Mentalization-based Therapy: A summary of the evidence and new developments

1
Mentalization-based Therapy A summary of the
evidence and new developments

• Dawn Bales, Helene Andrea, Ab Hesselink
  
  Psychotherapeutic Center de Viersprong,
  Viersprong Institute for Studies on Personality
  Disorders (VISPD)
• The Netherlands
• WPA International Congress Florence, april 4,
  2009

2

• Research team
• De Viersprong Roel Verheul, Maaike Smits, Fieke
  vd Meer, Nicole v Beek
• Erasmus University Rotterdam Sten Willemsen,
  Jan van Busschach
• Tilburg University Marieke Spreeuwenberg
• MBT Staff
• (De Viersprong, Bergen op Zoom, The Netherlands)

Internet www.vispd.nl / presentations Email
helene.andrea_at_deviersprong.nl
3
Content

• Mentalization-Based Treatment (MBT)
• A summary of the evidence
• Does MBT work?
• Are the effects lasting?
• Wat does it cost?
• New Developments and future plans
• Does MBT work in another dosage?
• Does MBT work for addiction problems?
• MBT for caregivers
• Other new developments

4
Mentalization-based Therapy

• Psychoanalytically oriented based on attachment
  theory
• Developed in the UK by Bateman Fonagy
• Evidence-based treatment for patients with severe
  BPD
• Maximum duration of 18 months
• Focus increasing patients capacity to mentalize

5
What is mentalization?
Making sense of the actions of oneself and others
on the basis of intentional mental states, such
as desires, feelings, and beliefs. It involves
the recognition that what is in the mind is in
the mind and reflects knowledge of ones own and
others mental states as mental states.
6
Schematic Model of BPD
Retrieval of negative affect laden memories and
cognitions
Constitutional factors
Activating (provoking) risk factors
Hyper-activation of the attachment system
BPD Pre-mentalistic subjectivity
Trauma/ Stress
Poor affect regulation
Inhibition of judgements of social
trustworthiness, paranoid thoughts and
mentalizing failure
Early attachment environment
Formation risk factors
Vulnerability risk factors
7
MBT developmental model of BPD

• Constitutionally vulnerable
• Insecure attachment
• ?
• Inhibited capacity to mentalize
• ?
• Symptoms and interpersonal problems
• Focus MBT enhancing mentalization within the
  context of attachment relationship

8
Goals

• To engage the patient in treatment
• To reduce general psychiatric symptoms,
  particularly depression and anxiety
• To decrease the number of self-destructive acts
  and suicide attempts
• To improve social and interpersonal function
• To prevent reliance on prolonged hospital stays

9
Essential features of the program

• Highly structured
• Consistent and reliable
• Intensive
• Theoretically coherent all aspects aimed at
  enhancing mentalizing capacity
• Flexible
• Relationship focus
• Outreaching
• Individualized treatment plan
• Individualized follow-up

10
A summary of the evidence

• Does MBT work?
• RCT Day-hospital (1999 UK)
• Partial Replication Study (2009 NL)
• Are the effects lasting?
• 18 month Follow-up (2001 UK, 2009 NL)
• Long term follow-up (2008 UK)
• Cost-effectiveness (2003 UK)
• Does MBT work in another dosage?
• RCT IOP (2009 UK)
• Future plans

11
IntroductionMBT-effectiveness United Kingdom

• RCT Day hospital MBT versus TAU for BPD
  patients
• Results
• MBT patients showed significant improvement in
  all outcome measures (Depressive symptoms,
  suicidal and self-mutilatory acts, reduced
  inpatient days, better social and interpersonal
  function)
• TAU patients showed limited change or
  deterioration over the same period
• Conclusion
• MBT superior to standard psychiatric care

Bateman Fonagy, American Journal Psychiatry
1999 2001 2008
12
MBT De Viersprong

• First MBT setting outside UK
• Naturalistic setting (instead of RCT)
• Research question
• What is the treatment outcome
• for severe BPD patients
• after 18 months of day hospital Mentalization
  Based Treatment
• in the Netherlands?

