Beginning with the Endpoint in Mind: BPD Indication

Pragmatic Considerations
Understanding the Indication
Epidemiology
Targets
Expected Effect(s)
Choosing an Endpoint
Valid and Fits the condition:
Reliable: Test-Retest, Inter-rater
Acceptable respondent burden
Clinically Meaningful
Regulatory acceptability
Resilient to remote assessment
BPD trials can expect high susceptibility
to subjectivity and temporal variance
BPD trials can expect high susceptibility
to subjectivity and temporal variance

There are no lab tests

for Borderline Personality Disorder
Co-occurrence of Category A Personality Disorder
may be a moderator of treatment response
 Assessments used in BPD Trials
ALS –Affective Lability Scale
BEST –Borderline Evaluation of Severity over Time
BDI –Beck Depression Inventory
BHS –Beck Hopelessness Scale
BIS –Barratt Impulsiveness Scale
BPDCL –Borderline Personality Disorder Checklist
BPRS = Brief Psychiatric Rating Scale
BSI –Brief Symptom Inventory
BSS –Beck Scale for Suicide Ideation
CAPS5 –Clinician Administered PTSD Scale for DSM-5
CGI –Clinician Global Impression
DES Dissociative Experiences Scale
GAF –Global Assessment of Functioning
HAM-A –Hamilton Rating Scale for Anxiety
HAM-D-17 –Hamilton Rating Scale for Depression
LSDS –Lifetime Self-Destructiveness Scale
MOAS –Modified Overt Aggression Scale
PGI –Patient Global Impression
SCL-90-R –Symptom Checklist 90 –Revised
SCID-5-PD –Structured Clinical Interview, DSM-5, Personality Disorders
SDS –Sheehan Disability Scale
S-STS –Sheehan-Suicidality Tracking Scale
Suicide Att. –Number of Suicide Attempts
YMRS –Young Mania Rating Scale
ZAN-BPD –Zanarini Rating Scale for Borderline Personality Disorder
 Choice Point: Off the shelf or Custom made?
Off the Shelf Outcome Measures

Considerations

• Extant Pertinent Scales

• Training/ Training Requirements

• Manualized

• Sufficient Psychometric data

 Choice Point: Off the shelf or Custom made?
Off the Shelf Outcome Measures

Considerations

• Extant Pertinent Scales

• Extant Pertinent Scales
• Training/ Training Requirements
• Training/ Training Requirements
• Manualized
• Manualized
• Sufficient Psychometric data
• Sufficient Psychometric data
• Resilience (site and remote assessment)
 Some Potential Off the Shelf Endpoints for BPD RCTs
Basis Type Examples
Difficulty with Emotional Regulation (DERS)
Borderline Evaluation of Severity Over Time (BEST)
Barratt Impulsiveness Scale
Zan-BPD SR
PRO
Subjective Q-LES-Q
Report Ecological Momentary Assessments
Zan-BPD
The PANSS Excited Component (PANSS-EC)
ClinRO Modified Overt Aggression Scale (MOAS)
CGI
Heart Rate Variability
Sensor
Actigraphy
Objective
Facial Expression
Measure AI
Vocal Pattern
Performance N-back
 Gold Standard for Comparison= Zan-BPD
Most commonly used, Face validity
DSM-5 BPD Criteria and ZAN-BPD: Same 9 items but different approach to scoring
DSM 5 Criteria/= ZAN-BPD Item SCID: Threshold required Zan- BPD Items (Severity 0-4)
1 Frantic efforts to avoid real or imagined abandonment. several examples None, Mild, Moderate, Serious, Severe

2 A pattern of unstable and intense interpersonal relationships either one prolonged relationship or several None, Mild, Moderate, Serious, Severe
characterized by alternating between extremes of idealization briefer relationships in which the alternating
pattern occurs at least twice
and devaluation.

