Content uploaded by Jennifer S Cheavens
Author content
All content in this area was uploaded by Jennifer S Cheavens on Apr 24, 2016
Content may be subject to copyright.
Dialectical Behavior Therapy for Adolescents: Theory, Treatment
Adaptations, and Empirical Outcomes
Heather A. MacPherson
•
Jennifer S. Cheavens
•
Mary A. Fristad
Ó Springer Science+Business Media New York 2012
Abstract Dialectical behavior therapy (DBT) was originally
developed for chronically suicidal adults with borderline
personality disorder (BPD) and emotion dysregulation.
Randomized controlled trials (RCTs) indicate DBT is
associated with improvements in problem behaviors,
including suicide ideation and behavior, non-suicidal self-
injury (NSSI), attrition, and hospitalization. Positive out-
comes with adults have prompted researchers to adapt DBT
for adolescents. Given this interest in DBT for adolescents,
it is important to review the theoretical rationale and the
evidence base for this treatment and its adaptations. A solid
theoretical foundation allows for adequate evaluation of
content, structural, and developmental adaptations and
provides a framework for understanding which symptoms
or behaviors are expected to improve with treatment and
why. We first summarize the adult DBT literature,
including theory, treatment structure and content, and
outcome research. Then, we review theoretical underpin-
nings, adaptations, and outcomes of DBT for adolescents.
DBT has been adapted for adolescents with various psy-
chiatric disorders (i.e., BPD, mood disorders, externalizing
disorders, eating disorders, trichotillomania) and problem
behaviors (i.e., suicide ideation and behavior, NSSI) across
several settings (i.e., outpatient, day program, inpatient,
residential, correctional facility). The rationale for using
DBT with these adolescents rests in the common under-
lying dysfunction in emotion regulation among the
aforementioned disorders and problem behaviors. Thus, the
theoretical underpinnings of DBT suggest that this treat-
ment is likely to be beneficial for adolescents with a broad
array of emotion regulation difficulties, particularly und-
erregulation of emotion resulting in behavioral excess.
Results from open and quasi-experimental adolescent
studies are promising; however, RCTs are sorely needed.
Keywords Dialectical behavior therapy Adolescents
Emotion dysregulation Treatment adaptation
Introduction
Dialectical behavior therapy (DBT) is a cognitive behav-
ioral treatment originally developed by Linehan (1993a, b)
for the treatment of chronically suicidal individuals, often
with borderline personality disorder (BPD). Positive results
from randomized controlled trials (RCTs) with adults have
prompted researchers to adapt DBT for adolescents who
exhibit similar behavioral and emotional dysregulation.
Given this interest in DBT for adolescents, it is important
to review the theoretical rationale and the evidence base for
this treatment and its adaptations. A solid theoretical
foundation allows for adequate evaluation of content,
structural, and developmental adaptations and provides a
framework for understanding which symptoms or behav-
iors are expected to improve with treatment and why. We
first summarize the adult DBT literature, including theory,
treatment structure and content, and outcome research.
Then, we review theoretical underpinnings, adaptations,
and empirical outcomes of DBT for adolescents. Regarding
the outcome literature of DBT for adolescents, studies of
youth with BPD features, suicide ideation, suicide behav-
ior, and/or non-suicidal self-injury (NSSI) are reviewed in
H. A. MacPherson (&) M. A. Fristad
Department of Psychiatry, The Ohio State University, 1670
Upham Drive, Suite 460, Columbus, OH 43210-1250, USA
e-mail: heather.macpherson@osumc.edu
H. A. MacPherson J. S. Cheavens M. A. Fristad
Department of Psychology, The Ohio State University,
1835 Neil Avenue, Columbus, OH 43210-1250, USA
123
Clin Child Fam Psychol Rev
DOI 10.1007/s10567-012-0126-7
Table 1; studies of youth with oppositional defiant disorder
(ODD), bipolar disorder (BD), binge eating disorder
(BED), anorexia nervosa (AN), bulimia nervosa (BN), and
trichotillomania (TTM) are reviewed in Table 2; and
studies that investigated DBT for adolescents in diverse
settings (i.e., correctional facilities, residential treatment
centers, long-term inpatient units, day treatment programs)
are reviewed in Table 3. We conclude with a discussion of
limitations in the adolescent DBT literature and also offer
considerations for future research. Review of research
efforts suggests DBT may be beneficial for adolescents
with a broad array of emotion regulation difficulties, par-
ticularly underregulation of emotion resulting in behavioral
excess. However, RCTs are needed to provide more
definitive evidence for the efficacy of DBT for adolescents.
Dialectical Behavior Therapy for Adults
DBT incorporates aspects of behavioral science, dialectical
philosophy, and Zen practice. Through a balance of change
and acceptance techniques within in a dialectical frame-
work, DBT aims to extinguish maladaptive behaviors and
shape and reinforce adaptive behaviors within a validating
environment, with the goal of helping clients build a life
worth living. The following overview first discusses dia-
lectical and biosocial theoretical underpinnings of DBT
and treatment components, modes, and strategies, as
delineated in the individual therapy and skills training
manuals by Linehan (1993a, b). This section concludes
with a summary of the adult DBT outcome literature.
Theory
Both dialectical philosophy and the biosocial theory underlie
the DBT framework. Dialectical philosophy posits a
worldview emphasizing wholeness, interrelatedness, and
process. It also suggests that there is no absolute truth and
instead emphasizes the existence of opposing forces simul-
taneously (i.e., thesis and antithesis). Dialectical change or
progress comes from the resolution of opposing forces,
through the recognition of the truth or validity in each pole,
into a synthesis. Regarding therapeutic dialog and relation-
ship, dialectics refer to change by persuasion, making use of
oppositions inherent in the therapeutic relationship, and
continually questioning what is being left out of under-
standing, to reduce polarized thoughts and behaviors. In
DBT, dialectics inform a worldview and communication
strategies used to elicit change (e.g., by the therapist high-
lighting opposing viewpoints and simultaneously looking for
truth in each perspective). DBT assumes opposing views can
exist within a person at the same time (e.g., desire to live and
desire to die), which can result in conflict; however,
highlighting and accepting this tension can help both thera-
pist and client move past a treatment standstill and foster
change (Rizvi et al. 2012). The central dialectic in DBT is the
intrinsic tension between acceptance and change (Linehan
1997). For a review of dialectics in DBT, see Fruzzetti and
Fruzzetti (2008).
DBT is a theoretically derived treatment in which skills
and therapeutic techniques were developed to target spe-
cific deficits outlined in Linehan’s (1993a) biosocial theory
of BPD. The biosocial theory suggests that BPD is pri-
marily a dysfunction of emotion regulation (Linehan
1993a), or the ability to monitor, evaluate, and modulate
one’s affective state (i.e., when and what emotions occur,
and how one experiences and expresses those emotions) in
order to accomplish one’s goals (Gross 1998; Thompson
1994). Specifically, the biosocial theory posits that
the emotional, behavioral, interpersonal, cognitive, and
selfdysregulation of individuals with BPD are developed
and maintained through transaction between a biological
tendency toward emotion dysregulation and an invalidating
environment. An early biological vulnerability, expressed
in childhood as impulsivity, has also been identified as a
precursor to the development of BPD (Crowell et al. 2009).
Emotion dysregulation stems in part from emotional vul-
nerability, resulting in frequent and intense emotional
experiences, combined with an inability to adequately
regulate emotions. Characteristics of emotional vulnera-
bility include high sensitivity to emotional stimuli, emo-
tional intensity, and slow return to emotional baseline.
An invalidating environment negates, punishes, and/or
responds erratically and inappropriately to private experi-
ences, punishes emotional displays and intermittently
reinforces emotional escalation, and oversimplifies the ease
of problem solving. Invalidation has also been associated
with increased levels of negative affect and physiological
arousal (Shenk and Fruzzetti 2011). As a result, emotion-
ally vulnerable individuals who experience invalidating
environments have never learned how to label and regulate
emotions, how to tolerate distress, or when to trust their
emotional responses. They tend to invalidate their emo-
tional experiences, look to others for accurate reflections of
reality, and oversimplify the ease of problem solving. From
a biosocial perspective, BPD behaviors resulting from the
transaction between emotional vulnerability and an inval-
idating environment function to regulate emotions or are
consequences of failed emotion regulation. Recent empir-
ical research supports the central role of emotion dysreg-
ulation not only in BPD (Chapman et al. 2008; Glenn and
Klonsky 2009; Gratz et al. 2006; Hughes et al. 2012;
Putnam and Silk 2005; Reeves et al. 2010; Selby and Joiner
2009) but also across broad areas of psychopathology
(Aldao et al. 2010; Kring and Sloan 2010; Nolen-Hoek-
sema 2012).
Clin Child Fam Psychol Rev
123
Table 1 Studies of dialectical behavior therapy for adolescents with borderline personality disorder symptoms, suicide ideation, suicide
behavior, and/or non-suicidal self-injury
Authors Design/Setting N, % Female Age M or
Range
Inclusion Criteria % Completed DBT Format Outcomes
Miller et al.
(2000)
Pre–post
Outpatient
33
85 % F of 27
youth with
complete
data
14–19 Self-injury in past
16 weeks or
current suicide
ideation; 3 BPD
symptoms
N/A (only
examined
treatment
completers)
12 weeks: weekly
individual therapy and
multifamily skills group,
telephone coaching,
consultation team
Significant posttreatment
improvement in BPD
symptoms; all skills rated
moderately to extremely
helpful; distress tolerance
(self-soothe) and
mindfulness skills (do
what works, stay focused)
rated most helpful
Woodberry
and
Popenoe
(2008)
Pre–post
Community
clinic
46
89 % F
13–18 History of suicide
attempts, self-
injury, and/or
unstable affect
or relationships
in past
3–6 months
63 % 15 weeks: weekly
individual therapy and
multifamily skills group,
telephone coaching,
consultation team
Significant posttreatment
improvements in
depression (d = 0.76 to
0.84), anger (d = 0.94),
dissociation (d = 0.69),
overall psychiatric
symptoms and functional
difficulties (d = 0.63),
suicide ideation
(d = 0.73), thoughts of
NSSI (d = 0.62), parent
report of own depression
(d = 0.72); nonsignificant
improvements in
internalizing (d = 0.55),
externalizing (d = 0.60),
total problems (d = 0.65)
James et al.