13
Study population
45 patients referred to MBT(Aug.04 Apr. 08)
Excludedn2 no DSM-IV BPDn2 refusedn1 early
dropout
40 PATIENTS INCLUDED
14
Demographic and clinical characteristics study
population (N 40)
Clinical characteristics Study population (N40) Mean Sd
Age 31.7 7.5
N
Female sex 28 70
At least one Axis-I diagnosis 38 95
More than one Axis-I diagnosis 32 80
Anxiety Disorders 17 43
Mood disorders 14 35
Eating disorders 13 33
Substance abuse dependency start treatment 26 66
PTSD 5 13
More than 1 comorbid axis II diagnosis 28 70
Paranoïd personality disorder 9 23
Avoidant personality disorder 9 23
Dependant personality disorder 6 15
Histrionic personality disorder 4 10
Antisocial personality disorder 3 8
15
Prospective naturalistic study design

• Measurements start treatment, 6, 12, and 18
  months
• Continuous outcomes GEE (SPSS)- correction for
  missing values- age and sexe as covariates-
  effect sizes corrected for data dependency
• Categorical outcomes univariate statistics
• Baseline n406 months n31 12 months n19 18
  months n16

16
Results Treatment engagement

• Low dropout rate (n5 12.5)
• n3 dropouts
• n2 push-outs
• Average treatment length 15.1 months (sd 4.2
  months range 4-18 months)

17
Results Symptomatic functioning (SCL90, BDI,
EQ-5D)
Effectsizes 0.75 1.79
Bales et al, 2009 Submitted do not quote
18
Results Social and interpersonal functioning
(IIP, OQ)
Effectsizes 1.17 1.56
Bales et al, 2009 Submitted do not quote
19
Domain personality pathology
SIPP Verheul et al, 2008
Effectsizes 1.08 1.58 large very large
20
Results care consumption domain
21
Conclusions

• Significant improvement on all outcome measures
  with effect sizes ranging from large to very
  large
• Low drop-out rate despite limited exclusion
  criteria
• Results similar to results of Bateman Fonagy
  (1999)

22

• (Methodological) limitations
• Working mechanisms mentalization
• Low N and missing values
• Causality

23
MBT Research

• Does MBT work?
• RCT Day-hospital (1999 UK)
• Partial Replication Study (2008 NL)
• Are the effects lasting?
• 18 month Follow-up (2001 UK, 2009 NL)
• Long term follow-up (2008 UK)
• Cost-effectiveness (2003, UK)
• Does MBT work in another dosage?
• RCT IOP (2009, UK)
• Future plans

24
Treatment of Borderline Personality Disorder With
Psychoanalytically Oriented Partial
hospitalization An 18 month Follow-up Bateman
Fonagy, American Journal of Psychiatry
(2001) Summary follow-up trial MBT patients
maintained and even showed additional improvement
of symptomatic and clinical gains during 18
months follow-up
25
8-Year follow-up of Patients treated for
Borderline Personality Disorder
Mentalization-Based Treatment versus Treatment as
usual Bateman Fonagy 2008 American Journal of
Psychiatry
26
8 year follow-up UK

• Study the effect of MBT-PH vs. TAU
• N41 patients from original trial
• 8 years after entry in to RCT, 5 years after all
  MBT treatment was complete
• Method
• interviews (research psychologists blind to
  original group allocation)
• structured review medical notes

