Ten-Year Course of Borderline

Personality Disorder

Mary C. Zanarini, Ed.D.

McLean Hospital
Harvard Medical School
 Borderline Personality Disorder (BPD)
Is Now Seen as a Valid Disorder
► According to the criteria of Robins
and Guze (1970)
Z It can be delimited from other
psychiatric disorders
Z Something of its etiology (both
environmental and biological) is known
Z It “runs” in families
Z It has a complex but increasingly
known course
Robins E, Guze SB. Am J Psychiatry. 1970;126:983-987.
 Borderline Personality Disorder Is Now
Recognized as a Common Disorder
► 1.8% of American adults
meet criteria for BPD
(range 1.6-5.9%)

► About as common as
bipolar I disorder

► More common than

schizophrenia

APA. DSM-IV-TR; 2000.

 Continuum of Borderline
Psychopathology
► Some people with BPD recover
spontaneously and are never patients
► Some use nonintensive outpatient
treatment and are never hospitalized
► Others become severely ill and use
large amounts of mental health
services, including repeated
inpatient stays
 Continuum of Borderline
Psychopathology (cont.)
► The latter group has defined BPD
for generations of clinicians
► Until
very recently, most research
studies have focused on
inpatient-level patients
► Thispresentation deals with this
type of severely ill patient
 McLean Study of
Adult Development (MSAD)
► First NIMH-funded prospective study of the
longitudinal course of BPD
► 362 McLean inpatients assessed
at baseline
► 8 waves of blind follow-up are complete: 2, 4,
6, 8, 10, 12, 14, and 16-year data
► 18-year wave began in July of 2010
► 20-year wave began in July of 2012
 Subjects

► 290 patients meeting DIB-R and

DSM-III-R criteria for BPD

► 72 axis II comparison subjects

meeting DSM-III-R criteria for
another personality disorder (but
neither study criteria set for BPD)

DIB-R=Revised Diagnostic Interview for Borderlines.

 DIB-R: Sectors of Psychopathology

► Dysphoric affect

► Disturbed cognition

► Impulsive behaviors

► Troubled relationships
Zanarini MC, et al. J Pers Disord. 1989;3:10-18.
 DIB-R: Definition of Borderline
Personality Disorder

► Symptoms in each of these

4 domains of borderline
psychopathology must be
present at the same time
► Results in a somewhat smaller
and more homogeneous group
of patients than DSM criteria
Zanarini MC, et al. J Pers Disord. 1989;3:10-18.
 Earlier Studies of Course of
Borderline Personality Disorder
► 17 small-scale, prospective studies of the short-
term course of BPD
Z Patients with BPD do poorly in the short-run

► 4 large-scale, follow-back studies of the long-term

course of BPD
► McGlashan: Chestnut Lodge
► Stone: New York State Psychiatric Institute
► Paris: Jewish General Hospital in Montreal
► Plakun: Austin Riggs
Z Patients with BPD do substantially better in the long-run
 Limitations of Earlier Studies

► Use of chart review or clinical

interviews to diagnose BPD
► No comparison group or the use
of less than optimal comparison
subjects
► Reliance on small size samples
with high attrition rates
Limitations of Earlier Studies (cont.)
► Only very basic data collected
at baseline and follow-up
► Typically,
only 1 postbaseline
reassessment
► Nonblindpostbaseline
assessments
► Variable number of years of
follow-up in the same study
 MSAD Subject Retention at 10-year
Follow-up

► 92% of surviving patients with BPD

still participating

► 85% of surviving axis II comparison

subjects still participating
 Time-to-Symptomatic Remission*

2-Year 4-Year 6-Year 8-Year 10-Year

Follow-Up Follow-Up Follow-Up Follow-Up Follow-Up
% % % % %

34.9 55.2 75.6 87.6 93.0

*Remission defined as no longer meeting either criteria set for BPD (DIB-R and DSM-III-R) for two years.
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.
 Time-to-Sustained
Symptomatic Remission*
4-Year 6-Year 8-Year 10-Year
Follow-Up Follow-Up Follow-Up Follow-Up
% % % %