(2008)
Pre–post with
follow-up
Community
clinic
16
100 % F
15–18 History of
[6 months of
severe and
persistent
deliberate self-
harm (all
had C 5 BPD
symptoms)
87.5 % 1 year (2 6-month blocks):
weekly individual
therapy and adolescent
skills group, telephone
coaching
Significant posttreatment
improvements in
depression, hopelessness,
NSSI, general
functioning; gains
maintained at 8-month
follow-up
James et al.
(2011)
Pre–post
Community
clinic
25
88 % F
13–17 History of
[6 months of
severe and
persistent
deliberate self-
harm (all
had C 5 BPD
symptoms)
72 % 1 year (2 6-month blocks):
weekly individual
therapy and adolescent
skills group, telephone
coaching, consultation
team, outreach strategies
(e.g., meals,
transportation, caregiver
consultation)
Significant posttreatment
improvements in
depression, hopelessness,
NSSI, general
functioning; findings
maintained with intent-to-
treat analyses
Fleischhaker
et al.
(2011)
Pre–post with
follow-up
Outpatient
12
100 % F
13–19 NSSI or suicide
behavior in past
16 weeks; BPD
diagnosis or C3
BPD symptoms
75 % 16–24 weeks: weekly
individual therapy and
multifamily skills group,
telephone coaching,
consultation team
Significant improvements at
1-year follow-up in
suicide behavior (8 youth
attempted pretreatment;
no attempts during study
or follow-up), NSSI
(d = 0.92), psychosocial
adjustment (d = 1.30 to
3.40 for significant
improvements),
psychopathology
(d = 0.54 to 2.14 for
significant
improvements), BPD
symptoms (pretreatment
M = 5.8, SD = 1.3;
follow-up M = 2.75,
SD = 1.9)
Clin Child Fam Psychol Rev
123
Using dialectical philosophy and biosocial theory,
Linehan (1993a) described common dialectical dilemmas
of individuals with BPD, characterized as six classes of
behaviors that represent the extremes of three continua. At
one end of each continuum is a class of behaviors
hypothesized to be biologically driven, and at the other end
is a class of behaviors thought to be socially determined
and maintained. The first dialectical dilemma, emotional
vulnerability versus self-invalidation, is a tendency to
vacillate between intense, uncontrollable emotional suf-
fering and dismissal, judgment, and invalidation of suf-
fering. Active passivity versus apparent competence, the
second dialectical dilemma, involves passivity in solving
one’s problems while actively engaging others to solve
problems, coupled with the tendency of others to overes-
timate the capabilities of the individual with BPD. Lastly,
unrelenting crises versus inhibited grieving is a tendency to
experience life as a series of extreme problems contrasted
with an inability to experience emotions associated with
trauma or loss. Individuals with BPD are likely to vacillate
between these polarities, causing distress (Linehan and
Schmidt 1995). In line with dialectical philosophy, the
overarching target of treatment is to help patients find the
truth in each end of the dialectic and create a synthesis that
reduces the distress associated with extreme vacillation.
Treatment Functions, Modes, and Strategies
Dialectical philosophy and biosocial theory inform DBT
functions, structure, and strategies. DBT has five functions:
enhancing behavioral capabilities; improving motivation;
assuring generalization of gains to the natural environment;
structuring the environment so it reinforces functional
rather than dysfunctional behaviors; and enhancing
Table 1 continued
Authors Design/Setting N, % Female Age M or
Range
Inclusion Criteria % Completed DBT Format Outcomes
Rathus and
Miller
(2002)
Quasi-
experimental
Outpatient
DBT = 29
TAU = 82
93 % F in
DBT
73 % F in
TAU
DBT = 16.1
TAU = 15.0
Suicide attempt
in last
16 weeks or
current suicide
ideation; BPD
diagnosis
or C 3 BPD
symptoms
DBT = 62 %
TAU = 40 %
12 weeks:
DBT = weekly individual
therapy and multifamily
skills group, telephone
coaching, consultation
team
TAU = weekly individual
psychodynamic
psychotherapy and
family therapy
Posttreatment, DBT
adolescents demonstrated
significantly fewer
psychiatric
hospitalizations (0 %
versus 13 %) and higher
treatment completion
compared with TAU; 1
suicide attempt in DBT
versus 7 in TAU; DBT
adolescents demonstrated
significant reductions in
suicide ideation,
depression, anxiety,
general psychiatric
symptoms, global
severity, BPD symptoms
posttreatment (not
measured in or compared
with TAU)
Katz et al.
(2004)
Quasi-
experimental
with follow-
up
Inpatient
DBT = 32
TAU = 30
84 % F in
total
sample
14–17 Recent suicide
attempt or
suicide
ideation;
agreement to
stay in hospital
for the duration
of treatment
N/A
(treatment
completion
required)
2 weeks:
DBT = 4 individual
therapy sessions, 10
adolescent skills group
sessions, consultation
team, DBT milieu
TAU =C1 per week
individual and daily
group psychodynamic
psychotherapy, case
management,
psychodynamic milieu
Posttreatment, DBT
adolescents demonstrated
significant reduction in
number of violent
incidents on unit
compared with TAU;
significant reduction in
total number of violent
incidents on unit
comparing 6-months
before and after DBT;
both groups demonstrated
significant reductions in
NSSI (DBT d
= 0.63;
TAU d = 0.73),
depression (DBT
d = 1.67; TAU
d = 1.05), suicide
ideation (DBT d = 2.12;
TAU d = 1.36) over
1-year follow-up
DBT dialectical behavior therapy, F female, BPD borderline personality disorder, N/A not applicable, NSSI non-suicidal self-injury, TAU treatment as usual
Clin Child Fam Psychol Rev
123
therapist capabilities and motivation. These functions are
apparent over the course of four stages of treatment, each
with a hierarchy of treatment targets, and four modes of
therapy. In the pretreatment stage, the therapist orients the
individual to treatment and obtains commitment to the
therapist–client relationship and to work on goals. In stage
one, the therapist helps the client attain basic capabilities
(i.e., adding to the skill repertoire) by reducing life-
threatening behaviors (e.g., suicide behavior, self-injury),
therapy-interfering behaviors (e.g., noncompliance, nonat-
tendance), and quality-of-life-interfering behaviors (e.g.,
homelessness, psychiatric disorders), and by increasing
Table 2 Studies of dialectical behavior therapy for adolescents with oppositional defiant disorder, bipolar disorder, eating disorders, and
trichotillomania
Authors Design/
Setting
N, % Female Age
or
Range
Inclusion Criteria %
Completed
DBT Format Outcomes
Nelson-
Gray
et al.
(2006)
Pre–post
Outpatient
54
15 % F of
32 youth
who
completed
treatment
10–15 Oppositional
defiant disorder
diagnosis
69 %
(5 youth
repeated)
16 weeks: weekly
adolescent skills group
with adaptations to
improve attendance and
homework completion
(e.g., pizza, financial
incentives)
Significant posttreatment
improvements in
positive behaviors (i.e.,
interpersonal strength),
oppositional defiant
disorder and
externalizing behaviors,
depressive symptoms,
internalizing symptoms,
total problem behaviors
Goldstein
et al.
(2007)
Pre–post
Specialty
outpatient
clinic
10
80 % F
14–18 Bipolar I, II, or
NOS diagnosis
with acute
manic, mixed, or
depressive
episode in last
3 months
90 % 1 year: 24 weekly then 12
monthly sessions
alternating individual
therapy with individual
family skills training,
telephone coaching, BD
adaptations (e.g., BD
psychoeducation)
High satisfaction and
significant posttreatment
improvements in suicide
ideation (d = 0.9 and
1.2), emotion
dysregulation (d = 0.3),
depression (d = 0.7);
nonsignificant
improvement in NSSI
(d = 0.8)
Safer
et al.
(2007)
Case study
with
follow-up
Specialty
outpatient
clinic
1 F 16 Binge eating
disorder
diagnosis
100 % 21 weeks: weekly
individual therapy with
skills, diary card, chain
analyses review, 4 family
sessions, telephone
coaching, ED adaptations
(e.g., DBT model of EDs)
Reduced frequency and
severity of binge
episodes posttreatment
and at 3-month follow-
up
Salbach-
Andrae
et al.
(2008)
Case series
Outpatient
AN = 6
BN = 6
100 % F
12–18 Anorexia nervosa
or bulimia
nervosa
diagnosis
92 % 25 weeks: weekly
individual therapy and
adolescent skills group (8
multifamily groups),
telephone coaching,
consultation team, ED
adaptations (e.g., review
nutrition, body image)
Significant posttreatment
improvements in
restricting (d = 1.2),
bingeing (d = 1.9),
purging (d = 1.7),
general psychopathology
(d = 0.43 to 1.10); AN
adolescents
demonstrated significant
improvement in body
mass index (d = -2.6)
Welch
and Kim
(2012)
Case study
with
follow-up
Outpatient
1 F 16 Trichotillomania
diagnosis
100 % 16 weeks: weekly
individual therapy with
skills and chain analyses
review, parent check-in
meetings, TTM
adaptations (e.g., TTM
psychoeducation, habit
reversal, stimulus control)
Improvements in hair
pulling, emotion
regulation, anxiety,
depression
posttreatment; slight
worsening of hair pulling
from posttreatment to
follow-up
DBT dialectical behavior therapy, F female, NOS not otherwise specified, BD bipolar disorder, NSSI non-suicidal self-injury, ED eating disorder,
AN anorexia nervosa, BN bulimia nervosa, TTM trichotillomania
Clin Child Fam Psychol Rev
123
Table 3 Studies of dialectical behavior therapy for adolescents in diverse settings
Authors Design/
Setting
N, % Female Age M or
Range
Inclusion Criteria % Completed DBT Format Outcomes
Trupin
et al.
