Dialectical Behavior Therapy For Borderline Personality Disorder

ANRV307-CP03-08 ARI 21 February 2007 15:49
Dialectical Behavior
Therapy for Borderline
Personality Disorder
Annu. Rev. Clin. Psychol. 2007.3:181-205. Downloaded from www.annualreviews.org
Thomas R. Lynch,1 William T. Trost,2

Nicholas Salsman,3 and Marsha M. Linehan4
by University of Washington on 01/03/12. For personal use only.
1
Departments of Psychology & Neurosciences and Psychiatry & Behavioral Sciences,
Duke University and Duke University Medical Center, Durham, North Carolina
27710; email: lynch011@mc.duke.edu
2
Psychiatric and Psychological Associates, Durham, North Carolina 27704;
email: williamttrost@gmail.com
3
Department of Psychology, University of Washington, Seattle, Washington 98195;
email: salsman@brtc.psych.washington.edu
4
Department of Psychology, University of Washington, Seattle, Washington 98195;
email: linehan@brtc.psych.washington.edu
Annu. Rev. Clin. Psychol. 2007. 3:181–205 Key Words

First published online as a Review in intentional self-injury, suicide, DBT, emotion regulation,
Advance on December 8, 2006
mindfulness
The Annual Review of Clinical Psychology is online
at http://clinpsy.annualreviews.org Abstract
This article’s doi: Since the introduction of Linehan’s treatment manuals in 1993,
10.1146/annurev.clinpsy.2.022305.095229
dialectical behavior therapy (DBT) has been widely disseminated
Copyright c 2007 by Annual Reviews. throughout multiple therapeutic settings and applied to a variety of
All rights reserved
diagnoses. The enthusiasm with which it was embraced by clini-
1548-5943/07/0427-0181$20.00 cians early on led some to question whether DBT’s popularity was
outstripping its empirical foundation. Most of the specific concerns
raised regarding DBT’s early empirical base have been meaningfully
addressed in subsequent randomized controlled trials. This review
provides a brief introduction to DBT, followed by a critical appraisal
of empirical support for the treatment and a discussion of current
research trends.
181
Contents
INTRODUCTION . . . . . . . . . . . . . . . . . 182 Linehan et al. 2006b . . . . . . . . . . . . . . 193
OVERVIEW OF DIALECTICAL RANDOMIZED CONTROLLED
BEHAVIOR THERAPY TRIALS OF DIALECTICAL
TREATMENT APPROACHES . . 183 BEHAVIOR THERAPY FOR
Functions and Modes of CLIENTS WITH OTHER
Dialectical Behavior DIAGNOSES . . . . . . . . . . . . . . . . . . . 195
Therapy . . . . . . . . . . . . . . . . . . . . . . 184 Dialectical Behavior Therapy for
Four Stages of Dialectical Behavior Depression and Other
Therapy . . . . . . . . . . . . . . . . . . . . . . 185 Personality Disorders . . . . . . . . . . 195
Therapy Adherence in Dialectical Dialectical Behavior Therapy for
Behavior Therapy . . . . . . . . . . . . . 186 Eating Disorders . . . . . . . . . . . . . . 196

RANDOMIZED CONTROLLED Quasi-Experimental Studies. . . . . . . 196
TRIALS OF DIALECTICAL GENERAL ISSUES . . . . . . . . . . . . . . . . 197
BEHAVIOR THERAPY FOR CURRENT DEVELOPMENTS

BORDERLINE PERSONALITY AND FUTURE DIRECTIONS . . 198
DISORDER . . . . . . . . . . . . . . . . . . . . . 187 Are Direct Comparisons with
Linehan et al. 1991, 1993, 1994 . . . 187 Other Borderline Personality
Koons et al. 2001 . . . . . . . . . . . . . . . . . 187 Disorder Treatments
Turner 2000a. . . . . . . . . . . . . . . . . . . . . 189 Needed? . . . . . . . . . . . . . . . . . . . . . . 198
Van den Bosch et al. 2002, 2005; Dismantling Studies and Testing
Verheul et al. 2003 . . . . . . . . . . . . 190 Mechanisms of Change . . . . . . . . 199
Linehan et al. 1999 . . . . . . . . . . . . . . . 192 CONCLUSION . . . . . . . . . . . . . . . . . . . . 200
Linehan et al. 2002 . . . . . . . . . . . . . . . 192
INTRODUCTION this review are to continue with this approach