29.6 46.9 67.1 86.0

*Sustained remission defined as no longer meeting either criteria set for BPD (DIB-R and DSM-III-R) for
four years.
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.
 Time-to-Symptomatic Recurrence*

2 years after 4 years after 6 years after 8 years after

1st remission 1st remission 1st remission 1st remission
% % % %

16.5 22.4 27.4 29.5

*Recurrence defined as meeting the study criteria for BPD for two years after meeting the criteria for
remission in a previous follow-up period.
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.
 Time-to-Loss of Sustained Remission*

2 years after 1st 4 years after 1st 6 years after 1st

remission remission remission
% % %

6.9 12.8 15.4

*Loss of sustained remission defined as meeting the study criteria for BPD for two years after meeting the
criteria for sustained remission in a previous follow-up period.
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.
 Completed Suicide

2-Year 4-Year 6-Year 8-Year 10-Year Total

Follow-Up Follow-Up Follow-Up Follow-Up Follow-Up Follow-Up
% % % % % %

1.7 1.4 0.7 – 0.3 4.1

(N=5) (N=4) (N=2) (N=1) (N=12)

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.

 Complex Model of Borderline
Psychopathology
► Hyperbolic temperament is
the outward “face” of the
neurobiological dimensions
that underlie borderline
psychopathology
► After “kindling” of some kind,
acute and temperamental
symptoms develop
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.
 Acute Symptoms
► Resolve relatively quickly
► Are the best markers for the disorder
► Are often the main reason for
expensive forms of psychiatric care,
such as inpatient stays
► Are akin to the positive symptoms
of schizophrenia
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.
 Temperamental Symptoms

► Resolve relatively slowly

► Are not specific to BPD
► Are associated with ongoing
psychosocial impairment
► Are akin to the negative
symptoms of schizophrenia
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.
 Examples of Symptoms

► Acute symptoms: self-mutilation,

suicide efforts, quasi-psychotic
thoughts

► Temperamental symptoms:
angry feelings and acts,
distrust and suspiciousness,
abandonment concerns
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.
 Time-to-Remission of Chronic Anger and
Self-mutilation

100

80
% of Patients

60

40

20

0
Baseline 2-Year 4-Year 6-Year 8-Year 10-Year
Follow-Up Follow-Up Follow-Up Follow-Up Follow-Up

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

 Time-to-Remission of Intolerance of
Aloneness and Suicide Efforts

100

80
% of Patients

60

40

20

0
Baseline 2-Year 4-Year 6-Year 8-Year 10-Year
Follow-Up Follow-Up Follow-Up Follow-Up Follow-Up

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

Course of 24 BPD Symptoms Studied

► Using two different methods of defining acute

and temperamental symptoms among
borderline patients
Z 12 symptoms were found to be acute in nature
Z And 12 symptoms were found to be
temperamental in nature

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

 Acute Symptoms I

► Affective Symptoms
Z Affective instability
► Cognitive Symptoms
Z Quasi psychotic thought
Z Serious identity disturbance
► Impulsive Symptoms
Z Substance abuse
Z Promiscuity
Z Self-mutilation
Z Suicide efforts
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.
 Acute Symptoms II

► Interpersonal Symptoms
Z Stormy relationships
Z Devaluation/manipulation/sadism
Z Demandingness/entitlement
Z Serious treatment regressions
Z Countertransference problems/”special” treatment
relationships

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

 Temperamental Symptoms I

► Affective Symptoms
Z Depression
Z Helplessness/hopelessness/worthlessness
Z Anger
Z Anxiety
Z Loneliness/emptiness
► Cognitive Symptoms
Z Odd thought (e.g., overvalued
ideas)/unusual perceptual experiences
(e.g., depersonalization)
Z Nondelusional paranoia
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.
 Temperamental Symptoms II

► Impulsive Symptoms
Z Other forms of impulsivity (e.g., eating
binges, spending sprees, reckless driving)
► Interpersonal Symptoms
Z Intolerance of aloneness
Z Abandonment/engulfment/annihilation
concerns
Z Counterdependency
Z Undue dependency/masochism