(2002)
Pre–post
with
control
group
Juvenile
detention
facility
DBT = 45
TAU = 45
100 % F
Mental health
unit = 14.8
General
population
unit = 15.5
TAU = 15.2
Incarcerated
females on
mental health
unit (DBT
n = 22) or
general
population unit
(DBT n = 23;
TAU n = 45)
N/A 10 months:
DBT ? TAU = 1–2
times/week adolescent
skills group
TAU = educational,
recreational,
vocational programs,
group meetings,
behavior modification
Mental health unit
adolescents showed
significant reduction in
behavior problems
(aggression, NSSI,
classroom disruption);
staff on mental health
unit (who received
more DBT training; 80
versus 16 h) showed
significant reduction in
punitive responses
compared to year prior;
no behavior or staff
changes on other units
Shelton
et al.
(2011)
Pre–post
secondary
analyses
Correctional
facility
38
0%F
16–19 Incarcerated
males with
impulsive
behavior
problems
68 % 16 weeks: weekly
adolescent skills group
Significant posttreatment
improvements in
coping, aggression
impulsive behaviors;
nonsignificant
improvements in
negative affect, self-
control
Sunseri
(2004)
Pre–post
compared
29 months
before and
after DBT
Residential
treatment
facility
68 (n = 42
before DBT;
n = 26 after
DBT)
100 % F
12–18 Resident at
treatment
facility;
commitment to
DBT
N/A 29 months: weekly
individual therapy,
twice weekly
adolescent skills
group, telephone
coaching, consultation
team
After DBT
implementation,
significant reductions
in premature
terminations due to
self-harm or
psychiatric
hospitalization (16.7 %
versus 0 %), number of
days spent in
psychiatric hospitals
due to NSSI (71 days
from 8 youth versus
42 days from 6 youth),
duration of physical
restraints and
seclusions (median of
20 min versus 11 min)
Wasser
et al.
(2008)
Pre–post
with
control
group
(matched)
Residential
treatment
facility
DBT = 12
STM = 12
25 % F
DBT = 14.7
STM = 14.6
Resident at
treatment
facility
N/A (selected youth
who already
completed DBT)
17 weeks:
DBT = weekly
individual therapy and
multifamily skills
group
STM = family, group,
individual, behavioral,
medication treatment
General psychiatric
symptoms improved
posttreatment for both
groups; DBT
adolescents had
significantly greater
reduction in
depression; STM
adolescents had
significantly greater
reduction in
psychomotor
excitation
Clin Child Fam Psychol Rev
123
behavioral skills. In stage two, the therapist helps the client
replace quiet desperation with normative emotional expe-
riencing by decreasing posttraumatic stress. In stage three,
the therapist helps the client achieve ordinary happiness
and unhappiness and resolve problems in living by
increasing respect for self and achieving individual goals.
Finally, in stage four, the therapist helps the client resolve a
sense of incompleteness and attain the capacity for freedom
and sustained contentment. Most of the empirical research
on DBT has focused on stage one targets; however, flexi-
bility offered by the DBT stages allows for the application
of DBT to individuals with varying degrees of dysfunction
(Lynch et al. 2007b).
Aforementioned functions and stages of treatment are
accomplished via four modes of therapy: weekly individual
therapy; weekly group skills training; as-needed telephone
coaching; and weekly therapist consultation team meetings
(Robins and Rosenthal 2011). The individual therapist is
responsible for addressing motivational problems, treat-
ment planning, working on progress toward goals, and
assessing and problem-solving crises and skill deficits.
Other modes of treatment revolve around the individual
therapy (Linehan 1993a). Individual therapy is organized
around and sequentially targets the aforementioned hier-
archy of behaviors occurring either in session or reported
on the client’s weekly diary card, a monitoring tool on
which clients record daily ratings of emotions, problem
behaviors, and skills use (Rizvi et al. 2012). For example, a
therapist treating a client in stage one would first address
suicide or self-injurious behavior, followed by any forms of
Table 3 continued
Authors Design/
Setting
N, % Female Age M or
Range
Inclusion Criteria % Completed DBT Format Outcomes
McDonell
et al.
(2010)
Pre–post
with
historical
control
group
Long-term
inpatient
unit
DBT = 106
(from 2000 to
2005)
Control = 104
(from 1995 to
1999)
58 % F
12–17 Admitted to
inpatient unit
N/A 1 year: 3 DBT intensity
levels (unknown
frequency) = DBT
milieu (chain analyses,
behavior interventions,
individual skills); DBT
milieu ? adolescent
skills group; DBT
milieu ? adolescent
skills group ?
individual therapy; all
with consultation team
Control = individual
and family therapy as
needed
DBT adolescents
demonstrated
significant
improvement in global
functioning and
significant reduction in
number of
medications; compared
with control, DBT
adolescents
demonstrated
significant reduction in
NSSI
Charlton
and
Dykstra
(2011)
Pre–post
Day
treatment
program
19 Unknown Enrolled in day
treatment
program for
developmental
and behavioral
health needs
52 % moved to less
restrictive setting;
16 % remained in
day program; 19 %
moved to more
restrictive setting;
16 % lost to
follow-up
19 months: weekly
individual therapy,
twice weekly
multifamily skills
group (when family
available), telephone
coaching, consultation
team, adaptations for
intellectual disabilities
(e.g., concrete and
simplified language
and handouts)
Adolescents
demonstrated
increased DBT skills
use, ability to identify
maladaptive emotions,
thoughts, actions;
significant correlation
between problem
behaviors (e.g., argued,
tried to avoid work,
tried to hurt self or
others, attempted
suicide), negative
thoughts, negative
feelings with month
(i.e., as number of
months in program
increased number of
problem behaviors,
negative thoughts,
negative feelings
decreased)
DBT dialectical behavior therapy, TAU treatment as usual, F
female, N/A not applicable, NSSI non-suicidal self-injury, STM standard therapeutic milieu
Clin Child Fam Psychol Rev
123
noncompliance or behaviors interfering with treatment,
followed by Axis I disorders or other life problems, and
finally followed by skill building. Strategies for addressing
problem behaviors are described below.
Clients also participate in weekly group skills training.
Groups are conducted with a primary and coleader and range
from 2 to 2.5 h, with the first half devoted to homework
review and the second half spent teaching new skills (i.e.,
mindfulness, distress tolerance, emotion regulation, inter-
personal effectiveness). Mindfulness involves finding the
synthesis between extremes by orienting to the truth in each
position. These skills also include focusing attention by
observing, describing, and participating in the present
moment without trying to change one’s present experience
and while assuming a nonjudgmental stance, focusing
awareness on one thing at a time, and developing effective-
ness (i.e., doing what is needed to achieve one’s goals).
Mindfulness skills are central to DBT and thus are woven
throughout the other skills modules. Distress tolerance tea-
ches impulse control, distracting, and self-soothing strate-
gies for tolerating aversive contexts, surviving crises, and
radically accepting situations that cannot be changed without
resorting to dysfunctional behavior. Emotion regulation
teaches methods for identifying and describing emotions,
determining whether an emotion is justified by current cir-
cumstances, modulating emotions via acting opposite to the
emotion or problem solving, reducing vulnerability to
unwanted negative emotions, and increasing experience of
positive emotions. Finally, interpersonal effectiveness tea-
ches assertiveness skills aimed to help clients achieve their
objectives in interpersonal interactions while also main-
taining positive relationships and their self-respect. These
skills are taught over 6 months and then repeated. Following
the treatment hierarchy, group skills training targets: ther-
apy-destroying behaviors; skills acquisition, strengthening,
and generalization; and therapy-interfering behaviors.
Clients are encouraged to use as-needed telephone
coaching calls if they are experiencing suicide or self-
injurious urges, if they need help utilizing a skill or do not
know what skill to use, or if there is a rupture in the
therapeutic relationship. These calls are typically of short
duration (5–15 min) and consist of the therapist quickly
assessing the client’s problem and helping to identify the
most effective skill to use in the current situation. How-
ever, clients are prohibited from calling the therapist within
24 h of suicide or self-injurious behaviors in order to avoid
inadvertent reinforcement via therapist attention and
because the client has already used a strategy to relieve
distress (albeit maladaptive) instead of seeking assistance
from the therapist in identifying an adaptive skill. Clients
may call during this 24-h period to receive coaching for
medical attention and/or if the client is having urges to self-
harm again.
Lastly, weekly therapist consultation team meetings
(1–2 h) hold therapists within the therapeutic frame, bal-
ance therapists’ interactions with clients, address problems
that arise in treatment, increase adherence to DBT princi-
ples, and increase therapists’ motivation and capabilities in
delivering DBT. During consultation team, mindfulness is
first practiced and then an agenda is set according to the
aforementioned target hierarchy and therapists’ needs (i.e.,
help with individual clients or support when feeling burned
out). Together, these treatment modalities (i.e., individual
therapy, group skills training, telephone coaching, consul-
tation team meetings) aim to reduce clients’ dysfunctional
behaviors in the presence of dysregulated emotion.
Finally, specific treatment strategies are used within the
four modes of treatment to achieve the functions and tar-
gets outlined in DBT (Robins and Rosenthal 2011). Dia-
lectical strategies foster change by highlighting opposing
viewpoints and simultaneously looking for truth in each
perspective. A dialectical therapeutic relationship is con-
stantly balancing acceptance and change, flexibility and
stability, nurturing and challenging, and a focus on capa-
bilities and deficits, with the goal of achieving syntheses.
Dialectical strategies also target behavioral extremes and
rigidity and highlight contradictions in the client’s thoughts
or behavior by offering alternative viewpoints, encouraging
synthesis between opposing perspectives, and promoting
dialectical thinking and acting. Validation strategies
involve the therapist’s acceptance of the client and serve to
communicate to the client that his or her responses make
sense within the current context or are what would be
expected of almost anyone in a given situation (Linehan
1997). Stylistic strategies refer to style and form of thera-
pist interaction and include both reciprocal (e.g., responsive,
genuine) and irreverent (e.g., matter-of-fact, unexpected)
communication. Together, dialectical and stylistic strategies
produce the movement, speed, and flow characteristic of
therapist–client interactions in DBT.
