NEW RESEARCH |

Emotion Regulation Training for Adolescents With

Borderline Personality Disorder Traits: A Randomized
Controlled Trial
H. Marieke Schuppert, M.D., Ph.D., Marieke E. Timmerman, Ph.D., Josephine Bloo, Ph.D.,
Tonny G. van Gemert, M.D., Herman M. Wiersema, M.D., Ruud B. Minderaa, M.D., Ph.D.,
Paul M.G. Emmelkamp, Ph.D., Maaike H. Nauta, Ph.D.

Objective: To evaluate the effectiveness of Emotion Regulation Training (ERT), a 17-session weekly
group training for adolescents with borderline personality disorder (BPD) symptoms. Method: One
hundred nine adolescents with borderline traits (73% meeting the full criteria for BPD) were
randomized to treatment as usual only (TAU) or ERT þ TAU. Outcome measurements included
severity of BPD symptoms, general psychopathology, and quality of life. Multilevel analyses were
conducted on an intent-to-treat basis. Clinical significant change was determined by normative
comparisons on a primary outcome measurement. Results: Independent of treatment condition, the
two groups improved equally on the severity of BPD symptoms, general psychopathology, and quality
of life. Nineteen percent of the ERT group was remitted according to the cutoff score after treatment (at
6 months) versus 12% of the control group. Follow-up assessments in the ERT group at 12 months
showed some further improvement (33% remittance). With regard to predictors of outcomes,
adolescents with higher levels of depression or attention-deficit/hyperactivity disorder or
oppositional-defiant disorder at baseline and who reported a history of abuse had worse outcomes,
regardless of treatment condition. The attrition rate for the ERT sessions was remarkably low
(19%). Conclusions: Early interventions for BPD symptoms in adolescence are feasible and
necessary. No additional effect of ERT over TAU could be demonstrated in the present study. There
is a clear need for developing effective interventions for adolescents with persistent BPD
symptomatology. Clinical trial registration information—Evaluation of Group Training for Adolescents
(Emotion Regulation Training) with Emotion Regulation Problems: A Randomized Controlled Clinical
Trial; http://trailregister.nl/; ISRCTN97589104. J. Am. Acad. Child Adolesc. Psychiatry; 2012;
51(12):1314-1323. Key Words: borderline personality, randomized controlled trial, treatment,
adolescents.

B
orderline personality disorder (BPD) is a
complex and severe disorder that usually
has its onset in adolescence.1 Diagnosing
BPD in adolescence has long been controversial,
despite the growing body of evidence of a valid
and reliable diagnosis before 18 years of age.2,3
There is convincing evidence for the continuity of
BPD from adolescence into adulthood.4 Early
symptoms of BPD are associated with several
serious functional and psychopathologic problems
This article is discussed in an editorial by Dr. David J. Miklowitz on
page 1238.
Supplemental content cited in this article is available online.
in the long term.4 BPD in adolescents has been
found to be a better predictor than Axis I disorders
for psychopathology and psychosocial dysfunction-
ing later in life.5 Furthermore, borderline symptoms
in adolescence are a predictor for social impairment
and lower life satisfaction, lower academic and
occupational functioning, less partner involvement,
and a higher consumption of health care services at
20-year follow-up.4
Although there are no reliable figures, the
prevalence of BPD in adolescence is estimated
at 1% to 3%.6,7 This figure increases to 10% to 14%
when milder cases are included6 or when self-
reports are used.8
Despite the high prevalence and adverse con-
sequences of BPD symptoms in the long term,

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

1314 www.jaacap.org VOLUME 51 NUMBER 12 DECEMBER 2012

only few treatment protocols have been devel-
oped and evaluated for adolescents. The available
interventions are rather intensive and therapists
generally need extensive training (e.g., Cognitive
Analytic Therapy [CAT], Dialectic Behavior Ther-
apy for Adolescents [DBT-A], Mentalization-
Based Treatment for Adolescents [MBT-A]). Of
those, only CAT has been evaluated in a rando-
mized controlled trial (RCT). In this RCT, 86
youngsters with BPD symptoms (15–18 years
old) were randomized to Good Clinical Care or
to CAT.2 CAT is comprised of 16 to 24 individual
sessions of psychotherapy based on elements of
psychoanalytic object relations theory and cogni-
tive psychology.9 These interventions showed
equal and significant improvements over a 2-
year period (Cohen d ¼ 0.54–1.38). The rate
of improvement was (moderately) faster for
the secondary measures, but there were no
differences between Good Clinical Care and
CAT on BPD symptoms.
Although DBT has frequently been evaluated in
adult samples, with good results,10 the adolescent
version has been evaluated only in nonrando-
mized, small samples.11,12 DBT focuses on (para)-
suicidal behavior, therapy-interfering behaviors,
and other dangerous or destabilizing behaviors.
The DBT-A consists of 16 weekly multifamily
sessions and family therapy can be added. All
studies found a decrease in (para)suicidal beha-
vior and/or depressive symptoms, with effect
sizes (Cohen d) ranging from 0.23 to 3.40.
The last treatment available for youngsters is
the MBT-A.13 Mentalizing is the capacity to make
sense of others and of oneself, to be aware, and to
understand the subjective states and mental
processes of oneself and of others. Patients with
BPD are considered to have a fragile mentalizing
capacity, which makes them vulnerable in inter-
personal relationships.14 The MBT-A has not yet
been evaluated.
All these interventions require extensive addi-
tional training for therapists. Moreover, the DBT-
A and MBT-A are directed at adolescents with
severe BPD symptoms and are time intensive.
However, early intervention might prevent
adverse outcomes in the long term.4,5 Therefore,
the authors developed low-threshold care, not
only for adolescents with full-syndrome BPD but
also for (referred) subsyndromal cases, which is
time limited and easy to implement in general
mental health care. Emotion Regulation Training
(ERT) is a manual-based group training for
JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY
VOLUME 51 NUMBER 12 DECEMBER 2012
EMOTION REGULATION TRAINING
adolescents (14–19 years old) with BPD traits15
and was developed as an add-on to treatment as
usual (TAU). The training is an adaptation of the
Systems Training for Emotional Predictability and
Problem Solving developed by Bartels et al.16
Problems in emotion regulation are often consid-
ered to form the core symptom of BPD.17,18 ERT
focuses on this symptom, using the structure of
the Systems Training for Emotional Predictability
and Problem Solving, complemented with ele-
ments of DBT skills training and cognitive-
behavior therapy. Age-specific adaptations are
the duration of the program (17 weeks), the length
of the sessions (105 minutes), and specific topics
to meet the developmental stage of self-
exploration.19 ERT has been evaluated in a ran-
domized pilot study.20 Forty-three adolescents
were randomized to TAU (n ¼ 20) or to a
combination of TAU and ERT (ERT þ TAU; n ¼
23). The two groups showed an equal significant
decrease in borderline symptoms over a 6-month
period. The ERT þ TAU group improved sig-
nificantly more on locus of control than the TAU
group. The ERT protocol has been adapted
according to the findings of the pilot study.15
The present study evaluated the effectiveness
of ERT at a larger scale. The study was conducted
in four mental health centers in the Netherlands.
Adolescents (N ¼ 109) were randomized to TAU
or ERT þ TAU. The authors hypothesized that
adolescents in the ERT þ TAU group would
improve more on borderline symptoms, general
psychopathology, quality of life, and locus of
control than those in the TAU group.
Next, the authors explored the predictive value
of a history of abuse and/or trauma and depressive
and externalizing symptoms. There is a paucity of
studies that have addressed the identification of
predictors in BPD treatment and hardly any in
adolescents. Barnicot et al.21 reported a systematic
review and meta-analysis on 41 treatment studies,
three of which concerned adolescents. Predicting
factors for dropout were a commitment to change,
impulsivity, and therapeutic alliance. In the present
study, the authors examined predicting factors for
the effectiveness of ERT.
METHOD
Sample
Participants were 109 adolescents 14 to 19 years old
(mean ¼ 15.98 years, SD ¼ 1.22 years) who were referred
for emotion regulation problems and/or BPD features to
www.jaacap.org 1315
SCHUPPERT et al.