8 year follow-up 2008 Bateman Fonagy
27
Zanarini Rating Scale for BPD mean (SD)
MBT-PH (n 22) TAU (n15) Significance
Positive criteria n () 3 (13.6) 13 (86.7) ?2 16.5 p.000004
Total mean (SD) 5.5 (5.2) 15.1 (5.3) F1,35 29.7 p.000004
Affect mean (SD) 1.6 (2.0) 3.7 (2.0) F1,35 9.7p.004
Cognitive mean (SD) 1.1 (1.4) 2.5 (2.0) F1,35 6.9 p.02
Impulsivity mean (SD) 1.6 (1.8) 4.1 (2.3) F1,35 13.9 p.001
Interpersonal mean (SD) 1.5 (1.7) 4.7 (2.3) F1,35 23.2p.00003
8 year follow-up 2008 Bateman Fonagy
28
Suicide attempts mean (SD)
MBT-PH TAU Significance
Total N mean (SD) .05 (0.9) 0.52 (.48) U 73 Z 3.9 p .00004
Any attempt N () 5 (23) 14 (74) ?2 8.7 df- 1 P .003
8 year follow-up 2008 Bateman Fonagy
29
Global Assessment of Function
MBT-PH TAU Significance
Mean (SD) 58.3 (10.5) 51.8 (5.7) F1,35 5.4 p.03
Number () gt 60 10 (45.5) 2 (10.5) ?2 6.5 df 1 p .02
8 year follow-up 2008 Bateman Fonagy
30
Vocational status
8 year follow-up 2008 Bateman Fonagy
31
Conclusions from long term follow-up

• MBT-PH group continued to do well 5 years after
  all MBT treatment had ceased
• TAU did badly within services despite significant
  input
• TAU is not necessarily ineffective in its
  components but package or organization is not
  facilitating possible natural recovery
• BUT
• Small sample, allegiance effects (despite
  attempts being made to blind the data collection)
  limit the conclusions.
• GAF scores continue to indicate deficits.
  Suggests less focus during treatment on
  symptomatic problems greater concentration on
  improving general social adaptation

8 year follow-up 2008 Bateman Fonagy
32
MBT Research

• Does MBT work?
• RCT Day-hospital (1999 UK)
• Partial Replication Study (2008 NL)
• Are the effects lasting?
• 18 month Follow-up (2001 UK, 2009 NL)
• Long term follow-up (2008 UK)
• Wat does it cost? (2003, UK)
• Does MBT work in another dosage?
• RCT IOP (2009, UK)
• Future plans

33
Health Service Utilization Costs for Borderline
personality Disorder Patients Treated with
Psychoanalytically Oriented Partial
Hospitalization Versus General Psychiatric
Care Bateman Fonagy (2003) American Journal
of Psychiatry
34
Total Annual Health Care Utilization Costs
35
Cost-effectiveness

• Significantly lower cost during treatment
  compared to 6-month pretreatment costs for both
  MBT and General Care Group
• During FU period annual cost of MBT 1/5 of anual
  General Care costs

36
Content

• Mentalization-Based Treatment (MBT)
• A summary of the evidence
• Does MBT work?
• Are the effects lasting?
• Wat does it cost?
• New Developments and future plans
• Does MBT work in another dosage?
• Does MBT work for addiction problems?
• MBT for caregivers
• Other new developments

37
Treatment Outcome Studies UK Implementation of
Outpatient Mentalization Based Therapy
for Borderline Personality Disorder Bateman
Fonagy (2009)
38
Design of Intensive out-patient MBT RCT

• Referrals for IOP-MBT and SCM groups
• Random allocation (minimisation for age, gender,
  antisocial PD)
• Individual (50 mins) Group (1.5 hrs) weekly for
  18 months
• Assessments at admission, 6 months, 12 months, 18
  months
• Medication followed protocol

IOP vs. SCM Bateman Fonagy (2009)
39
Therapy

• SCM - weekly
• Support and structure
• Challenge
• Advocacy
• Social support work
• Problem solving
• Medication review
• Crisis management

• MBT - weekly
• Support and structure
• Challenge
• Basic mentalizing
• Interpretive mentalizing
• Mentalizing the transference
• Medication review
• Crisis management