by conservatively scrutinizing the status of
Following the publication of Linehan’s
DBT research and evaluating the rigor with
DBT: dialectical (1993a,b) treatment manuals and the first in-
behavior therapy which criticisms of prior research have been
tensive training for therapists outside of the
addressed to date.
BPD: borderline University of Washington in 1992, dialecti-
personality disorder Prior to any careful review of an existing
cal behavior therapy (DBT) for the treatment
treatment literature, it is useful to define the
Dialectic: the of borderline personality disorder (BPD) has
development of two necessary elements required for an interven-
grown increasingly popular among clinicians,
opposing positions tion to be classified as representative of the
patients, and mental health advocate groups.
(the thesis and treatment. These issues are particularly im-
However, excitement generated by new treat-
antithesis) that are portant considering the growing number of
resolved through ments should correspond with the empirical
adaptations developed based on DBT. For our
formation of a data supporting the efficacy of the new ap-
synthesis review, we focused primarily on randomized
proach. In an attempt to align enthusiasm
controlled trials (RCTs), and we used crite-
with empiricism, several reviewers compre-
ria for study inclusion based in part on a cur-
hensively critiqued the body of existent DBT
rently ongoing meta-analytic study of DBT
research in a special section of Clinical Psy-
(S. McMain, personal communication). Con-
chology: Science and Practice (Levendusky 2000,
sequently, we included studies if they met
Scheel 2000, Swenson 2000, Turner 2000b,
the following criteria: (a) At least one of the
Westen 2000, Widiger 2000). The goals of
182 Lynch et al.

treatment arms was DBT or described as with comorbid personality disorder (Lynch
cognitive behavior therapy and based on et al. 2006b), and eating-disordered individu-
treatment protocols specified in Linehan’s als (Safer et al. 2001, Telch et al. 2001).
Randomized
(1993a,b) books. (b) The DBT treatment must In this chapter, we first briefly review the controlled trial
have included individual therapy sessions, a principles and strategies associated with DBT (RCT): controls for
formal skills-training group, a therapist con- treatment and adherence measures used in factors that
sultation team, and some form of coach- treatment studies. Next, we critically review jeopardize internal
validity (history,
ing (typically telephone for outpatient care), findings from RCTs for BPD and other DBT
maturation of
and/or the function associated with each of adaptations that have been published. We participants, testing,
these modes was addressed in some fashion then mention published DBT studies that instrumentation,
(e.g., individual therapy conducted over the were not RCTs (e.g., quasi-experimental de- statistical regression,
telephone). (c) DBT treatment length must sign, open trial). Finally, we outline future di- selection, and
experimental
have been at least six months for outpatient rections for research.
mortality)
programs and at least two months for inpa-
Biosocial theory:
tient treatments. (d ) Outcome measures must
OVERVIEW OF DIALECTICAL the transaction
have included at least one scaled measure of
between an
self-injury. (e) The study specifically states BEHAVIOR THERAPY
invalidating rearing
that it is an RCT, or a review of methodology TREATMENT APPROACHES environment and a
reveals that the study meets RCT criteria (ran- DBT was originally developed as a treatment biological tendency
toward emotional
dom assignment of subjects to two or more for people who meet criteria for BPD, par-
vulnerability produce
treatment groups). We also report whether ticularly those who are highly suicidal. Since a dysregulation in
the study included measures of adherence. For then, DBT has been reformulated and con- the client’s emotional
RCT studies using DBT in non-BPD popu- ceptualized as a treatment for multidiagnostic system
lations, we used the same criteria with the ex- treatment-resistant populations. DBT draws
ception that a measure of self-harm was not its principles from behavioral science, dialec-
required. tical philosophy, and Zen practice. The ther-
Using this as our definitional criteria, we apy balances acceptance and change, with the
note that DBT has garnered considerably overall goal of helping patients not only to
greater empirical evidence for its efficacy in survive, but also to build a life worth living.
treating BPD since the critiques in 2000, war- In addition, DBT explicitly helps therapists
ranting designation as well-established when avoid becoming burned out, as often happens
utilizing criteria outlined by the Division in the treatment of behaviors associated with
12 Task Force (Chambless & Hollon 1998). BPD or multidiagnostic cases.
To be considered well-established accord- A guiding principle of DBT is summarized
ing to this criteria, a treatment must have in the biosocial theory elucidated by Linehan.
demonstrated efficacy in at least two rigor- Briefly, the biosocial theory of BPD asserts
ous RCTs with superiority over placebo con- that the client’s emotional and behavioral
trol conditions or another bona fide treatment dysregulation are elicited and reinforced by
(Chambless & Ollendick 2001). At this time, the transaction between an invalidating rear-
DBT has been evaluated and found to be ef- ing environment and a biological tendency
ficacious for the treatment of BPD in seven toward emotional vulnerability (Linehan
well-designed RCTs conducted across four in- 1993a). Practically speaking, this theory en-
dependent research teams (Koons et al. 2001; courages DBT therapists to view client be-
Linehan et al. 1991, 1993, 1994, 1999, 2002, haviors as natural reactions to environmental
2006b; Turner 2000a; Verheul et al. 2003). reinforcers. This theory also informs treat-
In addition, it has demonstrated efficacy in ment, which focuses on shaping and reinforc-
RCTs for chronically depressed older adults ing more adaptive behaviors while also pro-
(Lynch et al. 2003), older depressed adults viding clients with a validating environment.
www.annualreviews.org • Dialectical Behavior Therapy 183