FIGURE 1 Flowchart of assessment and randomization. Note: ERT ¼ Emotion Regulation Training.

Assessed for eligibiliy

N=133

Excluded n=24
• 6 lack of motivation
• 1 refused to sign informed consent
Randomized • 17 not enough enrollment on
N=109 location for randomization

Control Group Intervention (ERT)

n=55 n=54

Second
assessment Completed Did not
n=50 intervention complete
n=44 intervention
Second n=9
assessment Second
n=43 assessment
n=6

Follow-up Follow-up
n=38 n=3

one of four mental health centers in the northern
Netherlands. Recruitment took place from November
2007 through February 2010. The initial diagnostic
procedures were completed as customary for the center
concerned. In cases of emotion regulation problems or
BPD symptomatology as the main problem area, the
adolescents were referred to trained clinicians for
complementary screening. Next, the adolescents and
their parents or caretakers were referred for assessment
to an independent research psychologist who was
blinded to the treatment allocation. To participate in
the study, subjects had to meet at least two BPD criteria
(according to the Structured Clinical Interview for DSM-
IV Personality Disorders [SCID-II]).22 Anxiety disor-
ders, mood disorders, or attention-deficit/hyperactivity
disorder (ADHD) could be present as comorbid dis-
orders but were not the primary diagnosis. Exclusion
criteria were psychotic disorders, conduct disorder, and
substance dependence as assessed by the Schedule for
Affective Disorders and Schizophrenia for School-Age
Children (K-SADS),23 and an IQ below 80 as estimated
from the educational level.
Procedure
Figure 1 shows the flowchart of randomization
and assessments. Randomization took place in each
participating center until the center had 12 to 18
participants included in the study (with informed
consent and baseline assessments completed). To
improve the representativeness of the sample, stratified
sampling was applied by randomizing the adolescents
with a score of 15 or more on the total scale of a
structured interview on BPD severity (Borderline Per-
sonality Disorder Severity Index [BPDSI]; see Measure-
ments)24 and then the adolescents with a lower score.
Randomization by drawing lots was performed by an
independent, masked research assistant who was not
involved in the project.
Assessments were accomplished at baseline, after
treatment (i.e., after ending ERT or after a comparable
period), and 6-month follow-up (for the ERT þ TAU
group only). Adolescents in the TAU group were not
assessed at follow-up, because they were allowed to
enter ERT after the treatment assessment. Each parti-
cipant and parent/caretaker received a gift voucher of
h5 after each assessment.
Three independent research psychologists com-
pleted the assessments. Ratings on taped interviews
(15%) were made by a clinical psychology student and a
psychologist. Interviewers were trained during a half-
day course. Next, they observed two to three interviews
conducted by an experienced interviewer, and they had
regular consensus meetings. The intraclass correlation