3 Identity disturbance: markedly and persistently unstable self- acknowledges trait None, Mild, Moderate, Serious, Severe
image or sense of self.
4 Impulsivity in at least two areas that are potentially self- several examples indicating a pattern of None, Mild, Moderate, Serious, Severe
damaging impulsive behavior

5 Recurrent suicidal behavior, gestures, or threats, or self- ≥2 events (when not in a Major Depressive None, Mild, Moderate, Serious, Severe
mutilating behavior. Episode)

6 Affective instability due to a marked reactivity of mood acknowledges trait None, Mild, Moderate, Serious, Severe

7 Chronic feelings of emptiness acknowledges trait None, Mild, Moderate, Serious, Severe

8 Inappropriate, intense anger or difficulty controlling anger acknowledges trait and at least one example or None, Mild, Moderate, Serious, Severe
several examples

9 Transient, stress-related paranoid ideation or severe several stress-related examples that do not None, Mild, Moderate, Serious, Severe
dissociative symptoms. occur exclusively during a Psychotic
Disorder or a Mood Disorder With Psychotic
Features
Zan-BPD = Standard for Comparison
 Zan-BPD = Standard for Comparison
• Pertinent to BPD= Face Validity (for DSM criteria)
– Construct Validity- yes
– Convergent Validity- limited
– Discriminant Validity- ??

• Training/ Training Requirements- Known

– Existing: Rater Training
– Existing Certification process

• Manualized- Yes

• Psychometric data: Regulatory acceptability-?

– Prior study experience and data
– Inter-rater reliability – acceptable
– Intra-rater reliability- under study

• Resilience (site and remote assessment)

– Remote assessment: video assessment feasible
 Convergent Validity
of Interview and Self-report Versions of ZAN-BPD

(N=75)
ZAN-BPD Item Analysis: Olanzapine 3-
arm fixed-dose trial*

Placebo OLZ 2.5 mg OLZ 5-10 mg

* *
8 8.5
6.8

ZAN-BPD Total Change from Baseline to Endpoint (LOCF)

 ZAN-BPD Item Analysis: Olanzapine
3-arm fixed-dose trial*
Change from Baseline to Endpoint (LOCF)
*
1.3 1.3
Placebo Olz 2.5 mg OLZ 5-10 mg
1.1 1.1 1.1
Y *
1 1 1 1 Y 1
0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9
0.8 0.8 0.8 0.8
0.7 0.7

*
0.3 0.3
0.2

Intense Anger Affective Chronic Identity Paranoid Frantic Efforts Suicidal or self Impulsivity Unstable
Instability Emptiness Disturbance Ideation/ to avoid mutilating that is self interpersonal
Dissociation abandonment behavior damaging relationships
 Making sense of complexity
Assessment Probes Subject Intermediary Item
Type Response Judgment Scored
Scale None
PRO Questions Self Report By Subject

Interview
ClinRO Self Report By Rater
Questions
judgment
Clinical Rater

Interview By Computer
sClinCRO Questions Self Report algorithm
simulating
judgment
Algorithm
Performance
Interpretation Scored by tester
Measure

Sensor based Electronic

metrics Raw data transmission readout
CAN WE USE A PRO FOR BPD?
 BORDERLINE EVALUATION OF SEVERITY OVER TIME (BEST)
• INTERNAL CONSISTENCY AND ITEM ANALYSIS
Cronbach’s α coefficients at baseline for subjects with BPD and comparison subjects were 0.86 and 0.90, respectively, indicating
that test homogeneity was relatively high at baseline). When subjects with BPD were combined with comparison subjects, the test
homogeneity of the baseline scores remained high (α = 0.92). Cronbach’s α coefficient for the borderline subjects was 0.89 after
the first month of treatment, and remained high (0.90 to 0.92) during the 20-week treatment period. Item-total correlations and the
corresponding overall measure of internal consistency, Cronbach’s α coefficient, from these visits indicate that all items are
measuring the same dimension.
• TEST-RETEST RELIABILITY
Correlation between baseline and screening BEST total scores was moderate (r = 0.62, n =130, P < .001). As mentioned earlier,
we expected some test-retest instability due to real changes in borderline symptoms, but this was not seen. There was a mean (SD)
of 53.1 (45.6) days between screening and baseline assessments.
• CONVERGENT AND DISCRIMINANT VALIDITY
At the screening visit, the BEST correlated strongly with the ZAN-BPD score, SCL-90-R total score, the SAS total score, the CGI
severity score, and both the GAS and BDI scores.
The BEST correlated more strongly with the SCL-90-R total score than with any of the other scales at the screening visit, but this
was only a matter of degree because all relationships were significant. At each time point throughout the study (baseline through
week 20), each instrument score remained significantly related to the BEST total score, yet the relationship between the BEST
and the BDI scores produced the highest coefficients (0.67–0.80), while the relationship with the CGI severity score and the SAS
total score produced the lowest (0.33–0.59, and 0.41–0.59, respectively).
• SENSITIVITY TO CLINICAL CHANGE
The BEST total score was sensitive to clinical change that occurred among all subjects with BPD who participated in the STEPPS
treatment study. In Table 3, we present observed and modeled means of BEST total score, CGI Severity, patient-rated global
improvement, and BDI score. The modeled means, obtained using a repeated measures model with first-order
autoregressive covariance structure, are estimates of the mean that would be observed if subjects were not lost to follow-up. The
BEST total score decreased from a mean of 38.7 (SD = 11.3) at baseline to a mean of 32.9 (SD = 12.0) at week 20 of the study.
The overall time effect for visits at baseline through week 20 was significant (P < .001). The CGI severity scale, the patient-rated
global improvement scale, and the BDI were also sensitive to clinical change by week 20 (P < .001). However, follow up contrast
tests which compared each weekly visit to the screening visit revealed that the CGI severity scores were not significantly different
 CAN WE USE A PRO FOR BPD?
DIFFICULTIES IN EMOTIONAL REGULATION SCALE
Please indicate how often the following statements apply to you by writing the appropriate number from the scale below on the line beside each item.