Problem-solving strategies are the primary change
strategies in DBT and involve first understanding and
labeling a selected problem behavior (e.g., suicide behav-
iors, self-injury) via a behavioral chain analysis that iden-
tifies vulnerabilities, events, thoughts, feelings, sensations,
and behaviors that led up to the problem behavior, as well
as consequences of the behavior. Subsequently, a solution
analysis is conducted to identify points of intervention that
would disrupt the chain of events and prevent the problem
behavior from recurring, with emphasis on rehearsal and
troubleshooting. DBT has four sets of change procedures:
skills training, contingency management, exposure strate-
gies, and cognitive modification. Skills training teaches the
client new skills. Contingency management provides a
consequence that influences the probability of a client’s
behavior occurring again. Exposure provides non-
Clin Child Fam Psychol Rev
123
reinforced exposure to cues associated previously, but not
currently, with a threat. Cognitive modification changes the
client’s dysfunctional assumptions or beliefs. Finally, when
problems in the client’s environment interfere with func-
tioning or progress, the therapist employs case manage-
ment strategies by either consulting with the client on how
to interact effectively with the environment or intervening
directly when the environmental contingencies are very
powerful. Collectively, these treatment functions, modes,
and strategies aim to reduce problematic behaviors asso-
ciated with dysregulated emotions while shaping and
reinforcing more effective, adaptive behaviors. For a
review of DBT in clinical practice, see Dimeff and Koerner
(2007) and Rizvi et al. (2012).
Empirical Outcomes
Numerous randomized controlled trials (RCTs) with
adults have demonstrated DBT’s efficacy in treating BPD
and a range of other psychiatric disorders across various
settings. Reviewed below are RCTs of DBT for adults,
empirical findings from these studies, and proposed
mechanisms of change. To date, standard outpatient DBT
(including all four modes of therapy) for adults with BPD
has been evaluated in nine RCTs, three of which included
adults with BPD plus substance use disorders. Five RCTs
compared DBT with treatment as usual (TAU; Carter
et al. 2010; Koons et al. 2001; Linehan et al. 1991, 1999;
Verheul et al. 2003), while four RCTs compared DBT
with active treatments (Clarkin et al. 2007; Linehan et al.
2002, 2006; McMain et al. 2009). Active comparison
treatments included comprehensive validation with
12-step (Linehan et al. 2002), community treatment by
experts (primarily psychodynamic treatment; Linehan
et al. 2006), transference-focused therapy or supportive
treatment (Clarkin et al. 2007), and general psychiatric
management (psychodynamic treatment plus medication
management; McMain et al. 2009). A recent meta-anal-
ysis including eight RCTs and eight non-RCTs also
examined the efficacy of standard DBT for adults with
BPD (Kliem et al. 2010).
Two recent RCTs of DBT utilized broader inclusion
criteria than BPD diagnosis. One RCT evaluated DBT
versus TAU in an outpatient publicly funded service setting
for adults with any cluster B personality disorder (i.e.,
borderline, antisocial, narcissistic, histrionic; Feigenbaum
et al. 2012). The other RCT evaluated DBT versus opti-
mized TAU (supervision provided by non-cognitive
behavioral expert) in a college counseling center for stu-
dents who were suicidal, reported at least one lifetime
NSSI or suicide attempt, and endorsed three or more BPD
symptoms (Pistorello et al. 2012).
DBT for adults with BPD has also been evaluated
adjunctive to medication (Linehan et al. 2008; Simpson
et al. 2004; Soler et al. 2005). Though these studies were
RCTs, all participants received DBT and only the medi-
cation condition (active medication versus placebo) dif-
fered between groups.
Nine additional RCTs evaluated adapted DBT for adults
with depression, eating disorders (EDs), attention-deficit/
hyperactivity disorder (ADHD), and BD. Rationale for use
of DBT with these disorders rests in the common under-
lying dysfunction in emotion regulation (Kring and Sloan
2010). Two RCTs evaluated DBT plus antidepressant
medication versus antidepressant medication alone for
depressed older adults (Lynch et al. 2003) and depressed
older adults with at least one comorbid personality disorder
(Lynch et al. 2007a). DBT in these studies consisted of
group skills training and telephone coaching (Lynch et al.
2003) or individual therapy and group skills training
(Lynch et al. 2007a). One RCT for treatment-resistant
depression evaluated DBT group skills training versus
waitlist control (WLC; Harley et al. 2008). Two RCTs for
BED evaluated DBT group skills training versus WLC
(Telch et al. 2001) or an active group therapy comparison
(Safer et al. 2010). Two RCTs evaluated individual DBT
(with some skills training review) versus WLC for BN
(Safer et al. 2001) and binge eating and purging episodes
(Hill et al. 2011). One RCT for ADHD evaluated DBT
group skills training versus structured group discussion
control (Hirvikoski et al. 2011). Lastly, one RCT for BD
evaluated DBT group skills training versus WLC (Van Dijk
et al. 2012).
In addition to outpatient settings, where most of the
aforementioned RCTs were conducted, DBT has been
successfully implemented with adults in inpatient units
(e.g., Bohus et al. 2000, 2004; Kro
¨
ger et al. 2006, 2010),
community mental health centers (e.g., Comtois et al.
2007; Pasieczny and Connor 2011; Prendergast and
McCausland 2007), and forensic settings (e.g., Berzins and
Trestman 2004; Bradley and Follingstad 2003; Evershed
et al. 2003). However, these studies were not RCTs.
Results from RCTs cumulatively suggest that partici-
pation in DBT is associated with: reduced frequency and
severity of suicide behavior and/or NSSI (Carter et al.
2010; Clarkin et al. 2007; Feigenbaum et al. 2012; Koons
et al. 2001; Linehan et al. 1991, 1993, 1999, 2006, 2008;
McMain et al. 2009, 2012; Pistorello et al. 2012; van den
Bosch et al. 2002, 2005; Verheul et al. 2003) and suicide
ideation (Koons et al. 2001; Linehan et al. 2006); decreased
BPD symptoms (McMain et al. 2009,
2012; Pistorello et al.
2012), substance abuse/dependence (Harned et al. 2008;
Linehan et al. 1999, 2002; van den Bosch et al. 2005), ED
symptoms (Hill et al. 2011; Safer et al. 2001, 2010; Telch
et al. 2001), ADHD symptoms (Hirvikoski et al. 2011),
Clin Child Fam Psychol Rev
123
hopelessness (Koons et al. 2001), depression (Clarkin et al.
2007; Feigenbaum et al. 2012; Harley et al. 2008; Koons
et al. 2001; Linehan et al. 2006, 2008; Lynch et al. 2003,
2007a; McMain et al. 2009, 2012; Pistorello et al. 2012;
Simpson et al. 2004; Soler et al. 2005; Van Dijk et al.
2012), anger/irritability (Feigenbaum et al. 2012; Koons
et al. 2001; Linehan et al. 1993, 1994, 1999, 2008; McMain
et al. 2009, 2012), aggression (Linehan et al. 2008; Soler
et al. 2005), and affective control (Van Dijk et al. 2012);
reduced health service utilization and/or inpatient psychi-
atric days (Carter et al. 2010; Koons et al. 2001; Linehan
et al. 1991, 1993, 2006; McMain et al. 2009, 2012; Van
Dijk et al. 2012); and improved social and global adjust-
ment (Clarkin et al. 2007; Feigenbaum et al. 2012; Linehan
et al. 1993, 1994, 1999; Pistorello et al. 2012; Simpson
et al. 2004), treatment retention (Linehan et al. 1991, 1999,
2006; Safer et al.
2010; van den Bosch et al. 2002; Verheul
et al. 2003), quality of life (Carter et al. 2010; McMain
et al. 2009, 2012), and interpersonal functioning (McMain
et al. 2009, 2012).
While all RCTs demonstrated DBT improved emotional
and behavioral symptoms following treatment, some
studies conducted by researchers not affiliated with the
treatment developers (e.g., Carter et al. 2010; Feigenbaum
et al. 2012) and studies that compared DBT with active
treatments (especially treatments specifically designed for
individuals with BPD: Clarkin et al. 2007; McMain et al.
2009, 2012) did not always yield significant between-group
differences. Results from the meta-analysis of standard
DBT for adults with BPD by Kliem et al. (2010) also found
good treatment retention (27.3 % drop-out rate), a moder-
ate global effect size, and a moderate effect size for suicide
and self-injurious behaviors. However, this effect size
decreased to small when DBT was compared with BPD-
specific treatments, and a small reduction in effects was
shown at follow-ups. Thus, numerous studies of DBT for
BPD and other psychiatric disorders in various settings
have yielded positive results and suggest efficacy in
improving various emotional and behavioral symptoms in
adults, though not always to a significantly greater degree
than active treatments.
Though growing evidence supports the efficacy of DBT
for various adult psychiatric disorders, mechanisms of
change and necessary components linked with clinical
improvements are not well understood (Robins and Chap-
man 2004). As aforementioned, most RCTs have evaluated
the efficacy of standard DBT for BPD. However, a recent
RCT demonstrated efficacy of 3 months of DBT group
skills training alone versus psychodynamic-oriented group
skills training control among adults with BPD (Soler et al.
2009). Other RCTs have demonstrated efficacy of group
skills training alone among adults with depression (Harley
et al. 2008; Lynch et al. 2003), BED (Safer et al. 2010;
Telch et al. 2001), ADHD (Hirvikoski et al. 2011), and BD
(Van Dijk et al. 2012). In addition, a recent examination of
mediators in three RCTs of DBT for BPD revealed that
DBT skills fully mediated the decrease in suicide attempts
and depression and the increase in control of anger over
time (Neacsiu et al. 2010). DBT skills also partially med-
iated the decrease in NSSI over time. Efficacy of DBT
group skills training in aforementioned studies supports a
skills deficit model of these psychiatric disorders.
However, some RCTs that evaluated adapted DBT for
other psychiatric disorders found support for individual
therapy alone (with some skills training review) among
adults with BN (Safer et al. 2001) and binge eating and
purging episodes (Hill et al. 2011). In addition, a recent
non-RCT found similar positive outcomes among adults
with BPD who received 1 year of standard DBT versus
individual DBT (with incorporated skills training; Andio
´
n
et al. 2012). The role of the therapeutic relationship in DBT
has recently been examined using data from a previous
RCT (Linehan et al. 2006). Specifically, relative to com-
munity treatment by experts, DBT participants developed
significantly greater self-affirmation, self-love, self-pro-
tection, and less self-attack (Bedics et al. 2012a). In addi-
tion, DBT participants who perceived their therapist as
affirming and protecting reported less frequent NSSI.