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

1316 www.jaacap.org VOLUME 51 NUMBER 12 DECEMBER 2012

coefficient for the subscales proved to be excellent
(0.98–1.00, mean ¼ 0.98), with an exception on the
subscale Identity Disturbance (0.89).
Interventions
Emotion Regulation Training (ERT) is a manual-
based group training for adolescents (14–19 years old)
with BPD features.15 The focus of ERT is to improve the
feeling of control over intense, strong emotions by
increasing cognitive, social, and behavior coping skills.
ERT is based on Cognitive Behavior Therapy (e.g., chain
analysis, homework forms, cognitive restructuring)
and elements of Dialectical Behavior Therapy,17 such
as psychoeducation on emotion regulation and
mindfulness-based relaxation exercises. ERT consists
of 17 weekly sessions of 105 minutes, followed by two
booster sessions at 6 and 12 weeks after the weekly
course. A detailed session-by-session outline is pre-
sented in Table S1 (available online).
TAU consisted of (a combination of) pharmacother-
apy, individual psychotherapy (mostly with [elements
of] Cognitive Behavior Therapy), counseling, family
therapy, inpatient psychiatric care, and emergency care
in case of self-mutilation or suicidal behavior.
Therapists
For the present study, 13 therapists for the ERT
groups were selected by the staff of the mental health
center for which they worked. Ten therapists held a
master’s degree and three therapists held a bachelor’s
degree with postgraduate training. All therapists had
previous therapy experience in psychiatry (mean ¼ 12.4
years, range ¼ 3–34 years), with at least 2 years of
experience in therapy with adolescents with borderline
features. At least one of the two therapists per ERT
group was a licensed cognitive behavior therapist or
clinical psychologist. Therapists received one-day
training in the ERT program by the author(s) of the
ERT program and had supervision sessions by tele-
phone or e-mail contact as needed. Next, the therapists
had two supervision sessions at the site during the
weekly course. The therapists used a therapist manual
with a detailed session-by-session outline next to the
manual for the adolescents. The ERT therapists were
not the individual therapists of the adolescents.
To keep the study as naturalistic as possible, there
were no specific criteria for the therapists in the TAU
condition. The number and type of contacts in TAU was
checked in retrospect by the first author.
To increase treatment adherence and comparability
among the centers, the manuals were highly structured.
The treatment integrity of a random sample of 20
audiotaped sessions was checked by an independent
rater. The check for treatment adherence focused on
adherence to the content of the program and on the
competence of the therapists. The competence of all
JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY
VOLUME 51 NUMBER 12 DECEMBER 2012
EMOTION REGULATION TRAINING
therapists was found to be accurate. On average, the
sessions covered 92.4% of the ERT manual.
Measurements
The assessments were conducted by research psy-
chologists who were blinded to the treatment condition.
The assessments (interviews and questionnaires) took
2.5 to 3 hours. In some cases, the second and third
assessments were assessed at home or by telephone.
The SCID-II BPD section22 consists of nine items in
accord with the DSM-IV criteria. The instrument was
developed for adults but is frequently used in adoles-
cents.2 The SCID-II BPD section was used to character-
ize the sample in borderline pathology level.
Different modules of the SADS -Present and Lifetime
version were used.23 The modules disruptive behavior
disorders and psychotic disorders were administered at
baseline to obtain information on the exclusion criteria.
The modules traumatic experiences and posttraumatic
stress disorder were assessed after treatment and were
used to obtain additional information on trauma as a
possible predictor for therapeutic change (indicated as
‘‘abuse’’).
The BPDSI (also referred to as the BPDSI-IV-ado)24 is
a semistructured interview that consists of 72 items,
spread over the nine criteria for BPD in the DSM-IV,1
and is a reliable and valid instrument for the assessment
of the severity of borderline symptoms in adolescents in
the previous 3 months. The total scale (range ¼ 0–90)
and the subscale Affective Instability (range ¼ 0–10)
were used as the primary outcome measurements.
The Symptoms Checklist-90-R25 contains 90 items
that assess general psychological complaints. The
questionnaire is used frequently and shows good
psychometric properties.25 The total scale was added
as a secondary outcome measurement.
The Youth Quality-of-Life—Research Version is a
validated questionnaire that contains 56 items, divided
in 41 perceptual items and 15 contextual items.26 Higher
scores indicate a better quality of life. The Perceptual
subscale was used as a secondary outcome measurement.
The Life Problems Inventory, subscale Emotional
Dysregulation27 (15 items), was added as secondary
outcome measurement as a self-report on emotional
dysregulation. The psychometric properties of the Life
Problems Inventory were found to be good.27
The Multidimensional Emotion Regulation Locus
of Control is based on the Multidimensional Health
Locus of Control.28 The subscale Intern (six items) was
included as a secondary outcome measurement. Higher
scores indicate a better locus of control.
The Children’s Depression Inventory29 is a 27-item
questionnaire developed for children and adolescents
to measure symptoms of depression in the previous
2 weeks. The psychometric properties have been shown
to be good.29 The total score was used as a predictor.
www.jaacap.org 1317
SCHUPPERT et al.
The Swanson, Nolan, and Pelham Rating Scale is a
frequently used parent report in ADHD research.30 It
addresses ADHD and oppositional-defiant disorder
symptoms as described in the DSM-IV.1 The psycho-
metric properties were found to be good,30 and the total
score was used as a predictor.
Statistical Analyses
Analyses were conducted according to the intent-to-
treat principle, with multilevel analysis in the statistical
program MLwiN (University of Bristol, UK),31taking
into account all the available data at all data points. Six
models were built, one each for BPDSI total score, BPDSI
affective instability, Symptoms Checklist-90-R, Youth
Quality-of-Life—Research Version, Life Problems
Inventory (subscale Emotional Dysregulation), and
Multidimensional Emotion Regulation Locus of Con-
trol. In the multilevel model, the statistical significance
of the fixed effects was tested using the approximate t
test, and of the random effects with the deviance test,
with the significance level set at .05. The modeling
strategy was as follows. First, dummy variables were
used to represent the condition effect and the time effect
for after treatment and follow-up (coded such that each
parameter expressed the change between the measure-
ment concerned and the baseline measurement). The
effects of condition, time, and its interaction were of
primary interest and therefore retained in all models.
Second, to explore the possible effects of relevant
individual characteristics, some moderators were tested
as fixed effects, namely abuse, depression, and ADHD,
and their interactions with time and/or condition. The
variables were preserved in the model only when
significant. The continuous moderator variables were
centered on the sample mean, implying that the para-
meter estimates for the effects of condition, time, and its
interaction reflect the effects for a patient with a score
equal to the sample mean of the moderator variables in
the model concerned. The between-individual and
within-individual variances were estimated as random
effects. To examine the possible differential effects for
the more homogenous threshold BPD and completers,
the multilevel analyses were repeated for threshold BPD
and completers.
Effect sizes were calculated using the Cohen d
(Cohen d ¼ mean 1 – mean 2/s pooled), baseline to
after intervention for the two groups, baseline to follow-
up intervention for the ERT group, and after interven-
tion between the groups.
Clinical significant change was determined through
normative comparisons on borderline severity. A cutoff
of 6 was used based on a sample of control adolescents
described in a previous study on this outcome
measurement.24
To examine possible differential effects for the more
homogenous threshold BPD and completers, multilevel
analyses were repeated for threshold BPD and completers.
JOURNAL
1318 www.jaacap.org
RESULTS
Participants
Figure 1 presents a flowchart of assessment and
randomization. Initially, 133 adolescents were
referred for ERT, and 109 adolescents were rando-
mized to one of the two conditions. The mean
number of DSM-IV criteria was 6.17 (range ¼ 3–9);
73% fulfilled five or more criteria. Nine ERT groups
were performed. The mean number of therapy
sessions was 12.1 (SD ¼ 4.8, range ¼ 0–17). The
attrition rate was 19% (defined as having attended
less than half the training [nine sessions]).
Participants were free to use mental health care
services, and the frequency of contacts from
baseline to after the intervention was equal across
conditions (F ¼ 2.0, p ¼ .84). The mean numbers of
individual contacts were 5.2 (SD ¼ 5.9; range ¼ 0–
22) and 3.2 (SD ¼ 4.4, range ¼ 0–18) for family
contacts. The number of participants admitted to
inpatient care or daycare was equal (six in the ERT
þ TAU group and eight in the TAU group, w2
analysis).
Demographic Characteristics
There were no significant differences between
conditions or among the four sites for the main
characteristics (w2 analysis): age (mean ¼ 15.98
years, SD ¼ 1.22 years), gender (96% female),
divorce of the parents (49% divorced), contacts
with the justice system (30%), or ethnicity (19%
with a non-Dutch parent).
The use of addictive drugs and psychotropic
medication was considerable (29% in the two
groups). Alcohol misuse was defined as excessive
according to legislation (no alcohol before 16 years
of age) or recommendations of the Dutch govern-
ment (no more than two alcoholic drinks [male] or
one alcoholic drink [female] on average and no
binge drinking). Only 34% of the youngsters did
not use any alcohol, whereas 33% reported social
use and 33% reported excessive use. There were
no significant differences between groups with
regard to addictive drugs, alcohol, and medica-
tion at baseline or follow-up (Table S2, available
online; medication use). Fifty-five percent of the
adolescents reported a history of physical and/or
sexual abuse, with no significant differences
between groups.
Effectiveness of ERT
Results of the primary and secondary outcome
measurements are presented in Tables 1 and 2,
OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY
VOLUME 51 NUMBER 12 DECEMBER 2012
 EMOTION REGULATION TRAINING