Table 1 Interventions that serve the five functions of dialectical ing the client utilize skills that function to
behavior therapy reduce stress and the long-term negative con-
Function Example interventions sequences of repeated self-injury). The mid-
1. Enhance capabilities Behavioral skills training, modeling, dle path approach of dialectics is an inherent
behavioral rehearsal, psychoeducation, feature of Zen, and DBT utilizes these prin-
coaching and feedback, homework ciples in an effort to help clients behave more
2. Increase motivation Behavioral assessment, chain analysis, effectively and live more balanced lives.
contingency management,
exposure-based strategies, cognitive
modification Functions and Modes of Dialectical
3. Enhance generalization Phone and email consultation, homework, Behavior Therapy
to the natural environment in vivo interventions, client review of DBT is a comprehensive treatment designed
therapy tapes to serve five functions (see Table 1) through
4. Structure the Case management, family or marital interventions delivered in four modes of ther-
environment interventions apy. The first mode of therapy involves a tra-
5. Enhance therapist Weekly consultation team meeting, ditional dyadic relationship between the client
capabilities and motivation treatment manuals, supervision,

and his or her individual therapist. The indi-
to treat effectively continuing education
vidual DBT therapist takes primary respon-
sibility for a client’s treatment by overseeing
In DBT, therapists pay particularly close at- progress toward therapy goals, integration
tention to the factors that maintain dys- of therapy modes, and management of life-
functional behaviors, such as reinforcers of threatening behaviors and crises. Individ-
Intentional
self-injury: self-injurious behavior and aversive conse- ual DBT therapy is organized around the
nonfatal, intentional quences of more effective behavior. Whereas following target hierarchy: (a) eliminating
self-harm resulting behavioral principles focus on changing inef- life-threatening behaviors including suicide
in tissue damage, fective behavior, a great challenge in treating attempts and intentional self-injury, (b) elimi-
illness, or risk of
individuals with BPD is to balance the efforts nating therapy-interfering behavior including
death or ingestion of
drugs or other to change with acceptance and validation. In nonattendance or not doing homework, and
substances with clear general, a dialectical philosophy, which syn- (c) ameliorating behaviors and factors leading
intent to cause thesizes an initial proposition or thesis that is to decreased quality of life including home-
bodily harm or death opposed by a contradictory antithesis, helps to lessness, drug dependence, or other severe axis
provide this balance. For example, an organiz- I disorders.
ing assumption dialectically considers clients The second mode of therapy, skills train-
to radically be doing the best that they can ing, is a more didactic intervention that
while at the same time recognizing that they teaches clients four primary skill sets: mind-
need to do better and behave more effectively. fulness, distress tolerance, emotion regula-
In the case of BPD, one of the most fre- tion, and interpersonal effectiveness. Mind-
quent dialectical tensions is that a behavior, fulness primarily has to do with the quality
such as self-injury behavior, is both functional of awareness that an individual brings to the
(it helps the patient reduce distress in the short present experience. Mindfulness practice of-
term) and dysfunctional (the self-injury pro- ten involves letting go of attachments and be-
duces negative effects on health and inter- coming one with current experience, with-
personal functioning in the long term, and out judgment or any effort to change what
is associated with the risk of suicide). The is. At the same time, mindfulness involves
dialectical tension is resolved by finding the the use of skillful means and the finding of
synthesis, by seeking to find what is being left a middle path between extremes or polari-
out of the thesis and antithesis (e.g., validat- ties. Skills taught in this module include ob-
ing the need to relieve distress while help- serving, describing, fully participating, being
184 Lynch et al.