1---------------------------2---------------------------3---------------------------4---------------------------5
almost never Sometimes about half the time most of the time almost always
(0-10%) (11-35%) (36-65%) (66-90%) (91-100%)
__ 19) When I’m upset, I feel out of control.
__ 1) I am clear about my feelings. __ 20) When I’m upset, I can still get things done.
__ 2) I pay attention to how I feel. __ 21) When I’m upset, I feel ashamed at myself for feeling that way.
__ 3) I experience my emotions as overwhelming and out of control. __ 22) When I’m upset, I know that I can find a way to eventually feel better.
__ 4) I have no idea how I am feeling. __ 23) When I’m upset, I feel like I am weak.
__ 5) I have difficulty making sense out of my feelings. __ 24) When I’m upset, I feel like I can remain in control of my behaviors.
__ 6) I am attentive to my feelings. __ 25) When I’m upset, I feel guilty for feeling that way.
__ 7) I know exactly how I am feeling. __ 26) When I’m upset, I have difficulty concentrating.
__ 8) I care about what I am feeling. __ 27) When I’m upset, I have difficulty controlling my behaviors.
__ 9) I am confused about how I feel. __ 28) When I’m upset, I believe there is nothing I can do to make myself feel better.
__ 10) When I’m upset, I acknowledge my emotions. __ 29) When I’m upset, I become irritated at myself for feeling that way.
__ 11) When I’m upset, I become angry with myself for feeling that way. __ 30) When I’m upset, I start to feel very bad about myself.
__ 12) When I’m upset, I become embarrassed for feeling that way. __ 31) When I’m upset, I believe that wallowing in it is all I can do.
__ 13) When I’m upset, I have difficulty getting work done. __ 32) When I’m upset, I lose control over my behavior.
__ 14) When I’m upset, I become out of control. __ 33) When I’m upset, I have difficulty thinking about anything else.
__ 15) When I’m upset, I believe that I will remain that way for a long time. __ 34) When I’m upset I take time to figure out what I’m really feeling.
__ 16) When I’m upset, I believe that I will end up feeling very depressed. __ 35) When I’m upset, it takes me a long time to feel better.
__ 17) When I’m upset, I believe that my feelings are valid and important. __ 36) When I’m upset, my emotions feel overwhelming.
__ 18) When I’m upset, I have difficulty focusing on other things.

Reverse-scored items (place a subtraction sign in front of them) are numbered 1, 2, 6, 7, 8, 10, 17, 20, 22, 24 and 34.