TABLE 1 Pre- and Postintervention and Follow-Up (FU) Scores (Means [SD]) for Outcome Measurements
Effect Size
(Cohen d )a

Treatment Pre Post FU

Measure Group Mean (SD) n Mean (SD) n Mean (SD) n Pre-Post Pre-FU

BPDSI—total ERT þ TAU 18.23 (9.67) 54 13.29 (9.53) 48 11.27 (8.98) 36 0.51 0.75
score
TAU 20.35 (11.16) 55 15.39 (9.00) 49 0.49
BPDSI—affective ERT þ TAU 5.22 (2.57) 54 3.88 (2.37) 48 3.55 (2.19) 36 0.54 0.70
instability
TAU 5.31 (2.53) 55 4.29 (2.48) 49 0.41
SCL-90-R ERT þ TAU 209.43 (64.89) 49 190.19 (73.58) 48 170.98 (66.58) 38 0.29 0.58
TAU 216.72 (63.67) 54 193.00 (63.63) 48 0.37
YQOL-R ERT þ TAU 53.83 (12.29) 51 54.51 (13.17) 48 59.61 (11.74) 37 0.05 0.48
TAU 52.93 (12.19) 55 54.72 (13.31) 49 0.14
LPI-ed ERT þ TAU 40.38 (10.86) 51 32.87 (11.61) 46 28.36 (13.67) 37 0.67 0.97
TAU 41.77 (11.15) 55 36.37 (11.63) 49 0.47
MERLC-intern ERT þ TAU 19.29 (5.40) 51 21.33 (6.54) 46 20.16 (7.56) 38 0.34 0.13
TAU 19.74 (5.38) 55 20.27 (6.82) 49 0.09

Note: The number differs owing to missing values. ERT ¼ Emotion Regulation Training; SCL-90-R ¼ Symptoms Checklist 90-Revised; TAU ¼ Treatment as Usual.
a
Effect sizes (Cohen d) between groups after the intervention: Borderline Personality Disorder Severity Index-IV–adolescent version (BPDSI total) ¼ 0.23;
BPDSI affective instability ¼ 0.17; SCL90 ¼ 0.04; Youth Quality-of-life Research Version (YQOL-R) ¼ 0.02; Life Problem’s Inventory emotional
dysregulation (LPI-ed) ¼ 0.13; Multidimensional Emotion Regulation Locus-of-control intern (MERLC-intern) ¼ 0.16.

respectively. In summary, there were no signifi-
cant differences between groups on any measure-
ment. The two groups showed improvement from
baseline to after the intervention on all measure-
ments except quality of life. Also, the ERT group
showed significant improvement from after the
intervention to follow-up.
Severity of Borderline Symptoms. Tables 1 and 2
indicate that the mean total scores of the BPDSI
and subscale Affective Instability decreased sig-
nificantly from baseline to after the intervention.
However, there were no significant differences
between the ERT þTAU and TAU groups.
Secondary Outcome Measurements. Tables 1 and 2
present a significant improvement in general
psychopathology and emotional dysregulation
from baseline to after the intervention, but not
in quality of life and locus of control. No sig-
nificant differences between the two groups
were found.
A further significant improvement over time
was found in the ERT þ TAU group at follow-up.
No comparison with the TAU condition could be
made because the follow-up assessment was
conducted only in the active condition.
Table 2 also presents the results of the search for
predictors of change. It was investigated whether
depressive symptoms, symptoms of ADHD, and
a history of abuse were associated with improve-
ment on the outcome measurements. More
depressive symptoms at baseline were found to
be associated with a lesser decrease of borderline
severity and general psychopathology after inter-
vention, but this was unrelated to the condition.
After treatment and at follow-up, having more
depressive symptoms (at baseline) was associated
with higher borderline severity and general psy-
chopathology and at follow-up with less quality
of life. Further, having more ADHD symptoms
was associated with less quality of life, more
emotional dysregulation, and more locus of con-
trol at follow-up (ERT þ TAU group only). Also, a
history of abuse was associated with less
improvement in borderline severity and general
psychopathology after treatment and less
improvement in general psychopathology and
quality of life at follow-up. The number of
individual and/or family sessions in the TAU
group was not associated with outcome.
All analyses were repeated with threshold BPD
cases only and completers only. The analyses
showed similar effects in the completer group
as in the full sample, except for emotional dysre-
gulation, where completers in the ERT þ TAU

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 51 NUMBER 12 DECEMBER 2012 www.jaacap.org 1319
SCHUPPERT et al.