Support has also been demonstrated for therapists’ bal-
ancing of autonomy and control, maintaining a non-
pejorative stance, and using warmth and autonomy (Bedics
et al. 2012b). These studies support the importance of
individual therapy components in DBT (e.g., behavior
therapy strategies, combination of acceptance and change
interventions, dialectical strategies, nonjudgmental
assumptions about patients) and the quality of the thera-
peutic relationship in ensuring positive clinical outcomes.
Thus, while additional research is needed to examine the
utility of specific treatment modes and strategies and their
role in the efficacy of DBT, results from RCTs support the
use standard DBT for adults with BPD, with growing
evidence for adaptations of DBT for other psychiatric
disorders.
Dialectical Behavior Therapy for Adolescents
Given positive outcomes with adults, recent research has
adapted and evaluated DBT for adolescents. The follow-
ing section reviews the theoretical underpinnings
informing use of DBT with adolescents, summarizes
treatment adaptations originally proposed by Miller et al.
(1997, 2007b), and concludes with a review of empirical
studies of DBT for adolescents. Limitations of current
studies and considerations for future research are also
discussed.
Clin Child Fam Psychol Rev
123
Theory
As reviewed above, DBT has been found to be efficacious
for adults with BPD (www.div12.org/Psychological
Treatments/treatments/bpd_dbt.html). Thus, adaptation of
DBT for adolescents with BPD symptoms or diagnosis may
be warranted and beneficial. Though most evaluations of
DBT for adolescents have included youth with BPD fea-
tures, other studies have targeted youth with various psy-
chiatric disorders (i.e., mood disorders, externalizing
disorders, EDs, TTM) and problem behaviors (i.e., suicide
ideation and behavior, NSSI) across several settings (i.e.,
outpatient, day program, inpatient, residential, correctional
facility). The rationale for using DBT with these adoles-
cents rests in the common underlying dysfunction in
emotion regulation among the aforementioned disorders
and problem behaviors.
Most adolescent DBT studies targeted youth with BPD
features (Fleischhaker et al. 2011; James et al. 2008, 2011;
Miller et al. 2000; Rathus and Miller 2002; Woodberry and
Popenoe 2008). Though somewhat controversial, research
suggests that the prevalence, reliability, and validity of
BPD diagnoses in adolescent samples are largely compa-
rable to those found among adult samples (Miller et al.
2008). Adolescents with BPD present with similar symp-
toms and functional impairment as adults with BPD
(Becker et al. 2002; Chanen et al. 2007). However,
research on the stability of BPD over time is mixed. While
for some severely affected adolescents the diagnosis of
BPD remains stable over time, a less severe subgroup of
youth moves in and out of diagnosis (Miller et al. 2008).
These findings are consistent with research suggesting that
BPD diagnostic status in adults is not particularly stable
(Zanarini et al. 2010). Symptoms related to temperament,
such as abandonment fears, have higher positive predictive
power when making the diagnosis of BPD in adolescents
(Becker et al. 2002) and also endure longer than other BPD
symptoms (e.g., those related to impulsivity) in adult
samples (Zanarini et al. 2007). Thus, research indicates that
the diagnosis of BPD in adolescents is comparable in terms
of symptom constellation, functional impairment, and
temporal stability to the diagnosis when made in adult
samples. Therefore, adaptation of DBT, an evidence-based
treatment for adults with BPD, for adolescents who exhibit
BPD features or diagnosis is a logical extension.
Although most empirical studies of DBT have included
adults with BPD, DBT was originally developed to treat
suicide-related behavior and extreme emotional and
behavioral dysregulation (Robins and Rosenthal 2011). As
such, within the DBT framework, BPD is conceptualized
primarily as a disorder of emotion regulation. Problematic
behaviors are viewed as efforts to regulate extreme emo-
tions or consequences of failed emotion regulation
(Linehan 1993a). Given that adolescents can also present
with similar dysregulated emotions and problematic
behaviors, and emotion dysregulation has been linked with
the development of various forms of psychopathology in
adolescents (McLaughlin et al. 2011), extension of DBT to
a broader group of adolescents (as opposed to just those
with BPD) may be warranted.
All of the behaviors and disorders that have been tar-
geted in studies of DBT for adolescents can be conceptu-
alized by poor emotion regulation. For example, all
evaluations of adolescents with BPD features (Fleischhaker
et al. 2011; James et al. 2008, 2011; Miller et al. 2000;
Rathus and Miller 2002; Woodberry and Popenoe 2008)
and one study of hospitalized adolescents (Katz et al. 2004)
also incorporated suicide ideation, suicide behavior, and/or
NSSI as study inclusion criteria. Indeed, suicide ideation
(Orbach et al.
2007), suicide behavior (Tama
´
s et al. 2007;
Zlotnick et al. 1997), and NSSI (Adrian et al. 2011; Nock
and Prinstein 2004; Nock et al. 2009) have been shown to
be related to emotion dysregulation in youth. For example,
the most common self-reported reasons for adolescent
NSSI are automatic positive reinforcement (i.e., to create a
desirable physiological state) and automatic negative
reinforcement (i.e., to escape from an averse physiological
state; Nock and Prinstein 2004; Nock et al. 2009). In
addition, the automatic negative reinforcement function of
NSSI has been associated with a history of suicide attempts
in adolescents (Nock and Prinstein 2005), thus supporting
an emotion regulation function of suicide ideation, suicide
behavior, and NSSI in adolescents.
Adaptations of DBT for youth with ODD (Nelson-Gray
et al. 2006), BD (Goldstein et al. 2007), BED (Safer et al.
2007), AN (Salbach-Andrae et al. 2008), BN (Salbach-
Andrae et al. 2008), and TTM (Welch and Kim 2012) can
also be tied to a common underlying dysfunction in emo-
tion regulation. For example, the diagnostic criteria for
ODD include emotion dysregulation (e.g., often loses
temper, spiteful and vindictive), interpersonal difficulties
(e.g., argues with adults, annoys others on purpose), and
poor distress tolerance (e.g., easily annoyed, angry and
resentful; Nelson-Gray et al. 2006). In addition, early
emotion dysregulation has been linked with the develop-
ment of ODD (Stingaris et al. 2010), while recent research
suggests that early ADHD and ODD symptoms predict
subsequent development of BPD symptoms (Burke and
Stepp 2012; Stepp et al. 2012). Similarly, research posits
that the core feature of adolescent BD is emotion dysreg-
ulation (Carlson and Meyer 2006; Dickstein and Leibenluft
2006; Leibenluft et al. 2003). In addition, BD in adoles-
cents is associated with suicide behavior (Goldstein et al.
2005), NSSI (Esposito-Smythers et al. 2010), interpersonal
deficits (Goldstein et al. 2006), and treatment noncompli-
ance (Coletti et al. 2005), all of which are DBT targets, and
Clin Child Fam Psychol Rev
123
DBT has been successfully implemented with adults with
BD in a recent RCT with promising results (Van Dijk et al.
2012). Thus, both ODD and BD in adolescents are asso-
ciated with dysfunction in emotion regulation as well as
other problem behaviors targeted in and responsive to
DBT.
Emotion dysregulation has also been linked to EDs and
TTM. Regarding EDs, an adapted biosocial theory posits
that EDs develop through transaction between an invali-
dating environment and a biological vulnerability to reg-
ulating emotions and/or to the hunger/satiety system
(Wisniewski and Kelly 2003; Wisniewski et al. 2007). ED
behaviors (bingeing, purging, restricting) are viewed as
behavioral attempts to avoid painful emotions, in the case
of AN, or change painful emotions, in the case of BED and
BN. Some empirical evidence also supports the role of
emotion dysregulation in ED symptoms in youth (Sim and
Zeman 2005). In addition, adolescents with EDs commonly
present with suicide ideation, suicide behavior, and NSSI,
which are targets in DBT (Bjarehed and Lundh 2008;
Peebles et al. 2011; Ruuska et al. 2005). Also, DBT has
been adapted for adults with EDs and demonstrated posi-
tive results in RCTs (Hill et al. 2011; Safer et al. 2001,
2010; Telch et al. 2001). Regarding TTM, research with
adults and youth indicates hair pulling is automatic/habit-
ual or functions to regulate emotions, with the latter cued
by negative emotions, intense thoughts or urges, or
attempts to create symmetry (Christenson and Mackenzie
1994; Diefenbach et al. 2008; Flessner et al. 2007, 2008,
2009; Shusterman et al. 2009). Also, DBT has been
adapted for adults with TTM and demonstrated promising
results in a case study (Keuthen and Spirch 2012) and open
trial (Keuthen et al. 2010, 2011). Thus, EDs and TTM in
adolescents are associated with emotion dysregulation and
problem behaviors targeted in DBT, and studies of DBT for
adults with EDs and TTM demonstrated positive findings.
Some researchers have investigated DBT for adoles-
cents in particular settings with a transdiagnostic focus
rather than targeting certain psychiatric disorders or
behavioral problems. Specifically, DBT has been imple-
mented with youth in correctional facilities (Shelton et al.
2011; Trupin et al. 2002), residential treatment facilities
(Sunseri 2004; Wasser et al. 2008), long-term inpatient
units (McDonell et al. 2010), and day treatment programs
(Charlton and Dykstra 2011). Again, rationale for using
DBT with these adolescents is based on the underlying
dysfunction in behavioral and emotional regulation. Youth
who participated in DBT in aforementioned studies pre-
sented with a number of psychiatric diagnoses (e.g., BPD,
substance abuse/dependence, EDs, mood disorders, post-
traumatic stress disorder, ADHD, ODD, conduct disorder)
and impairing behaviors (e.g., suicide ideation and
behavior, NSSI, aggression, impulsivity, disruptive
behavior, running away). Therefore, DBT in these settings
is applied transdiagnostically with the aim of reducing the
myriad symptoms related to behavioral and emotional
dysregulation and that have demonstrated improvement in
adult RCTs of DBT.
