behavioural manifestations of BPD remitting at a quicker rate than internal, primarily affective experiences.^30,33 In general, behavioural symptoms of PDs are less stable than the personality traits associated with BPD over time.³⁴ Despite early reductions in symptoms of self-harm and suicidality, the risk of completed suicide remains. After 27 years of follow-up, about 10% of patients completed suicide, typically when patients were in their 30s and after multiple failed treatments.³² The MSAD found a suicide rate of about 5% and an identical rate of death by all other causes.³¹ In this sample, several factors were associated longitudinally with suicide attempts, including comorbid MDD, SUD, posttraumatic stress disorder, family history of suicide, and some specific symptoms of BPD.³⁵ However, based on previous research in retrospective studies, the rate of suicide in BPD can be expected to fall between 8% and 10%.^36,37

Comorbid psychiatric disorders are extremely common in patients with BPD. Mood disorders, particularly MDD, SUDs, anxiety disorders, and eating disorders are more frequently seen in patients with BPD, compared with patients with other PDs. Even after 6 years of follow-up, over 60% of BPD patients met criteria for a mood disorder and a similar rate for an anxiety disorder.³⁸ Presence of a BPD diagnosis is also associated with longer duration to recovery from some psychiatric illness, such as MDD.³⁹ Comorbid PDs are also more frequent in patients with BPD, and comorbid avoidant, dependent, and self-defeating PDs were associated with lower rates of remission from BPD.⁴⁰ Comorbid medical conditions are also significant problems for patients with BPD, including irritable bowel syndrome,⁴¹ osteoarthritis, diabetes, obesity, and others.⁴² Patients with BPD also experience more frequent and more intense pain⁴³ with increased use of opiate medications.⁴⁴ In general, rates of psychiatric and medical treatment use, including psychiatric medications, are much higher in patients with BPD and, although there is decline during the first 8 years of follow-up, treatment use remains relatively elevated and unchanged after that.^42,45,46 It is also notable that overall rates of mortality in BPD are elevated, compared with the general population.³²

Treatment Implications

Understanding the course of BPD can have a significant impact on the clinical management of patients with the disorder. One of the first issues is accurately diagnosing the disorder. As previously discussed, BPD can often be identified in adolescents and young adults, and it is at this age that clinicians should start to look for symptoms of the disorder. The course of the disorder appears unstable during adolescence, yet the long-term course for most patients is essentially identical to what is seen in adults. As evidence-based treatments are available—although evidence in youth is not as robust as it is in adults⁴⁷—patients should be directed to these psychotherapies as early as possible. There is no rationale to wait to provide diagnosis and treatment, and many of the arguments for waiting (based on the assumption that symptoms are transient) are not based on evidence.⁴⁸ Finally, providing an accurate diagnosis and providing information on the generally positive outcomes for BPD can instill hope in patients who, all too often, feel completely hopeless.⁴⁹ Evidence suggests that even a single session of psychoeducation about the diagnosis can be beneficial.⁵⁰

One of the all-too-frequent reasons to avoid the diagnosis of BPD in all age groups is the fear of stigmatization. BPD is known to be highly stigmatized in mental health settings,^6,7,51,52 particularly in youth mental health, and part of the reason for the stigma is that BPD is seen as a lifelong disorder that is untreatable. Actually, patients with BPD can be cautiously optimistic about their prognosis, and many treatments exist that can profoundly improve their lives. Much of the stigma occurs when clinicians encounter these patients in crisis settings, such as the emergency department, crisis clinics, and general psychiatric inpatient wards. As these are not the settings in which treatment occurs, clinicians develop a biased perspective on these patients. Few clinicians will have the opportunity to see these patients improve over time in specialized treatment clinics; nonetheless, these patients improve and are treatable. Awareness of the course of BPD is one effective way of reducing stigma against the diagnosis.

Although the course of BPD is generally positive, some patients do experience relapses over time. Most of the specialized treatments for BPD are time-limited and generally of 1 to 3 years’ duration. This makes them expensive, and some experts suggest that a better model of care for patients with BPD would be intermittent psychotherapy.⁵³ Intermittent treatment may allow patients to address different problems at different points in their lives. For example, DBT focuses primarily on self-harm and suicidality,⁵⁴ which tend to be problems early on in the lives of patients with BPD. As these symptoms remit with time, patients who are older may want to deal with problems relating to emptiness and fear of abandonment, which are not directly addressed by DBT, and are symptoms that are slower to remit.³³ An intermittent psychotherapy model would allow patients to focus on specific problems at specific times in their lives, using the therapeutic approach that has the best evidence for that problem.

A related issue is the poor long-term functional status of patients with BPD. Zanarini et al³¹ demonstrated that recovery is significantly less likely in patients with BPD than in an Axis II comparison group. Slightly more than one-half the patients with BPD achieved recovery, which was defined as remission from symptoms as well as good, full-time vocational or educational functioning and at least one stable and supportive relationship with a friend or partner. Patients with BPD were also quite likely to rapidly lose their recovery. Several factors were identified as predictors to recovery, including not being hospitalized (prior to the index hospitalization), higher IQ, prior good vocational functioning, absence of a cluster C comorbidity, and the trait measures of high extraversion and high agreeableness.⁵⁵ Recovery was also associated with both marriage and being a parent, although often at an older age than nonrecovered patients.⁵⁶ Another longitudinal study demonstrated that the Global Assessment of Functioning scores were lower in the BPD group and significantly fewer of these patients achieved levels that represent good functioning³⁰ and were also more likely to receive social assistance.

As functional recovery is so difficult for patients with BPD to attain and maintain, this becomes an important long-term goal for intervention. Currently, the only long-term study to directly assess vocational or educational functioning indicates that mentalization-based treatment, provided in an 18-month partial hospitalization setting, is associated with a greater chance of being employed or in school up to 8 years later.⁵⁷ Other studies have also demonstrated improved social and global functioning, but these were not primary outcomes and were not significantly different between the different treatments; also, duration of follow-up was too short.^58,59 Thus there is evidence that, while treatment is likely to lead to improvements in functional recovery, there is a notable absence of studies that focus on this as a primary outcome. Consideration should be given to providing patients with BPD as much support in attaining functional recovery as possible. This may include psychosocial programs aimed at returning to work, and evaluation and involvement from occupational therapists, vocational counsellors, or other specialists who can help patients develop the skills necessary for any sort of work.

Dr Biskin has previously received an honorarium from Elsevier for publications related to borderline personality disorder.

The Canadian Psychiatric Association proudly supports the In Review series by providing an honorarium to the authors.