TABLE 2 Parameter Estimates (SE) of Multilevel Models of Outcome Measurements

BPDSI Total BPDSI Affective MERLC-

Score Instability SCL-90-R YQOL-R LPI-ed intern

Fixed effects
Intercept (baseline 19.18 5.26 (0.20)*** 212.68 53.33 41.21 19.51
score) (0.73)*** (4.63)*** (0.88)*** (0.94)*** (0.58)***
Post-treatment 6.14 1.08 35.66 2.41 5.46 0.65
(1.19)*** (0.29)*** (8.30)*** (1.25) (1.36)*** (0.78)
Follow-up 8.27 1.78 59.46 8.84 11.94 1.21
(1.09)*** (0.33)*** (10.48)*** (2.00)*** (1.57)*** (0.89)
Post-treatment 1.57 0.27 11.16 1.47 2.65 1.19
 condition (1.30) (0.38) (8.90) (1.64) (1.80) (1.05)
CDI (baseline) 0.84 0.15 6.34 1.06 0.75 0.14
(0.09)*** (0.02)*** (0.54)*** (0.09)*** (0.09)*** (0.06)*
Post-treatment  CDI 0.21 1.26 (0.62)*
(0.09)*
Post-treatment
 CDI  condition
Follow-up  CDI 0.38 3.77 0.41
(0.14)** (0.88)*** (0.15)**
SNAP (baseline)
Post-treatment 
SNAP
Post-treatment 
SNAP  condition
Follow-up  SNAP 2.31 1.97 1.26
(0.79)** (0.90)* (0.50)**
Abuse (baseline)
Post-treatment  3.45 21.40 (8.87)*
abuse (1.27)**
Post-treatment 
abuse  condition
Follow-up  abuse 30.08 5.48
(13.66)* (2.53)*
Random effects
Between-individual 32.08 1.97 (0.44)*** 920.66 37.94 41.07 18.60
variance (6.19)*** (221.53)*** (8.35)*** (9.49)*** (3.77)***
Residual variance at 25.90 2.45 (0.30)*** 1302.76 44.24 53.43 17.20
measurement (3.17)*** (160.49)*** (5.40)*** (6.57)*** (2.11)***
occasions

Note: Values are presented as estimate (standard error). Empty cells indicate data not included in the model (because of nonsignificance). Abuse ¼ history of
sexual and/or physical abuse; BPDSI ¼ Borderline Personality Disorder Severity Index-IV—Adolescent Version; CDI ¼ Children’s Depression Inventory;
condition ¼ Emotion Regulation Training plus treatment as usual or treatment as usual; LPI-ed ¼ Life Problem’s Inventory, Emotional Dysregulation subscale;
MERLC-intern ¼ Multidimensional Emotion Regulation Locus-of-Control, Intern subscale; SCL-90-R ¼ Symptoms Checklist-90–Revised; SNAP ¼ Swanson,
Nolan, and Pelham Attention Deficit and Hyperactivity Disorder Measure; YQOL-R ¼ Youth Quality-of-Life Research Version.
*p o .05, **p o .01, ***p o .001.

group showed significantly greater improvement
after treatment than in the TAU group (Table S3,
available online).
Analysis of Noncompleters
Table 3 presents the differences between completers
and noncompleters. In general, adolescents who
did not complete the intervention reported less
symptomatology at baseline and a higher quality of
life. Multilevel analyses for treatment effectiveness
were repeated without the noncompleters. The
results indicated the same pattern of results, with
no additional positive effect of the ERT intervention
on all outcome measurements.