Thus, DBT has been adapted for adolescents with BPD,
suicide ideation and behavior, NSSI, ODD, BD, EDs, and
TTM. DBT has also been implemented in diverse settings
with youth who present with varied psychiatric and
behavioral impairment. Rationale for initiating DBT with
these adolescents rests in the common problems in emotion
regulation. Linehan (1993a) conceptualized BPD as a dis-
order of emotion regulation in the initial development of
the treatment, and as such, DBT is comprehensive and
flexible in a way that allows for use with clients presenting
with varied diagnoses, in diverse settings, across a rela-
tively larger age range.
Treatment Adaptations
Miller et al. (1997, 2007b) were the first to propose
adaptations of DBT for adolescents and subsequently
developed a treatment manual. Their adaptations targeted
youth exhibiting suicide ideation and behavior, NSSI, and
BPD features. Subsequent adaptations for other adolescent
presenting problems are modeled after and closely resem-
ble the Miller et al. (2007b) manual. DBT for adolescents
generally follows the same format as standard DBT,
including theoretical framework, functions, treatment tar-
gets, treatment modes, and strategies (Klein and Miller
2011). However, Miller et al. (2007b) introduced modifi-
cations to make DBT more developmentally appropriate
for adolescents and their families. The following summary
provides an overview of the adaptations to DBT for ado-
lescents, as delineated in the DBT manual for suicidal
adolescents by Miller et al. (2007b).
DBT for adolescents includes seven main adaptations of
standard DBT. First, family members, usually parents, are
included in multifamily skills training groups to enhance
generalization and reinforcement of skills and structure
adolescents’ environments (Miller et al. 2007a). In this
way, parents can serve as models and coaches for their
adolescents by utilizing and implementing skills. Parental
participation in skills training is designed to provide a
common vocabulary for therapeutic techniques within
families and enhance parents’ ability to provide validation,
support, and effective parenting. Including family members
in skills groups also offers the added benefits of providing
in vivo opportunities to role play skills, fostering inter-
family support, reducing adolescents’ disruptive behaviors
in group, and enhancing treatment compliance. Family
members may also receive telephone coaching and con-
sultation from the skills group therapist for skills
Clin Child Fam Psychol Rev
123
generalization, while adolescents receive telephone
coaching from the primary individual therapist (Steinberg
et al. 2011).
Second, family therapy sessions are conducted on an
as-needed basis. Although individual sessions with sig-
nificant others are incorporated into standard DBT for
adults, adapted DBT for adolescents focuses more
explicitly on this mode of treatment (Miller et al. 2002;
Woodberry et al. 2002). Family therapy sessions were
added because much of the turmoil in the lives of suicidal
adolescents involves their primary support system. Family
sessions are conducted when the relationship with a
family member is a central source of conflict or when a
crisis erupts within the family. The therapist may also
initiate family sessions if the treatment would be
enhanced by educating family members about particular
skills or aspects of treatment or if contingencies in the
home are too powerful for the adolescent to ignore and
continue to reinforce dysfunctional behavior. Goals of
family sessions include preparing the adolescent for
family interactions, increasing parental understanding of
adolescent’s emotional vulnerability, addressing parents’
own emotion dysregulation, improving familial commu-
nication, modifying contingencies in the familial envi-
ronment, and crisis management. Typically, selected
family members will attend 3 to 4 sessions out of the
adolescent’s 16 weeks of individual therapy, though more
or fewer sessions can be scheduled as needed.
A third adaptation involves the development and
teaching of three adolescent–family dialectical dilemmas
(Rathus and Miller 2000). Similar to the original dialectical
dilemmas proposed by Linehan (1993a), these adolescent–
family dialectical dilemmas are considered secondary
behavioral targets in DBT. The first dialectical dilemma,
excessive leniency versus authoritarian control, involves
placing too few behavioral demands or limits on the ado-
lescent, or being excessively permissive, versus enacting
coercive parenting methods limiting freedom, autonomy,
and independence. Normalizing pathological behaviors
versus pathologizing normative behaviors, the second
dialectical dilemma, involves viewing developmentally
normal adolescent behaviors as deviant versus failing to
address or perceive deviant adolescent behaviors as such.
Lastly, forcing autonomy versus fostering dependence
involves acting in ways that inhibit an adolescent’s
autonomy (e.g., excessive caretaking, overreliance on
parents) versus parents’ severing ties with the adolescent
such that he or she is prematurely forced to separate and
become self-sufficient. Adolescents and families tend to
vacillate between these polarities, causing extreme distress.
Thus, the central dilemma of treatment is to help adoles-
cents and parents move to a balanced position representing
synthesis.
Fourth, the treatment length was reduced from 1 year to
16 weeks. This may be the biggest change from standard
DBT because the time in treatment is significantly reduced
but the content (e.g., dialectical dilemmas, skills training
modules) is increased. According to Miller et al. (2007b),
treatment length was modified so it would be more
appealing to adolescents, given that suicidal adolescents
tend to complete only a limited number of therapy sessions.
For example, up to 77 % of adolescents who attempt sui-
cide do not attend therapy appointments or fail to complete
treatment (Trautman et al. 1993). Also, Miller et al.
(2007b) aimed to offer a brief treatment because they were
including many clients with first-time NSSI or suicide
attempts, many of whom did not meet full criteria for BPD.
Thus, they believed they could treat many of these ado-
lescents with a short-term treatment and offer optional
additional therapy (i.e., a graduate group or repeat of first
phase of treatment) for those who continued to exhibit
behavioral dyscontrol. Treatment length was also reduced
for pragmatic concerns so that clients who could not afford
extended therapy could still receive meaningful treatment,
which was in line with the current healthcare climate (e.g.,
acceptable to insurance companies).
A fifth adaptation, also involving the structure of DBT,
is a second phase of treatment: a 16 week optional graduate
group (with other treatment modes utilized as needed) for
clients who continue to exhibit difficulties following the
first phase of therapy (Miller et al. 2007a). Youth may
repeat the graduate group as many times as necessary in
order to achieve their identified goals. Both phases of
treatment address only the DBT stage one targets of
reducing life-threatening behaviors, reducing therapy-
interfering behaviors, reducing quality-of-life-interfering
behaviors, and increasing behavioral skills. The graduate
group is designed to address the DBT treatment functions
of improving capabilities, improving motivation, and pro-
moting generalization of skills, but in a way that requires
less intensive adolescent participation and fewer program
resources. The goal of the graduate group is to reinforce
and generalize skills previously taught. Group sessions
involve adolescents reviewing and teaching skills to peers
and consulting and problem solving with group members to
foster peer coaching and support rather than reliance on the
therapist. During this phase, the therapist consultation team
also continues, addressing the functions of treating the
therapist and structuring the environment as needed.
Continuing treatment in a separate, second phase with
reduced intensity allows for clients to feel an increased
sense of mastery without removing structural resources that
may be helping to maintain progress. Further, increasing
the length of treatment with a graduate group offers ado-
lescent clients the opportunity to use the skills that they
learned in the first stage of treatment to broaden treatment
Clin Child Fam Psychol Rev
123
goals once skills acquisition has occurred. Importantly, this
two-stage approach allows for reallocating staff resources
to ensure that therapists are available for more intensive
treatment of new clients who are beginning DBT.
Sixth, the number of skills taught within each module
was slightly reduced and a fifth adolescent-specific skills
module was added. Most of the original DBT skills were
maintained because there is no theoretical or empirical
basis for which skills to include or eliminate. In addition to
the four original DBT skills modules (i.e., mindfulness,
interpersonal effectiveness, emotion regulation, distress
tolerance), a fifth skills module, walking the middle path,
was developed for adolescents and their families. This
module teaches validation of self and others, behavioral
principles (i.e., how to reinforce, extinguish, punish, and
shape behavior), and three adolescent–family dialectical
dilemmas (described above) with the goal of finding the
middle path, or balanced synthesis, in each dilemma. The
dialectical dilemmas are introduced in the multifamily
skills training groups and are targeted in individual and
family therapy sessions.
Lastly, group skills handouts were modified to improve
their appeal and applicability to adolescents. Modifications
include simplification of terminology, streamlined language,
simplification of visual layout to decrease visual overstim-
ulation (via reduced amount of variability in font size, bold
print, underlining, and italicizing), and addition of adoles-
cent-geared graphics. Other important modifications when
teaching skills include adapting examples of each skill to
make them more applicable to adolescents and utilizing more
experiential and in vivo, rather than didactic, methods.
Thus, DBT for adolescents is based on the same theo-
retical underpinnings and generally follows the same
framework, including functions of treatment, targets,
modes, and strategies, as standard DBT for adults. How-
ever, adaptations involving inclusion of family members in
skills training, addition of family therapy sessions, devel-
opment of new adolescent–family dialectical dilemmas,
reduction of treatment length, addition of an optional
graduate group, implementation of a new skills module,
and modifications to handouts and delivery of content in
skills groups make DBT more applicable and appealing to
adolescents and their families.
Empirical Outcomes
To date, DBT for adolescents has been evaluated in 18
studies published in English-language journals. First, six
studies that targeted youth with BPD features plus suicide
ideation, suicide behavior, and/or NSSI (Fleischhaker et al.
2011; James et al. 2008, 2011; Miller et al. 2000; Rathus
and Miller 2002; Woodberry and Popenoe 2008) and one
study that targeted adolescents hospitalized for suicide
ideation or attempt (Katz et al. 2004) are reviewed. Then,
five studies that adapted DBT for other diagnoses associ-
ated with emotion dysregulation are summarized; specifi-
cally, one study each of youth with ODD (Nelson-Gray
et al. 2006), BD (Goldstein et al. 2007), BED (Safer et al.
2007), both AN and BN (Salbach-Andrae et al. 2008), and
TTM (Welch and Kim 2012). Lastly, six studies that
investigated DBT for adolescents in diverse settings rather
than with specific psychiatric or behavioral problems are
reviewed; including, correctional facilities (Shelton et al.