IOP vs. SCM Bateman Fonagy (2008?)
40
(Preliminary) Conclusions IOP

• MBT-IOP is surprisingly effective
• The sample was less disturbed than the partial
  hospital sample
• Most of the MBT subjects but also some of the SCM
  subjects lost their diagnosis
• Relatively few of the SCM patients improved in
  terms of subjective measures
• The MBT patients more reliably improved
• Even when improved, remains quite high scoring on
  pathology scales

IOP vs. SCM Bateman Fonagy (2009)
41
IOP in the Netherlands

• Course explicit mentalizing (CEM 8-10 sessions)
• Two times group psychotherapy, 75 min per week
• One individual contact per week
• Maximum duration 18 months

42
RCT

• IOP vs day hospital treatment
• Explosive ASPD is excluded
• Pilot randomisation
• N20
• gt70 cooperation

43
Content

• Mentalization-Based Treatment (MBT)
• A summary of the evidence
• Does MBT work?
• Are the effects lasting?
• Wat does it cost?
• New Developments and future plans
• Does MBT work in another dosage?
• Does MBT work for addiction problems?
• MBT for caregivers
• Other new developments

44
Substance abuse among MBT patientsPrevalence
and relation to treatment outcome

45
Background Aim

• Literature
• 57-67 BPD patients addiction problems -gt MBT?
• Combination BPD addiction -gt treatment
  prognosis worse
• Study objective
• What is the prevalence of DSM-IV substance
• abuse among MBT-patients?
• Additional explorative analysis
• Is substance abuse related to MBT treatment
  outcome?

46
Study population (1)
45 patients referred to MBT(Aug.04 Apr. 08)
Excludedn2 no DSM-IV BPDn2 refusedn1 early
dropout n1 no follow-up measurements
39 PATIENTS INCLUDED
47
Measurement Substance Abuse

• Composite International Diagnostic Interview
  (CIDI)
• Lifetime auto-version 2.1
• Substance Abuse Module (CIDI-SAM)
• Alcohol dependence or abuse (section J)
• Drugs / medication / other substance abuse or
  dependence (section L)

48
Study population (continued)
39 eligible patients
No CIDI availablen6 refused n9 untraceable
(not in treatment anymore)
24 PATIENTS with CIDI-SAM results
49
Results Prevalence substance abuse
No substance Diagnosis 21(N 5)
1 diagnosis 13 (N 3)
2 diagnoses 21 (N 5)
3-5 diagnoses 29 (N 7)
6-7 diagnoses 17 (N 4)
CIDI-SAM Abuse / dependence
Total population (N 24) 79.2 (N 19)
Specific prevalences 1. Alcohol 67 (N 16)
2. Cannabis 58 (N 14) 3. Cocaine 42 (N
10)
50

• Hypothesis from literature
• Prevalence liftetime substance abuse 50-70
• MBT population
• Prevalence 79
• Explorative analysis
• Association with treatment outcome?

51
Treatment outcome resultsExplorative
longitudinal analyses
Interaction Time x Lifetime substance abuse?
52
Interaction time Lifetime substance abuse

• Pattern for 50 of the outcome measurements
• Improvement for substance abusers and
  non-abusers
• Stronger improvement for no lifetime substance
  abuse
• However, only n5 no lifetime substance abuse!

53
New comparison subgroups

• N 5 no lifetime substance abuse

• N 19 lifetime substance abuse
• Diagnosis starttreatment?
• Yes N 13
• No N 6

Diagnosis start treatmentYes N 13 No N 11
(n 5 n 6)
54
Interaction time substance abuse start
treatment

• Pattern
• No significant interaction effect
• Improvement substance abusers start treatment
  (n13) resembles improvement non abusers
  start treatment (n11)

55
Interaction Time Substance abuse Summary

• Lifetime substance abuse
• N 19 yes, N 5 no
• Tendency towards stronger improvement forsmall
  group without lifetime substance abuse
• Substance abuse start treatment
• N 13 yes, N 11 no
• No difference improvement over time