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

1320 www.jaacap.org VOLUME 51 NUMBER 12 DECEMBER 2012

TABLE 3 Means (SD) Completers/Noncompleters of Outcome Measurements at Baseline; Mann-Whitney Test
Completers
Measure (n ¼ 41–44)
BPDSI total 19.26
BPDSI affective instability 5.51
SCL-90-R 216.48
YQOL-R 52.26
Note: Values are presented as mean (SD). The number differs owing to missing values. BPDSI ¼ Borderline Personality Disorder Severity Index-IV–Adolescent
Version; SCL-90-R ¼ Symptoms Checklist-90–Revised; YQOL-R ¼ Youth Quality-of-Life Research Version.
a
Mann-Whitney test.
*p ¼ .05, **p o .05.
Remission: Normative Means
Remission was defined as having an endpoint of
6 or lower on the BPDSI total score (BPD severity
in previous 3 months; Figure 2). A minority of
participants in the two groups reached remission:
19% of youth in the ERT þTAU group (n ¼ 9 of 48)
versus 12% in the TAU group (n ¼ 6 of 49) were
remitted. At 6-month follow-up, 67% of youths
(n ¼ 24 of 36) remained in the normative range of
borderline symptomatology.
DISCUSSION
To the authors’ knowledge, the present study
provides an account of the largest sample in a
treatment outcome study in adolescents with BPD
features and is the second published RCT for this
group.2 This study was conducted in general
mental health institutes, with few exclusion cri-
teria, thus increasing its external validity. Most
FIGURE 2 Borderline Personality Disorder Severity Index
(BPDSI-IV) mean values (total score) at baseline and after
intervention. Note: ERT ¼ Emotion Regulation Training; TAU
¼ treatment as usual.
JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY
VOLUME 51 NUMBER 12 DECEMBER 2012
EMOTION REGULATION TRAINING
Noncompleters
a
(n ¼ 8–10) U Test
(10.31) 13.70 (3.90) 151.00
(2.64) 3.92 (1.83) 133.00*
(63.50) 173.28 (63.61) 99.00
(12.34) 62.26 (8.28) 84.00**
adolescents (73%) fulfilled the full BPD criteria
according to the DSM-IV. In addition, 62.4% of
the adolescents had a minimum score of 16 on
the Children’s Depression Inventory, which is
the cutoff score to screen for depression.29 The
attrition rate was low. The present results can be
summarized as follows. First, adolescents
improved over time in symptoms of affective
instability, borderline symptomatology, and
general psychopathology, with moderate effect
sizes after treatment and moderate to high effect
sizes at follow-up. In addition, they reported a
higher quality of life at follow-up, with a moder-
ate effect size. Second, symptoms decreased
regardless of treatment condition. Third, the
authors were unable to identify subgroups
of adolescents who did benefit from the interven-
tion. Fourth, adolescents reporting more depres-
sive symptomatology or with a history of physical
and/or sexual abuse reported a lesser decrease of
complaints. This appeared unrelated to the treat-
ment condition and may merely reflect the pre-
dictors of the course of symptoms rather than the
predictors of treatment outcome. Fifth, noncompl-
eters and dropouts were characterized by less
symptomatology before the test. Completers
showed greater improvement in emotional dys-
regulation after treatment. Sixth, although the
decrease of complaints was in the moderate to
high range, most adolescents were still not within
the normative range of complaints after treatment
(85%) and at follow-up (67%). Research in adult
patients with BPD on intensive treatments, such
as Schema Focused Therapy, Transference
Focused Psychotherapy, and DBT, found compar-
able improvements after 6 months of
treatment.32,33
A significant improvement over time in BPD
symptoms and distress, after a 6-month treatment
www.jaacap.org 1321
SCHUPPERT et al.
period, is a positive and hopeful finding. It
implies that at least some adolescents profit from
the intervention period or that BPD symptoms
improve moderately over time and are less stable
than expected.
The present finding, that a relatively short
intervention does not seem to have substantial
additional benefits, is partly in line with the
authors’ pilot study20 and with a study by Chanen
et al.2 who found no added value of Cognitive
Analytic Therapy above Good Clinical Care.
These results require some considerations. First,
a recovery from BPD symptoms might require a
more intensive, structured, and long-term inter-
vention, such as evidence-based interventions
developed for adult patients with BPD.10,17,32–34
Moreover, there is an increasing notion that brief
interventions for BPD in adults may even have
iatrogenic effects.35 Second, ERT was developed
from different theoretical models that may be
inadequate. Much remains unclear on the devel-
opment and predictors of (early-onset) BPD. The
results of ERT might improve with an explicit
theoretical model based on new evidence about
the developmental pathways to BPD, including
the role of parental rearing,36 depression, and
posttraumatic stress disorder. Third, the attrition
rate was low and the participants under ERT
frequently mentioned peer relationships as an
important motivation to keep coming. However,
the group format leaves little room for individual
tailoring. Fourth, noncompleters showed an
improved quality of life and fewer symptoms at
baseline. This group may not have enough inter-
nal motivation for a weekly intervention. Fifth,
because of time constraints, the authors did not
assess Axis I disorders at baseline or sexual abuse
or trauma, leaving this to the individual therapist
before referral. Some adolescents might have
benefited more from specific treatments for
depression or posttraumatic stress disorder. Sixth,
although ERT focuses on the family/system of the
patient, this is limited. The intervention might
profit from an extension of parent/caretaker
involvement in a simultaneous group or in a
combination of ERT and systems therapy.
To the authors’ knowledge, this is the largest RCT
on the treatment of BPD symptoms in adolescents.
With a sample size of 109, the study was not
underpowered and small to medium effect sizes
between conditions should have been detected. The
low attrition rate (19%) in this study is remarkable
compared with other studies in adolescents (54% in
JOURNAL
1322 www.jaacap.org
CAT2 and 38% in DBT-A).12 The low attrition rate in
this study can partly be explained by peer contact,
as mentioned by the adolescents.
A strength of this study is the low dropout rate
in the assessment: 91% participated in the second
assessment and 76% in the third assessment.
ERT aims at several changes in the individual,
namely improved cognitive restructuring, improve-
ment of locus of control, improvement of emotion-
regulating strategies, and improvement in lifestyle.
In the present study, the authors did not investigate
the presumed responsible therapeutic mechanisms
underlying these changes, such as improved emo-
tion regulation and improved cognitive restructur-
ing. Therefore, they do not know whether ERT was
unable to change those factors or whether an
accomplished change in fact did not lead to changes
in borderline psychopathology.
Although there is a growing body of evidence for
effective interventions in adult patients with BPD,
research in this field for adolescents remains scarce.
This study may encourage clinicians to realize that
early intervention in adolescents with BPD symp-
toms is necessary: most adolescents still had a
borderline symptomatology even after 12 months
of treatment. The outcome was more negative for
adolescents with a history of abuse, which suggests
that trauma-focused therapy may be an important
add-on for this group of adolescents. The low
attrition rate shows that the treatment was feasible,
which may be seen as an indication of the clinical
need for specific treatment modules for BPD
symptoms in adolescence.
There is a strong need to identify effective
treatments for this group of youngsters. &
Accepted September 13, 2012.
Drs. Marieke Schuppert, van Gemert, Wiersema, and Minderaa are
with the University Medical Center Groningen. Dr. Timmerman and
Nauta is with the University of Groningen. Dr. Bloo is with the University of
Maastricht. Dr. Emmelkamp is with the University of Amsterdam.
This research received support from the Netherlands Organization for
Health Research and Development (Zon-Mw) grant 10-000-2030
(H.M.S.).
Dr. Timmerman served as the statistical expert for this research.
The study took place in the Netherlands in four centers: University
Medical Center Groningen, Geestelijke gezondheidsinstelling (GGZ)
Friesland, Jonx Groningen, and Dimence Zwolle. GGZ Groningen and
Adhesie Deventer gave permission to adapt the Dutch version of the
Systems Training for Emotional Predictability and Problem Solving for
adolescents.
The authors thank all participating adolescents, their parents, and all
therapists. They acknowledge Jaap Ringrose of GGZ Drenthe for his
contribution to develop the ERT, and thank Iris Rooke of Document-
Based Care Groep BV and Annelies Wolters and Evelien Miedema of
OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY
VOLUME 51 NUMBER 12 DECEMBER 2012