2011; Trupin et al. 2002), residential treatment centers
(Sunseri 2004; Wasser et al. 2008), long-term inpatient
units (McDonell et al. 2010), and day treatment programs
(Charlton and Dykstra 2011). See also Groves et al. (2012)
for a review of the adolescent DBT outcome literature
through 2008. The review concludes with a discussion of
limitations of current research and considerations for future
directions.
Five open trials of DBT for adolescents with BPD
symptoms plus suicide ideation, suicide behavior, and/or
NSSI demonstrated positive results (Fleischhaker et al. 2011;
James et al. 2008, 2011; Miller et al. 2000; Woodberry and
Popenoe 2008; see Table 1). These studies were conducted
predominantly with females in outpatient or community
clinic settings and most closely followed the DBT for ado-
lescents manual (including all four modes of standard DBT
plus family involvement; Miller et al. 2007b), aside from
variations in treatment length (ranging from 12 weeks
to 1 year). Results indicated improvements in suicide idea-
tion (Woodberry and Popenoe 2008), suicide behavior
(Fleischhaker et al. 2011), NSSI (Fleischhaker et al. 2011;
James et al. 2008, 2011), thoughts of NSSI (Woodberry and
Popenoe 2008), BPD symptoms (Fleischhaker et al. 2011;
Miller et al. 2000), depressive symptoms (James et al. 2008,
2011; Woodberry and Popenoe 2008), hopelessness (James
et al. 2008, 2011), dissociative symptoms (Woodberry and
Popenoe 2008), anger (Woodberry and Popenoe 2008),
overall psychiatric symptoms (Fleischhaker et al. 2011;
Woodberry and Popenoe 2008), general functioning (James
et al. 2008, 2011; Woodberry and Popenoe 2008), and psy-
chosocial adjustment (Fleischhaker et al. 2011). High com-
pletion rates were also reported (63–87.5 %), and in one
study, adolescents rated all skills moderately to extremely
helpful (Miller et al. 2000). Two studies demonstrated
maintenance of gains over 8-month (James et al. 2008)
and 1-year (Fleischhaker et al. 2011) follow-ups. Interest-
ingly, Woodberry and Popenoe (2008) also found signifi-
cant posttreatment improvement in parents’ depressive
symptoms.
Similarly, two quasi-experimental studies (i.e., lacking
random assignment) indicated improvement following
DBT when compared with TAU (psychodynamic psycho-
therapy) for mostly female adolescents with BPD features
Clin Child Fam Psychol Rev
123
plus suicide ideation or recent suicide attempt (Rathus and
Miller 2002) and adolescents hospitalized on an inpatient
unit for suicide ideation or attempt (Katz et al. 2004; see
Table 1). Rathus and Miller (2002) implemented DBT in
an outpatient setting and closely followed the Miller et al.
(2007b) manual; however, Katz et al. (2004) made adap-
tations to frequency of treatment modes to make DBT more
applicable on an inpatient unit (also, telephone coaching
was not used). Rathus and Miller (2002) found adolescents
who received 12 weeks of DBT demonstrated significantly
fewer psychiatric hospitalizations (0 versus 13 %) and
higher treatment completion (62 versus 40 %) compared
with TAU, despite youth in the DBT group having sig-
nificantly more psychopathology at baseline (i.e., depres-
sive and substance use disorders and BPD). There were no
significant between-group differences in suicide attempts,
likely due to low occurrence in both groups (7.3 %),
though only one DBT participant made an attempt during
the study versus seven in TAU. DBT participants also
demonstrated significant reductions in suicide ideation,
depression, anxiety, general psychiatric symptoms, global
severity, and BPD symptoms posttreatment; however, these
were not measured in the TAU group and thus could not be
compared. Katz et al. (2004) found adolescents who
received 2 weeks of DBT demonstrated a significant
reduction in the number of incidents on the inpatient unit
(e.g., violence toward self and others) when compared to
TAU at posttreatment. In addition, there was a significant
reduction in total number of incidents on the unit when
comparing the 6-month period before and after DBT
implementation. Both groups also demonstrated significant
reductions in NSSI, depression, and suicide ideation over
1-year follow-up.
Studies of DBT for ODD, BD, EDs, and TTM in out-
patient settings similarly demonstrated promising results;
however, these adaptations deviated significantly from the
Miller et al. (2007b) manual and lacked control compari-
sons (see Table 2). An open trial of 16 weeks of adoles-
cent-only group skills training with adaptations to improve
compliance (e.g., pizza, financial incentives) for youth with
ODD (mostly males) found significant posttreatment
improvements in positive behaviors (i.e., interpersonal
strength), ODD and externalizing behaviors, depressive
symptoms, internalizing symptoms, and total problem
behaviors (Nelson-Gray et al. 2006). An open trial of
1 year of DBT for youth (mostly females) with BD con-
sisting of acute treatment and continuation phase with BD
adaptations (e.g., psychoeducation about BD) and indi-
vidual therapy, individual family skills training, and tele-
phone coaching demonstrated high completion and
satisfaction and significant improvements in suicide idea-
tion, emotion dysregulation, and depression, and nonsig-
nificant improvement in NSSI (Goldstein et al. 2007).
One case study and one case series of DBT for youth
with EDs and one case study of DBT for an adolescent with
TTM also provide support for DBT with these populations
(see Table 2). DBT for adolescents with EDs incorporated
adaptations, such as reviewing the DBT model of disor-
dered eating behaviors and their association with dysreg-
ulated emotions, providing nutrition psychoeducation,
dispelling myths about food, and addressing negative body
issues. A case study of a 16-year-old female with BED who
received 21 weeks of individual therapy (with incorporated
skills review), 4 family therapy sessions, and telephone
coaching demonstrated reduced frequency and severity of
binge episodes posttreatment and at 3-month follow-up
(Safer et al. 2007). A case series of 25 weeks of DBT for
females with AN or BN consisting of weekly individual
therapy and adolescent group skills training (parents
attended 8 groups), telephone coaching, and consultation
team meetings found high treatment completion and sig-
nificant posttreatment improvements in behavioral symp-
toms of eating disorders (i.e., restricting, bingeing,
purging) and general psychopathology; AN youth also
demonstrated significant improvement in body mass index
(Salbach-Andrae et al. 2008). A case study of 16 weeks of
DBT for a 15-year-old female with TTM consisting of
weekly individual therapy with parent check-ins, psycho-
education about TTM, self-monitoring, chain analyses,
habit reversal, stimulus control, relapse prevention, and
DBT skills (mindfulness, emotion regulation, distress tol-
erance) found improvements in hair pulling, emotion reg-
ulation, anxiety, and depression by posttreatment, with
slight worsening of hair pulling at follow-up (Welch and
Kim 2012).
Further lending support to the use of DBT with ado-
lescents are six studies that adapted and examined DBT for
youth in specific settings (i.e., correctional facilities, resi-
dential treatment centers, long-term inpatient units, day
treatment programs) rather than with particular psychiatric
or behavioral targets, though many of these youth pre-
sented with numerous and severe psychiatric and behav-
ioral problems (see Table 3). These studies used open
designs, uncontrolled groups, or examination of time
periods before and after DBT implementation. DBT
adaptations also significantly deviated from the DBT for
adolescents manual (Miller et al. 2007b). Two studies in
correctional facilities implemented adolescent-only group
skills training in either a pre–post design with TAU control
and all females over 10 months (Trupin et al. 2002)oran
open design with males over 16 weeks (Shelton et al.
2011). Results indicated improvements in: behavior prob-
lems (e.g., aggression, NSSI, classroom disruption) and
punitive responses (Trupin et al. 2002); and coping,
aggression, impulsive behaviors, negative affect, and self-
control (Shelton et al. 2011). Two studies in residential
Clin Child Fam Psychol Rev
123
treatment facilities implemented either all four modes of
treatment over 29 months with females (compared
29 months before and after DBT implementation; Sunseri
2004) or individual therapy plus multifamily group skills
training over 17 weeks with mostly males and matched
standard therapeutic milieu control (STM; Wasser et al.
2008). Results demonstrated significant reductions in pre-
mature terminations due to self-harm or psychiatric hos-
pitalization (16.7 versus 0 %), number of days clients spent
in psychiatric hospitals due to NSSI (71 inpatient days
from 8 clients versus 42 inpatient days from 6 clients), and
duration of physical restraints and seclusions (median of
20 min versus 11 min) following implementation of DBT
(Sunseri 2004); and improvement in general psychiatric
symptoms, with DBT having a significantly greater impact
on depression and STM having a significantly greater
impact on psychomotor excitation (Wasser et al. 2008).
McDonell et al. (2010) compared youth receiving DBT
in a long-term inpatient unit to historical controls (who
received individual and family therapy as needed) over
1 year with three levels of DBT intensity (i.e., DBT milieu,
DBT milieu plus group skills training, or DBT milieu plus
group skills training and individual therapy) and found
significant improvement in global functioning and signifi-
cant reduction in number of medications, and significant
reduction in NSSI compared with control. Finally, exami-
nation of 19 months DBT adapted for youth with devel-
opmental and behavioral health needs in a day treatment
program (i.e., individual therapy, group skills training,
consultation team, telephone coaching, milieu behavior
management) found increased DBT skills use, ability to
identify maladaptive emotions, thoughts, and actions, and
significant correlation between problem behaviors (e.g.,
argued, tried to avoid work, tried to hurt self or others,
attempted suicide), negative thoughts, and negative feel-
ings with month (i.e., as number of months in DBT
increased number of problem behaviors, negative thoughts,
and negative feelings decreased; Charlton and Dykstra
2011). Collectively, findings from pre–post, uncontrolled,
and quasi-experimental studies examining DBT for ado-
lescents with a range of psychiatric disorders and problem
behaviors in various settings have yielded promising
results.
Limitations
Despite advances in research on DBT for adolescents,
significant limitations exist. First, although DBT was
originally adapted for adolescents with BPD features and
suicide ideation, suicide behavior, and/or NSSI, only five
open studies and one quasi-experimental trial have exam-
ined the efficacy of DBT for this population. One quasi-
experimental study also evaluated DBT for hospitalized
adolescents with suicide ideation or attempt. Open trials
lacked comparison groups; thus, it is possible that
improvements were due to nonspecific therapeutic factors,
uncontrolled medication use, passage of time, or other
factors unrelated to DBT. Quasi-experimental studies used
TAU control comparisons (psychodynamic psychotherapy)
but lacked random assignment. Thus, systematic differ-
ences between groups may have existed pretreatment and
affected outcome. Indeed, Rathus and Miller (2002) noted
youth who received DBT in their study presented with
significantly greater psychopathology than in the TAU
group.