Calculate total score by adding everything up. Higher scores suggest greater problems with emotion regulation.
SUBSCALE SCORING**: The measure yields a total score (SUM) as well as scores on six sub-scales:

Gratz, K. L. & Roemer, L. (2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the Difficulties in Emotion
Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54.
 We can use
a PRO for BPD, but

Cognitive
Perceptual
Distortion

Limits
reliability
of
self-report
Psychiatric symptoms and response quality to self-rated personality tests:
Evidence from the PsyCoLaus study
Marc Dupuisa,⁎, Emanuele Meiera, Dominique Rudazb, Marie-Pierre F. Strippolib,
Enrique Castelaob, Martin Preisigb, Roland Capela, Caroline L. Vandeleurb
a Institute of Psychology, University of Lausanne, Geopolis Building, CH-1015 Lausanne, Switzerland
b Centre for Research in Psychiatric Epidemiology and Psychopathology, Lausanne University Hospital, Department of Psychiatry, Switzerland

ABSTRACT
Despite the fact that research has demonstrated consistent associations between self-rated measures of personality
dimensions and mental disorders, little has been undertaken to investigate the relation between psychiatric symptoms
and response patterns to self-rated tests. The aim of this study was to investigate the association between psychiatric
symptoms and response quality using indices from our functional method.
A sample of 1,784 participants from a Swiss population-based cohort completed a personality inventory (NEOFFI) and
a symptom checklist of 90 items (SCL-90-R). Different indices of response quality were calculated based on the
responses given to the NEO-FFI. Associations among the responses to indices of response quality, sociodemographic
characteristics and the SCL-90-R dimensions were then established. Psychiatric symptoms
were associated with several important differences in response quality, questioning subjects’ ability to provide valid
information using self-rated instruments. As suggested by authors, psychiatric symptoms seem associated with
differences in personality scores. Nonetheless, our study shows that symptoms are also related to differences in terms
of response patterns as sources of differences in personality scores. This could constitute a bias for clinical assessment.
Future studies could still determine whether certain subpopulations of subjects are more unable to provide valid
information to self-rated questionnaires than others.
 Summary of conceptual levels and differences in observed scores.
Level Definition Meaning of individual differences
Trait The latent dimension that an instrument aims to measure, its Differences in a trait are actual differences that a
true score perfect measurement instrument is supposed to
ensure once measurement error and biases are
partialled out.
Responding The voluntary and involuntary strategy adopted to respond to Responding refers to a qualitative process, differences
a questionnaire. in responding are thus difficult to measure.
Responding includes very different patterns; some of them Nevertheless, they imply that two true scores of a
correspond to same trait are expressed in a different way which
response biases: situational social desirability bias, agreeing might not be comparable.
with every proposition of a questionnaire.
completing a questionnaire with little care, etc. Some
patterns result from social, and some from individual
characteristics (i.e. cognitive abilities, age, etc.).

Response The observed score on the latent construct, what the Differences in responses are observed differences.
instrument actually They consist of both potential differences in traits and
measures: the response is an observed score or an answer errors in measurement. Response-level differences
which is provided to a correspond to differences that neither result from
given item. differences in traits nor from differences in
responding, that-is-to-say to differences attributable
to error in
measurement.
Depression Drives Quality of Life
Depression Drives Quality of Life
 Depression Drives Quality of Life

Quality of life is typically highly correlated with depression score

 Depression Drives Quality of Life

Quality of life is typically highly correlated with depression score

Are MADRS and Q-LES-Q independent measures?

 PRO challenge:
Voluntary and involuntary strategy adopted to respond to a questionnaire influences outcome

Response patterns impacted by many factors

-Response biases:
Situational social desirability bias (i.e. faking either good or bad),
Acquiescence (agreeing with every proposition of a questionnaire)
-Effort/Motivation:
Completing a questionnaire with little care (i.e. insufficient effort
responding, time available), etc.
-Social factors (e.g. culture, gender, etc.)
-Individual characteristics (i.e. cognitive abilities, age, etc.).
 BPD RCT Endpoints
Other Clinician Rated Outcome Assessments
-Compare to Zan BPD
Existing: Rater Training and Certification process
Prior study experience and data
Inter-rater reliability – acceptable
Intra-rater reliability- under study
Clinically Meaningful- face validity
Remote assessment: video feasible
Regulatory acceptability-?