Accare, University Centre for Child and Adolescent Psychiatry, for their
tremendous work in the data collection.
Disclosure: Drs. Schuppert, van Gemert, and Wiersema are authors of
the manual on Emotion Regulation Training that is commercially
available in the Netherlands. Drs. Timmerman, Bloo, Minderaa,
Emmelkamp, and Nauta report no biomedical financial interests or
potential conflicts of interest.
REFERENCES
1. American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders. 4th ed, text rev. Washington DC:
American Psychiatric Association; 2000.
2. Chanen AM, Jovev M, Djaja D, et al. Screening for borderline
personality disorder in outpatient youth. J Pers Disord. 2008;
22:353-364.
3. Johnson JG, Cohen P, Skodol AE, Oldham JM, Kasen S, Brook JS.
Personality disorders in adolescence and risk of major mental
disorders and suicidality during adulthood. Arch Gen Psychiatry.
1999;56:805-811.
4. Winograd G, Cohen P, Chen H. Adolescent borderline symptoms
in the community: prognosis for functioning over 20 years. J Child
Psychol Psychiatry. 2008;49:933-941.
5. Chanen AM, Jovev M, Jackson HJ. Adaptive functioning and
psychiatric symptoms in adolescents with borderline personality
disorder. J Clin Psychiatry. 2007;68:297-306.
6. Bernstein DP, Cohen P, Velez CN, Schwab-Stone M, Siever LJ,
Shinsato L. Prevalence and stability of the DSM-III-R personality
disorders in a community-based survey of adolescents. Am J
Psychiatry. 1993;150:1237-1243.
7. Lewinsohn PM, Rohde P, Seeley JR, Klein DN. Axis II psycho-
pathology as a function of axis I disorders in childhood and
adolescence. J Am Ac Child Adolesc Psychiatry. 1993;36:1752-1759.
8. Chabrol H, Montovany A, Duconge E, Kallmeyer A, Mullet E,
Leichsenring F. Factor structure of the borderline personality
inventory in adolescents. Eur J Psychol Assess. 2004;20:59-65.
9. Ryle A. The contribution of cognitive analytic therapy to the
treatment of borderline personality disorder. J Pers Disord. 2004;
18:3-35.
10. Binks C, Fenton M, McCarthy L, Lee T, Adams CE, Duggan C.
Psychological therapies for people with borderline personality
disorder (review). Cochrane Database Syst Rev. 2006;1. CD005652.
11. Katz LY, Cox BJ, Gunasekara S, Miller AL. Feasibility of dialectical
behavior therapy for suicidal adolescent inpatients. J Am Acad
Child Adolesc Psychiatry. 2004;43:276-282.
12. Rathus JH, Miller AL. Dialectical behavior therapy adapted for
suicidal adolescents. Suicide Life Threat Behav. 2002;32:146-157.
13. Bleiberg E, Rossouw T, Sharp C. Mentalization based treatment for
adolescents: a framework to treat adolescent breakdown and
emerging personality disorders. In: Fonagy P, Bateman AW, eds.
Handbook of Mentalizing in Mental Health Practice. Washington,
DC: American Psychiatric Publishing; 2011.
14. Bateman A, Fonagy P. Mentalization based treatment for border-
line personality disorder. World Psychiatry. 2010;9:11-15.
15. Van Gemert TG, Ringrose J, Schuppert HM, Wiersema HM.
Emotieregulatie Training.Een Programma voor Adolescenten
met Emotieregulatie Problemen. Amsterdam: Uitgeverij Boom;
2009.
16. Bartels N, Crotty M, Blum N. The Borderline Personality Disorder
Skill Training. Iowa City, Iowa: University of Iowa; 1997.
17. Linehan MM. Cognitive-Behavioral Treatment of Borderline Per-
sonality Disorder. New York: Guilford Press; 1993.
18. Putnam KE, Silk KR. Emotion dysregulation and the development
of borderline personality disorder. Dev Psychopathol. 2005;
17:899-925.
19. Steinberg L, Morris AM. Adolescent development. Annu Rev
Psychol.. 2001;52:83-110.
JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY
VOLUME 51 NUMBER 12 DECEMBER 2012
EMOTION REGULATION TRAINING
Correspondence to H. Marieke Schuppert, M.D.,University Center
of Child and Adolescent Psychiatry, P.O. Box 660, 9700 AR
Groningen, The Netherlands; e-mail: m.schuppert@accare.nl
0890-8567/$36.00/C 2012 American Academy of Child and
Adolescent Psychiatry
http://dx.doi.org/10.1016/j.jaac.2012.09.002
20. Schuppert HM, Giesen-Bloo J, van Gemert TG, et al. Effectiveness
of an emotion regulation group training for adolescents—a
randomized controlled pilot study. Clin Psychol Psychother.
2009;16:467-478.
21. Barnicot K, Katsakou C, Marougka S, Priebe S. Treatment comple-
tion in psychotherapy for borderline personality disorder—a
systematic review and meta-analysis. Acta Psychiatr Scand.
2011;123:327-338.
22. Weertman A, Arntz A, Kerkhofs MLM. Gestructureerd Klinisch
Interview voor DSM-IV Persoonlijkheidsstoornissen (SCID-II).
Lisse: Swets and Zeitlinger; 2000.
23. Kaufman J, Birmaher B, Brent D, et al. Schedule for Affective
Disorders and Schizophrenia for School-aged Children—Present
and Life-time Version (K-SADS-PL): initial reliability and validity
data. J Am Acad Child Adolesc Psychiatry. 1997;36:980-988.
24. Schuppert HM, Bloo J, Minderaa RB, Emmelkamp PMG, Nauta
MH. Psychometric evaluation of the Borderline Personality Dis-
order Severity Index-IV Adolescent and Parent Version. J Pers
Disord. 2012;26:628-640.
25. Arrindell WA, Barelds DPH, Janssen ICM, Buwalda FM, van der
Ende J. Invariance of SCL-90-R dimensions of symptom distress in
patients with peri partum pelvic pain (PPPP) syndrome. Br J Clin
Psychol. 2006;45:377-391.
26. Patrick DL, Edwards TC, Topolski TD. Adolescent quality-of-life,
part II: initial validation of a new instrument. J Adolesc. 2002;
25:287-300.
27. Miller AL, Rathus JH, Linehan MM. Dialectical Behaviour Therapy
with Suicidal Adolescents. New York: Guilford Press; 2007.
28. Wallston KA. The validity of the Multidimensional Health Locus
of Control scales. J Health Psychol. 2005;10:623-631.
29. Roelofs J, Braet C, Rood L, et al. Norms and screening utility of the
Dutch version of the Children’s Depression Inventory in clinical
and nonclinical youth. Psychol Assess. 2010;22:866-877.
30. Bussing R, Fernandez M, Harwood M, et al. Parent and teacher
SNAP-IV ratings of attention deficit hyperactivity disorder symp-
toms. Psychometric properties and normative ratings from a
school district sample. Assessment. 2008;15:317-328.
31. Rabash J, Charlton C, Browne WJ, Healy M, Cameron B. MLwiN
Version 2.24. Bristol: Centre for Multilevel Modelling, University
of Bristol; 2009.
32. Giesen-Bloo J, van Dyck R, Spinhoven P, et al. Outpatient
psychotherapy for borderline personality disorder: randomized
trial of schema-focused therapy vs. transference-focused psy-
chotherapy. Arch Gen Psychiatry. 2006;63:649-658.
33. Verheul R, van den Bosch LMC, Koeter MWJ, de Ridder MAJ,
Stijnen T, van den Brink W. Dialectical behavior therapy for
women with borderline personality disorder: 12-month rando-
mized clinical trial in the Netherlands. Br J Psychiatry. 2003;
182:135-140.
34. Bateman A, Fonagy P. Psychotherapy for Borderline Personality
Disorder. Oxford, UK: Oxford Medical Publications; 2004.
35. Davidson KM, Tyrer P, Norrie J, Palmer SJ, Tyrer H. Cognitive
therapy v. usual treatment for borderline personality disorder :
prospective six-year follow-up. Br J Psychiatry. 2010;197:456-462.
36. Schuppert HM, Albers CJ, Minderaa RB, Emmelkamp PMG,
Nauta MH. Parental rearing and psychopathology in mothers of
adolescents with and without borderline personality symptoms.
Child Adolesc Psychiatry Mental Health. 2012;6:29.
www.jaacap.org 1323
SCHUPPERT et al.