56
Limitations

• Small N
• Retrospective measurement substance abuse (recall
  bias)
• Broader range of addictive problems
• Substance abuse outcome data not yet available

57
Conclusions

• Very high prevalence (79) lifetime substance
  abuse diagnosis among MBT patients
• Significant improvement possible for DD patients
  (severe BPD and substance abuse)

58
BPD and addiction Hannah

• 22 years old female
• Axis I polysubstance dependence (cannabis,
  cocaïne, XTC, speed) ADHD post-traumatic stress
  disorder sexual dysfunction
• Axis II borderline personality disorder
  histrionic personality disorder, paranoid
  features
• Low-level borderline/psychotic personality
  organisation (Kernberg)
• Unable to follow a whole day-program without
  drugs
• Completely integrated in drugscene

59
BPD and addiction Henry

• 46 years old
• Axis I polysubstance dependence (cocaine and
  alcohol) sexual dysfunction depression
• Axis II borderline personality disorder
  narcissistic personality disorder, avoidant
  personality disorder
• Fired from work because of drug dependence
• Divorced, two children
• Detoxification before start MBT
• Able to follow a day program without drugs
• Some social structure (volunteer, children
  visits, etc)
• No users as friends, not in drugscene

60
New Developments MBT-DD

• MBT-PH and IOP parallel low-frequent out-patient
  contact in addiction-center
• Plan integrated MBT- DD treatment
• Program
• inpatient detox
• day-hospital (PH)
• outpatient treatment
• Including system-oriented interventions

61
Content

• Mentalization-Based Treatment (MBT)
• A summary of the evidence
• Does MBT work?
• Are the effects lasting?
• Wat does it cost?
• New Developments and future plans
• Does MBT work in another dosage?
• Does MBT work for addiction problems?
• MBT for caregivers
• Other new developments

62
MBT for caregivers MBT-C

• A mentalizing parental program for high-risk
  parents and their children
• Goal promoting reflective parenting by enhancing
  the caregivers mentalizing with respect to
  him/herself and the child
• Population caregivers with severe BPD and their
  children up to seven years
• The interventions on caregiver-child interactions
  are based on principles from Minding the baby
  (Slade)

63
Plan MBT-C

• Program
• Course explicit mentalizing (8-10 sessions)
• Course explicit mentalizing for caregivers (6-8
  sessions)
• IOP MBT (1 gpt and 1 individual session)
• Interventions on caregiver-child interaction
  home-visitations and routine videotaping of
  mother-child interactions
• Research
• MBT-C versus TAU
• Hypothesis enhancing the caregivers
  mentalizing capacity results in less
  psychopathology in the children

64
Content

• Mentalization-Based Treatment (MBT)
• A summary of the evidence
• Does MBT work?
• Are the effects lasting?
• Wat does it cost?
• New Developments and future plans
• Does MBT work in another dosage?
• Does MBT work for addiction problems?
• MBT for caregivers
• Other new developments

65
Other New MBT Developments

• Adolescents (MBT-a, Viersprong, NL)
• Antisocial and BPD (Bateman, 2008 Viersprong,
  NL)
• Families (MBFT), (Viersprong, NL)
• Severe eating disorders (GGZ-MB, NL)
• Severe psychosomatic disorders (Eikenboom, NL)
• Children/parents (MBKT, NPi, NL)

66
Conclusions

• A summary of the evidence
• MBT does work for severe borderline patients
• The effects are lasting
• MBT shows considerable cost savings after
  treatment
• MBT-IOP also seems effective
• MBT is also promising for addiction
• Internationally many new developments

67

• www.vispd.nl/presentations
• dawn.bales_at_deviersprong.nl
• helene.andrea_at_deviersprong.nl
• ab.hesselink_at_deviersprong.nl