Studies of DBT for youth with other psychiatric disor-
ders or in specific settings are promising but also have
limitations. First, trials of DBT for ODD and BD were
evaluated via open trials, while adaptations of DBT for
BED and TTM were evaluated in case studies, and DBT for
AN and BN was evaluated in a case series. Lack of com-
parison conditions in these trials limits the conclusions that
can be made about the efficacy of DBT for these disorders
(i.e., improvements may be due to factors unrelated to
DBT). Second, six studies that examined implementation
of DBT for adolescents in specific settings (i.e., correc-
tional facilities, residential treatment centers, long-term
inpatient units, day treatment programs) did not specify
diagnostic or behavioral inclusion criteria. Though these
youth presented with comorbid conditions and significant
impairment, this design creates a heterogeneous sample of
youth with a range of psychiatric and behavioral problems,
some of which may be more responsive to DBT than
others. Also, although four of these studies utilized com-
parison conditions (i.e., pre–post intervention records with
TAU comparison, matched samples across agencies, his-
torical controls, and time periods before and after DBT
implementation), groups were uncontrolled and random
assignment was not used; thus, systematic group differ-
ences may have affected outcome.
Other limitations common to most aforementioned
studies of DBT for adolescents included relatively small
sample sizes consisting mostly of females. Though
McDonell et al. (2010) included 210 youth in their examina-
tion of DBT in a long-term inpatient unit (n = 106) versus
historical controls (n = 104), among outpatient imple-
mentations of DBT with adolescents, which is the recom-
mended form of treatment delivery outlined in both adult
(Linehan 1993a, b) and adolescent (Miller et al. 2007b)
manuals, sample sizes ranged from 1 to 111 (though of
N =
111, only 29 received DBT and 82 received TAU).
Some studies went to great lengths to improve compliance
and retention (e.g., financial incentives, meals, outreach
strategies; James et al. 2011; Nelson-Gray et al. 2006;
Woodberry and Popenoe 2008), which limits the ecological
validity and generalizability of these findings. Also, most
Clin Child Fam Psychol Rev
123
measures assessed symptoms and functioning through
adolescent self-report. Treatment fidelity was not specifi-
cally measured in any study and treatment length ranged
from 2 weeks to 29 months, with some adaptations devi-
ating considerably from the format and structure of DBT
outlined in manuals (Linehan 1993a, b; Miller et al.
2007b). In addition, only five studies included follow-up
data, and during these periods, treatment was uncontrolled.
Lastly, most trials either did not report medication use, or
this was uncontrolled. As a result of these deviations in
terms of treatment format, structure, and content, the dif-
ferent adolescent psychiatric disorders and problem
behaviors to which DBT was applied, and various study
designs and lengths of follow-up assessments, it is difficult
to synthesize and draw overarching conclusions about the
research on DBT for adolescents.
Future Directions
Given limitations of current studies examining DBT for
adolescents, additional research is needed. Research on
DBT for adolescents is relatively limited (18 studies pub-
lished in English-language journals), and to date, there are
no published RCTs. Given that the RCT design is the gold
standard for determining treatment efficacy (Chambless
et al. 1996, 1998; Chambless and Hollon 1998), multiple
RCTs are needed to evaluate whether DBT can be con-
sidered efficacious for adolescents. Currently, RCTs
examining DBT for adolescents are underway, the results
of which will direct the future of adolescent DBT research
considerably (Groves et al. 2012). Stringent RCTs
employing control comparisons similar to those used in
adult efficacy studies (i.e., starting with WLC or TAU
comparisons, followed by nonbehavioral active treatment
controls) would provide more definitive evidence for the
efficacy of DBT for adolescents. Such RCTs should also be
conducted by diverse research groups, measure and dem-
onstrate adherence to the manual, consider allegiance
effects, include semistructured assessment of adolescent
psychiatric symptoms, and assess functioning at long-term
follow-ups.
Because current empirical evidence is strongest for
adults and adolescents with BPD features plus suicide
ideation, suicide behavior, and/or NSSI, RCTs should first
target youth with these symptoms and behaviors. If efficacy
is demonstrated, additional RCTs examining different
disorders in adolescents with an underlying emotion reg-
ulation dysfunction could be initiated. Similarly, studies in
diverse settings should aim to create more homogenous
samples of youth with similar presenting problems and
defined inclusion/exclusion criteria to test the efficacy of
DBT for a specific disorder or problem behavior. Also,
before mediator, moderator, dismantling, effectiveness, or
dissemination studies are conducted, RCTs are needed to
determine for which adolescent disorders or problem
behaviors DBT is effective.
Miller et al. (2007b) provided a theoretically sound and
developmentally appropriate adaptation of DBT for sui-
cidal adolescents. Future research should aim to evaluate
clinical components and outcomes of this adaptation. For
example, optimal length of treatment should be investi-
gated empirically. Current studies rely on adaptations of
DBT with various lengths, ranging from 2 weeks to
29 months. Though the original manual (Miller et al.
2007b) called for 16 weeks of outpatient treatment with
optional continuation, adolescents with different presenting
problems or in different settings may benefit from alternate
lengths of treatment. In contrast, 16 weeks may indeed be
the optimal length of DBT for adolescents. Empirical
evaluation would provide a more definitive answer to this
question. In addition, evaluation of the most pertinent and
effective DBT components and skills for adolescents and
their families should be considered. Most of the original
DBT treatment modes and skills were maintained in the
adolescent DBT manual (Miller et al. 2007b) because there
is no theoretical or empirical basis for which components
to include or eliminate. However, some adaptations of
DBT for adolescents only included some of the treatment
strategies, modes, and skills. Similarly, particular skills
may be more effective than others for adolescents and their
families. Determination of the most pertinent treatment
components and skills may indicate specific strategies,
modes, and modules to emphasize, which would be espe-
cially informative since treatment of adolescents is typi-
cally much shorter in duration than the original DBT
protocol.
Conclusion
Given positive outcomes among adults with various psy-
chiatric and behavioral impairments, DBT has been adap-
ted for use with adolescents who present with similar
problems. Current adaptations of DBT target youth with
BPD features, suicide ideation and behavior, NSSI, ODD,
BD, EDs, and TTM. DBT has also been applied transdi-
agnostically among youth with varied psychiatric and
behavioral problems in correctional facility, residential,
long-term inpatient, and day treatment settings. Rationale
for using DBT with these adolescents rests in the common
underlying dysfunction in emotion regulation across ages,
diagnoses, and problem behaviors. Treatment adaptations
and length vary depending on the presenting problem and
setting. However, most adaptations are modeled after the
adolescent DBT manual (Miller et al. 2007b) and involve
inclusion of family members in skills training, addition of
Clin Child Fam Psychol Rev
123
family therapy sessions, inclusion of new adolescent–
family dialectical dilemmas, reduction of length of treat-
ment, addition of optional graduate group, implementation
of a new skills module, and modifications to handouts and
delivery of content in skills groups.
Although DBT for adolescents has been examined in
several studies, the research is still in its infancy. Quasi-
experimental studies demonstrated that, when compared
with TAU, DBT for adolescents was associated with sig-
nificant reductions in inpatient hospitalizations, attrition,
and behavioral incidents (e.g., violence toward self and
others). These studies also found DBT was associated with
significant reductions in suicide ideation, NSSI, BPD
symptoms, depression, anxiety, general psychiatric symp-
toms, and global severity, but improvements in these areas
were either not compared with adolescents receiving TAU
(Rathus and Miller 2002) or significant in both DBT and
TAU groups (Katz et al. 2004). Additional findings from
trials using less rigorous methodology demonstrated that
DBT was associated with significant reductions in disso-
ciative symptoms, ED symptoms, TTM symptoms, anger,
externalizing behaviors, impulsivity, hopelessness, emo-
tion dysregulation, general psychopathology, and medica-
tion usage, and significant improvements in interpersonal
strength, coping, general functioning, and psychosocial
adjustment. Thus, DBT appears to be a promising inter-
vention for adolescents presenting with a broad array of
emotion regulation difficulties; however, RCTs are sorely
needed to provide more definitive evidence for the efficacy
of DBT for adolescents.
References
Adrian, M., Zeman, J., Erdley, C., Lisa, L., & Sim, L. (2011).
Emotional dysregulation and interpersonal difficulties as risk
factors for nonsuicidal self-injury in adolescent girls. Journal of
Abnormal Child Psychology, 39, 389–400.
Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-
regulation strategies across psychopathology: A meta-analytic
review. Clinical Psychology Review, 30, 217–237.
Andio
´
n, O
´
., Ferrer, M., Matali, J., Gancedo, B., Calvo, N., Barral, C.,
et al. (2012). Effectiveness of combined individual and group
dialectical behavior therapy compared to only individual
dialectical behavior therapy: A preliminary study. Psychother-
apy, 49, 241–250.
Becker, D. F., Grilo, C. M., Edell, W. S., & McGlashan, T. H. (2002).
Diagnostic efficiency of borderline personality disorder criteria
in hospitalized adolescents: Comparison with hospitalized
adults. The American Journal of Psychiatry, 159, 2042–2047.
Bedics, J. D., Atkins, D. C., Comtois, K. A., & Linehan, M. M.
(2012a). Treatment differences in the therapeutic relationship
and introject during a 2-year randomized controlled trial of
dialectical behavior therapy versus nonbehavioral psychotherapy
experts for borderline personality disorder. Journal of Consulting
and Clinical Psychology, 80, 66–77.
Bedics, J. D., Atkins, D. C., Comtois, K. A., & Linehan, M. M.
(2012b). Weekly therapist ratings of the therapeutic relationship
and patient introject during the course of dialectical behavioral
therapy for the treatment of borderline personality disorder.
Psychotherapy, 49, 231–240.
Berzins, L. G., & Trestman, R. L. (2004).