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

Symptoms That Resolve Most Rapidly

► Those reflecting core areas of impulsivity

(e.g., self-mutilation, suicide efforts)
► Active attempts to manage interpersonal
difficulties (e.g., stormy relationships,
devaluation/manipulation/sadism)

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

 Most Stable Symptoms

► Affective symptoms reflecting areas of

chronic dysphoria (e.g., anger,
loneliness/emptiness)
► Interpersonal symptoms reflecting
abandonment and dependency issues
(e.g., intolerance of aloneness, counter-
dependency problems)

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

Clinical Implications of Symptomatic
Findings I

► There are five empirically-based

comprehensive forms of therapy for BPD
Z Dialectical Behavioral Therapy (DBT): Linehan
Z Mentalization-based Treatment (MBT): Bateman
and Fonagy
Z Transference-focused Psychotherapy (TFP):
Kernberg
Z Schema-focused Therapy (SFT): Young
Z General Psychiatric Management (GPM): McMain
and Links

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

Clinical Implications of Symptomatic
Findings II

► All five of these treatments are aimed at

acute symptoms
► Treatments aimed at temperamental
symptoms need to be developed

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2007;164:929-935.

 Broadly-defined Good Psychosocial
Functioning
► 78% of patients with BPD attain or
maintain broadly-defined good
psychosocial functioning over the
course of 10 years of prospective
follow-up
Z This goal is defined as at least
1 emotionally sustaining relationship with
a friend or romantic partner and
Z Both a good vocational performance and
a sustained vocational history

Adapted from: Zanarini MC, et al. Acta Psychiatr Scand. 2010;122:103-109.

 Narrowly-defined Good Psychosocial
Functioning
► 64% of patients with BPD attain or
maintain narrowly-defined good
psychosocial functioning over the
course of 10 years of prospective
follow-up
Z This goal is defined as at least
1 emotionally sustaining relationship with
a friend or romantic partner and
Z A good vocational performance, a
sustained vocational history, and full-time
vocational engagement
Adapted from: Zanarini MC, et al. Acta Psychiatr Scand. 2010;122:103-109.
 Stability of Good Psychosocial
Functioning Over Time

► Broadly-defined good psychosocial

functioning is more stable than
narrowly-defined good psychosocial
functioning

Adapted from: Zanarini MC, et al. Acta Psychiatr Scand. 2010;122:103-109.

 Sectors of Good Psychosocial
Functioning Over Time

► Almost all failures to attain or actual

losses of narrowly-defined good
psychosocial functioning were due
to problems in the vocational and
not the social realm

Adapted from: Zanarini MC, et al. Acta Psychiatr Scand. 2010;122:103-109.

 Psychosocial Functioning of Axis II
Comparison Subjects

► 93% maintained or attained

broadly-defined good psychosocial
functioning

► 92% maintained or attained

narrowly-defined good psychosocial
functioning
Adapted from: Zanarini MC, et al. Acta Psychiatr Scand. 2010;122:103-109.
 Clinical Implications of Psychosocial
Findings

► Rehabilitation model might be

useful for those who cannot work or
go to school full-time in an effective
and consistent manner

Adapted from: Zanarini MC, et al. Acta Psychiatr Scand. 2010;122:103-109.

 Collaborative Longitudinal Personality
Disorders Study (CLPS)

► Also NIMH-funded

► Now finished after following

subjects for 10 years

► Basically the same symptomatic

and psychosocial findings
Adapted from: Gunderson JG, et al. Arch Gen Psychiatry. 2011;68:827-837.
 Recovery from BPD

► Recovery is defined as having a

concurrent remission from BPD and
narrowly-defined good psychosocial
functioning

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.