TABLE S1 Session-by-Session Outline of the Emotion Regulation Training (ERT) for Adolescents

Sessions Session Content

Phase 1: Psychoeducation and Cognitive Model

1 psychoeducation regarding emotion dysregulation
2 introduction to cognitive model and behavioral chain analysis
3 automatic thoughts, basic assumptions, distortions
4 learning to recognize an emotional storm; introduction of skills training
Systems meeting psychoeducation for important system members, explanation of
ERT principles
Phase 2: Knowing Yourself
5 knowing yourself
6 using your self-knowledge, informing others about your problems
7 breaking out of an emotional storm
8 integration and validation of previous sessions, recognizing and
preventing emotional storms
Phase 3: Lifestyle Changes
9 introduction of lifestyle and ‘‘behavior modification plans,’’ detection
of high-risk lifestyle areas
10 selection of subject for behavior modification plan
11 eating and sleeping
12 handling loneliness, and balance in school/work/recreation
13 personal and mental health hygiene (including self-injurious behavior)
14 money
15 interpersonal relationships I
16 interpersonal relationships II
Evaluation and Booster Sessions
17 evaluation
2 booster sessions review of ERT, review of ‘‘modification plans,’’ plans for the future

TABLE S2 Medication Use at Baseline and at Follow-Up

Baseline After Treatment
ERT þ TAU TAU ERT þ TAU TAU
(n ¼ 54) (n ¼ 55) (n ¼ 49) (n ¼ 50)
No medication 39 38 30 35
SSRI 5 2 4 2
ADHD 8 5 10 2
Antipsychotic 0 5 0 2
Other 0 2 1 0
SSRI þ ADHD 1 1 0 0
SSRI þ antipsychotic 0 1 1 2
SSRI þ other 0 0 2 0
ADHD þ antipsychotic 0 0 0 2
ADHD þ other 1 0 1 2
2 antipsychotics 0 0 0 1
2 SSRIs 0 0 0 1
3 types of medication 0 1 0 1

Note: ADHD ¼ attention-deficit/hyperactivity disorder; ERT ¼ Emotion Regulation Training; SSRI ¼ selective serotonin reuptake inhibitor; TAU ¼ treatment
as usual.

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

1323.e1 www.jaacap.org VOLUME 51 NUMBER 12 DECEMBER 2012
 EMOTION REGULATION TRAINING

TABLE S3 Parameter Estimates of Multilevel Models of Outcome Measurements for Completers

BPDSI

Affective MERLC-
Total Score Instability SCL-90-R YQOL-R LPI-ed intern
Fixed effects
Intercept (baseline 19.27 5.14 208.22 54.28 40.88 19.74
score) (1.06)*** (0.29)*** (6.54)*** (1.22)*** (1.33)*** (0.80)***
Post-treatment 10.06 0.98 20.73 1.74 (1.32) 5.12 0.64
(2.31)*** (0.32)** (7.48)** (1.48)*** (0.84)
Follow-up 8.45 1.94 52.71 16.38 13.12 0.96
(1.24)*** (0.37)*** (8.73)*** (4.32)*** (1.79)*** (1.00)
Post-treatment  0.76 (1.59) 0.60 12.66 0.68 (1.96) 4.50 1.13
condition (0.46) (11.15) (2.21)* (1.26)
CDI (baseline) 0.76 0.15 6.27 1.13 0.75
(0.08)*** (0.02)*** (0.58)*** (0.10)*** (0.10)***
Post-treatment  CDI 1.39 (0.65)* 0.35
(0.16)*
Post-treatment 
CDI  condition
Follow-up  CDI 3.77 0.51
(0.88)*** (0.17)**
SNAP (baseline) 1.04
(0.40)**
Post-treatment 
SNAP
Post-treatment 
SNAP  condition
Follow-up  SNAP 2.58 2.01 (0.94)* 1.29
(0.83)** (0.52)*
Abuse (baseline)
Post-treatment  3.63 21.40 (8.87)*
abuse (1.34)**
Post-treatment 
abuse  condition
Follow-up  abuse 30.08 5.35
(13.66)* (2.61)*
Random effects
Between-individual 32.64 2.07 920.66 37.10 41.89 17.81
variance (6.66)*** (0.48)*** (221.53)*** (8.50)*** (10.08)*** (3.83)***
Residual variance at 28.03 2.51 1302.76 43.64 55.18 17.70
measurement (3.52)*** (0.31)*** (160.49)*** (5.52)*** (7.04)*** (2.25)***
occasions

Note: Values are presented as estimate (standard error). Empty cells indicate data not included in the model (because of nonsignificance). Abuse ¼ history of
sexual and/or physical abuse; BPDSI ¼ Borderline Personality Disorder Severity Index-IV–Adolescent Version; CDI ¼ Children’s Depression Inventory;
condition ¼ Emotion Regulation Training plus treatment as usual or treatment as usual; LPI-ed ¼ Life Problem’s Inventory, Emotional Dysregulation subscale;
MERLC-intern ¼ Multidimensional Emotion Regulation Locus-of-Control, Intern subscale; SCL-90-R ¼ Symptoms Checklist-90–Revised; SNAP ¼ Swanson,
Nolan and Pelham Attention Deficit and Hyperactivity Disorder measure; YQOL-R ¼ Youth Quality-of-Life Research Version.
*p o .05, **p o .01, ***p o .001.

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 51 NUMBER 12 DECEMBER 2012 www.jaacap.org 1323.e2