-PANSS excited Component

-Modified Overt Aggression Scale
-CGI-S/ CGI-I
 BPD RCT Endpoints
Potential for Sensor based measures
• Heart rate/ Heart rate variability
• Skin Conductance level (SCL), and skin conductance reactivity (SCR)
• Vocal expression
• Facial Expression

"Reduced HRV is associated with a variety of conditions such as diabetic

neuropathy, sepsis, myocardial infarction, and lately it has gained increased
interest in psychiatry due to the connection between autonomic dysfunction
and psychiatric pathologies”.
 Longitudinal Covariance of Resting State Heart Rate Variability and
Borderline Personality Disorder Symptoms in Adolescents with Non-
Suicidal Self-Injury
Julian Koenig1, Sindy Weise2, Lena Rinnewitz2, et al

Background: Resting state high-frequency heart rate variability (HF-HRV) – a potential trait marker of emotion regulation capacity -is
reduced in borderline personality disorder (BPD). In adolescents with non-suicidal self-injury (NSSI), HF-HRV is inversely correlated with
BPD symptoms. The study aimed to investigate if longitudinal changes in BPD symptoms are associated with changes in HFHRV in
adolescents with NSSI over time.
Methods: HF-HRV was recorded in female adolescents with NSSI (n=17) according to DSM-5 section 3 diagnostic criteria who
completed a baseline assessment and a one-year follow-up. Physiological data, structured clinical interviews and self-reports on
psychopathological distress were obtained at both time points. Covariance and predictors of change in clinical outcomes and HF-HRV
were assessed.
Results: Patients showed clinical improvements indicated by a reduction of depressive symptoms (z(34;17) 5 23.74, p,.0001), NSSI
frequency (z(34;17) 5 23.79, p,.0001), and
increases in the level of functioning (z(34;17) 5 2.87, p5.004). No significant differences were observed on resting state HFHRV (z(34;17)
5 20.94, p5.348) recorded at baseline and follow-up. Changes in BPD symptoms were significantly associated with changes in resting
HF-HRV (r(17)52.516, p=.033).
Conclusions: Longitudinal changes in BPD symptomatology in adolescents engaging in NSSI are associated with changes in resting state
HF-HRV, which gives support to HF-HRV as a trait marker of emotion regulation capacity. Results bear promise with respect to the
implementation of measures of HF-HRV in the monitoring of patients and outcome assessment within psychiatric research. Future
clinical studies are necessary to investigate the utility of HF-HRV to track treatment outcome in adolescents with BPD.

Biological Psychiatry May 15, 2017; 81:S140–S276 [Link]/journal

 BPD RCT Endpoints
Potential for Sensor based measures
Can Sensor based endpoint be used as BPD outcomes?
Reliability
Validity
Construct
Convergent
Discriminant
• What would it take to gain approval?
 Custom made Scale: Empirical Mosaic Approach

Symptom Domains Item Source

Subjective Objective
Cognitive-perceptual symptoms
Suspiciousness PANSS
Referential thinking YMRS
Paranoid ideation HAM-D
Illusions PANSS
Derealization DES
Depersonalization CAPS
Hallucination-like symptoms BPRS
Impulsive-behavioral dyscontrol
Impulsive aggression MOAS Heart rate variability
Deliberate self-harm CSSRS
Impulsive sexual behavior Zan-BPD
Substance abuse Zan-BPD Toxicology
Impulsive spending BIS
Affective dysregulation BEST Actigraphy
Mood lability
Rejection sensitivity Zan-BPD Facial/vocal expression
Intense anger out of proportion to the stimuli DERS Heart Rate Variability
Beginning with the Endpoint(s) in mind
Conclusions
Beginning with the Endpoint(s) in mind
Conclusions
• BPD can be reliably diagnosed
• Zan-BPD = Current standard for comparison
• Most experience with several Phase 3 RCTs in progress
• Need data on divergent validity, intra-rater reliability
• Other Endpoints available - varying degrees of specificity and psychometric support
• Subjective
• PRO
• ClinRO
• Objective
• Sensor-based (plausible)
New Scales could provide better coverage of condition and clinical meaningfulness (for patients)
Need to assess concordance across subjective and objective assessments
Beginning with the Endpoint(s) in mind
Conclusions
• BPD can be reliably diagnosed
• Zan-BPD = Current standard for comparison
• Most experience with several Phase 3 RCTs in progress
• Need data on divergent validity, intra-rater reliability
• Other Endpoints available - varying degrees of specificity and psychometric support
• Subjective
• PRO
• ClinRO
• Objective
• Sensor-based (plausible)
New Scales could provide better coverage of condition and clinical meaningfulness (for patients)
Need to assess concordance across subjective and objective assessments
• What would it take to get an outcome measure accepted this indication?
• Could new scales be based on items most sensitive to change (like MADRS)