Adult Victimization over Time

At baseline, borderline patients reported higher rates of adult rape and physical assault by a partner than Axis II comparison subjects. Four risk factors were found to significantly predict whether borderline patients had an adult history of being a victim of physical and/or sexual violence before their index admission: female gender, a substance use disorder that began before the age of 18, childhood sexual abuse, and emotional withdrawal by a caretaker (a form of neglect). At six-year and 10-year follow-up, borderline patients reported higher rates of being verbally, emotionally, physically, and sexually abused or assaulted than did Axis II comparison subjects. However, each of these forms of abuse declined significantly over time. The clinical implications of these prevalence and predictive findings are discussed.

Mental Health Treatment over Time

At baseline, borderline patients reported using high rates of outpatient treatment and more restrictive forms of treatment, such as inpatient psychiatric treatment. Over time, these rates have declined, particularly for more restrictive and costly forms of treatment. However, high rates of borderline patients remain in outpatient psychotherapy and continue taking standing medications in all major classes of psychotropic medications. Perhaps most concerning is that a substantial percentage of borderline patients have been treated with aggressive polypharmacy (three or more standing medications) despite the health consequences, such as obesity, and despite there being no empirical evidence for this common practice. Evidence-based psychosocial treatments are reviewed, and suggestions are made for treatment as usual.

Recovery from Borderline Personality Disorder

As important as our findings concerning the high rates of symptomatic remission and low rates of symptomatic recurrence are, the rates of recovery are even more important. This is so because we defined “recovery” as concurrent symptomatic remission and good social and good full-time vocational functioning. After 10 years of prospective follow-up, 50% of borderline patients achieved this important goal. After 16 years of prospective follow-up, 60% of borderline patients achieved this key outcome. In general, recovery is more difficult to achieve and maintain than remission. Two vignettes are presented in this chapter. The first deals with a patient who remitted but never recovered, and the second deals with a patient who both remitted and recovered.

Predictors of Time-to-Remission and Recovery

This chapter reports on the bivariate and multivariate predictors of two of our most important outcomes—remission and recovery. In terms of time-to-remission, seven variables that span five areas of prediction were found to be significant in multivariate analyses. These areas were: lack of chronicity, good premorbid vocational functioning, no history of childhood sexual abuse, no family history of substance abuse, and three aspects of temperament. In terms of time-to-recovery, five variables that span three areas of prediction were found to be significant in multivariate analyses. These areas were: lack of chronicity, competence, and temperament. The clinical implications of these significant models and the significant bivariate predictors that did not end up in these models are discussed in detail.

Prevalence and Predictors of Physically Self-Destructive Acts over Time

Self-mutilation and help-seeking suicide threats and attempts are among the few almost pathognomonic symptoms of BPD. This chapter assesses predictors of self-harm and reasons for self-harm over time. It also assesses predictors of suicide threats and attempts over the years of prospective follow-up. Each outcome has a different set of multivariate predictors, but some appear in several multivariate models. More specifically, sexual adversity in childhood and adulthood, major depression, and severity of dissociation are predictors of self-mutilation; and sexual adversity in adulthood, major depression, and severity of dissociation are predictors of suicide attempts. However, these factors do not play a role in predicting suicide threats. Instead, two dysphoric affective states and two outmoded interpersonal survival strategies are the best set of predictors of suicide threats.

Sexual Issues over Time

We studied two types of sexual issues over time. The first issue was sexual relationship difficulties, which we defined as avoiding sex for fear of becoming symptomatic, or becoming symptomatic after having sex. The second issue was sexual orientation and gender of relationship choice. At six-year follow-up, we found that sexual relationship difficulties were significantly more common among borderline patients than among Axis II comparison subjects, although the rate was declining in both study groups. At 16-year follow-up, the same pattern was found for non-recovered versus recovered borderline patients. In terms of the second issue, patients with BPD were significantly more likely than Axis II comparison subjects to report homosexual or bisexual orientation and intimate same-sex relationships. In addition, patients with BPD were significantly more likely than Axis II comparison subjects to report changing the gender of intimate partners, but not sexual orientation, at some point during the follow-up period.

Physical Health and Medical Treatment

Remitted borderline patients were found to have better physical health, make better health-related lifestyle choices, and use fewer costly forms of treatment, such as ER visits, than non-remitted borderline patients. This same pattern was found 10 years later for recovered vs. non-recovered borderline patients. At both time points, obesity was the most common serious health problem, and smoking and lack of exercise were the most common poor lifestyle choices. Obesity was found to be related to poor psychosocial functioning in most realms. Recovered borderline patients had better sleep quality and were not as troubled by dysfunctional attitudes about sleep as non-recovered borderline patients. Borderline patients also reported higher levels of physical pain than Axis II comparison subjects. However, a substantial minority were able to use opioid medications responsibly over time.

Additional Symptom Areas over Time

This chapter details the course of symptom areas that were not covered in Chapter 6, or were covered in a non-comprehensive manner. These symptom areas are anxiety, shame, dissociation, and 17 specific cognitions (e.g., overvalued ideas, ideas of reference). These symptoms tended to decrease over time, but they also tended to remain significantly more severe or more common among borderline patients than among Axis II comparison subjects. Multivariate predictive models were also found for the severity of anxiety and the severity of shame. Taken together, these symptoms represent areas of suffering that would benefit from more clinical attention and support.

Symptomatic Remissions and Recurrences of the Borderline Diagnosis

Many clinicians are reluctant to treat or actively avoid treating patients with BPD. This is so because of the interpersonal difficulties that tend to arise during such a treatment, and is partly due to the idea that BPD is a chronic disorder. This chapter, however, describes the reason for much of the new optimism about the borderline diagnosis. After 16 years of prospective follow-up, it was found that 99% of borderline patients achieved a two-year remission of their BPD, and 78% achieved an eight-year remission of this disorder. Additionally, recurrences of BPD were relatively infrequent, suggesting that remissions of this disorder were stable in nature. These findings suggest that BPD is a “good prognosis” diagnosis and not the chronic condition that many clinicians still believe.

The Long-Term Course of the Symptoms of Borderline Personality Disorder

This chapter reports on the prevalence rates of the 24 symptoms of BPD assessed in this study; prevalence rates that are declining over time. It also reports on rates of remission of each of these symptoms and rates of recurrence following periods of remission. The 24 symptoms are divided into 12 acute symptoms (e.g., self-mutilation, suicide efforts) and 12 temperamental symptoms (e.g., chronic feelings of anger, intolerance of aloneness). It was found that acute symptoms remit more rapidly and are less likely to recur. The clinical implications of these different symptom trajectories are discussed in detail.