 Time-to-Recovery from BPD*

2-Year 4-Year 6-Year 8-Year 10-Year

Follow-Up Follow-Up Follow-Up Follow-Up Follow-Up
% % % % %

14.3 26.8 36.0 42.8 50.3

*Recovery from BPD defined as concurrent remission from BPD and narrowly-defined good
psychosocial functioning.
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.
 Time-to-Loss of Recovery from BPD*

2 years after 4 years after 6 years after 8 years after

1st remission 1st remission 1st remission 1st remission
% % % %

12.6 19.8 28.7 33.6

*Loss of recovery from BPD defined as meeting the study criteria for BPD for two years after meeting the
criteria for remission in a previous follow-up period and/or loss of one of the four elements of narrowly-
defined good psychosocial functioning.
Adapted from: Zanarini MC, et al. Am J Psychiatry. 2010;167:663-667.
Time-to Remission, Sustained Remission, and Recovery
From BPD

100

80

60

40

20

0
2 YR FU 4 YR FU 6 YR FU 8 YR FU 10 YR FU

Recovery from BPD Sustained Remission Remisson from BPD

Time-to-Loss of Remission, Sustained Remission, and
Recovery from BPD

35
30
25
20
15
10
5
0
2 YRS 4 YRS 6 YRS 8 YRS
Loss of SustainedRemission Recurrence of BPD Loss of Recovery from BPD
 Predictors of Time to Remission
from Borderline Personality Disorder
►7 factors found to predict earlier
time to remission
Z Younger age
Z Good vocational record
Z No history of childhood sexual abuse
Z No family history of substance abuse
Z Absence of an anxious cluster
personality disorder
Z High agreeableness
Z Low neuroticism

Adapted from: Zanarini MC, et al. Am J Psychiatry. 2006;163:827-832.

 Nature of These Predictors
►4 factors commonly assessed in
clinical practice
Z Younger age–demographics
Z Good vocational record–psychosocial
functioning
Z No history of childhood sexual
abuse–adverse childhood events
Z No family history of substance
abuse–family history of psychiatric
disorder
 Nature of These Predictors (cont.)
►3 factors commonly noticed but rarely
discussed in clinical practice
► All 3 are aspects of temperament
Z Absence of anxious cluster PD–low
levels of shyness and undue
dependency
Z High agreeableness–not particularly
argumentative or manipulative
Z Low neuroticism–does not typically
feel inferior or ashamed
 Psychiatric Treatment

► Mostly treated in community

► Over 70% of patients with BPD are
in individual therapy and taking
standing medications during all
5 follow-up periods
► However, rate of psychiatric
hospitalization declined from 79% at
baseline to 29% at 10-year follow-up
Adapted from: Hörz, et al. Psychiatr Serv. 2010;61:612-616.
 Polypharmacy at 10-Year Follow-Up

40
BPD
30 OPD
% of Patients

20

10

0
≥3 Medications ≥4 Medications ≥5 Medications

Adapted from: Hörz, et al. Psychiatr Serv. 2010;61:612-616.

 Polypharmacy and
Borderline Personality Disorder
► No empirical evidence for its efficacy
► Associated with high rates of obesity
► Which, in turn, is associated with
elevated rates of
Z Osteoarthritis
Z Diabetes
Z Hypertension
Z Chronic back pain
Z Urinary incontinence
Z Gastroesophageal reflux disorder
Z Gallstones
Adapted from: Frankenburg FR & Zanarini MC. J Pers Disord. 2006;20:71-80.
 Main Findings
► 93% of patients with BPD experience
a remission of their BPD
► Recurrences of BPD are relatively rare
► Thecourse of BPD is very different
from that of mood disorders where
remission occurs more rapidly but
recurrences are more common
 Main Findings (cont.)

►Completed suicide is substantially

less common than the expected
10%
►Thismay be due to more trauma-
sensitive or supportive treatments
 Main Findings (cont.)

► BPD seems to be comprised of two

types of symptoms
Z Acutesymptoms
Z Temperamental symptoms
 Main Findings (cont.)

► Almost 80% of patients with BPD

attain broadly-defined good
psychosocial functioning
► But only 64% attain narrowly-defined
good psychosocial functioning
► Social functioning is less impaired
than vocational functioning
 Main Findings (cont.)

►Recovery from BPD is more

difficult to attain than remission
from BPD alone
►However, it is relatively stable
once attained
 Main Findings (cont.)

►Prediction of time to remission

is multifactorial in nature
Z Involvesfactors that are routinely
assessed in treatment
Z And other factors, particularly
aspects of temperament, that are
not
 Conclusions

► Taken together, the results

of this study suggest that the
prognosis for most, but not all,
patients with BPD is better than
previously recognized