nonjudgmental, focusing on one thing in the the patient’s cognitive, behavioral, and emo-
present moment, and being effective over be- tional regulation systems when he or she ex-
ing right. Distress-tolerance training attempts periences intense emotions (Linehan 1993b).
to equip clients with a range of specific meth- Consequently, the primary goal of treatment
ods aimed at improving the client’s capac- is to help the patient to engage in functional,
ity to tolerate aversive situations, feelings, or life-enhancing behavior, even when intense
thoughts; to survive crises; and to radically ac- emotions are present. For example, mindful-
cept that which cannot be changed. Emotion- ness skills and opposite action (i.e., behav-
regulation training tends to be more change ing opposite to the action urges of an unjus-
focused and includes specific methods de- tified emotion) are hypothesized to work by
signed to identify what emotion is being ex- encouraging nonreinforced engagement with
perienced, to decide whether the emotion is emotionally evocative stimuli, while blocking
justified or fits the current circumstances, and dysfunctional escape, avoidance behaviors, or
then to learn ways to modulate the emotion other ineffective responses to intense emo-
if the client decides he or she would like to tions (Lynch et al. 2006a).
change his or her emotional experience. Fi-
nally, interpersonal effectiveness training is

designed to help clients interact with others Four Stages of Dialectical Behavior
in ways that allow them to improve relation- Therapy
ships while simultaneously maintaining their DBT is a flexible treatment that varies in its
own personal values and self-respect. approach depending on the client’s current
A third mode of therapy in DBT, skills level of disorder. This tailoring of approach
generalization, focuses on helping clients in- to the client’s current needs can be roughly
tegrate the skills and principles taught in DBT operationalized into four stages of treatment.
into real-life situations. In practice, this usu- A patient engaging in imminently dangerous
ally translates into telephone contact outside or deadly behaviors, such as suicidal behav-
of normal therapy hours (i.e., coaching calls). iors or severe heroin addiction, enters DBT
These calls are typically brief interactions fo- at the first stage of treatment. Treatment dur-
cused on helping clients apply specific skills in ing this stage is focused on eliminating the
specific circumstances. most severely disabling and dangerous behav-
The fourth mode of therapy employed iors. Once behavioral dysfunction is under
in DBT is a consultation team designed to control, patients move to stage two of treat-
support the therapists in working with diffi- ment, which focuses on shifting from quiet
cult clients. The teams serve several impor- desperation to emotional experiencing. Stage
tant functions, including reducing therapist two may include helping clients experience
burnout, providing therapy for the thera- emotions after a lifetime of avoiding emo-
pist, improving phenomenological empathy tions or inhibited grieving associated with
for clients, and providing consultation for in- posttraumatic stress disorder. Stage three ad-
dividual therapists or group skills trainers re- dresses problems in living, such as uncompli-
garding specific client difficulties. cated axis I disorders, career problems, and
The goal of the treatment approaches out- marital problems. Finally, stage four involves
lined above can be distilled down into the helping the client develop the capacity for
following process: the reduction of ineffec- freedom and joy. Treatment targets in stage
tive action tendencies linked with dysregu- four may include working on reducing feel-
lated emotions (Chapman & Linehan 2006). ings of emptiness or loneliness and increasing
The core problem in BPD is hypothesized experiences associated with feeling complete.
to not be excessively intense emotions, but Stage one targets are the focus of most of the
instead the pervasive habitual breakdown of empirical research available on DBT to date.

However, the ultimate goal of DBT is to pro- Anchor points for each item range between
vide a comprehensive treatment designed to 0 and 5. Each item is scored according to an
help clients at all levels of psychological dis- expert judgment-scoring algorithm based on
tress achieve optimal functioning. the DBT adherence strategy manual. Condi-
tions for scoring in the form of if-then rules
take into account the necessity and sufficiency
Therapy Adherence in Dialectical of each strategy given the context of the ses-
Behavior Therapy sion and the prescriptions/proscriptions of the
There is no agreed upon approach for con- DBT treatment manual. Scores of above 3.9
ducting manipulation checks in psychother- indicate an adherent session. Inter-rater reli-
apy outcome studies (Am. Psychiatr. Assoc. abilities of mean scores of the strategy items
2001). In addition, in a review of the literature, range from 0.78 to 0.83. Correlations between
only approximately 26% of recently com- the mean score of the items and the global
pleted psychotherapy outcome studies even rating range from 0.89 to 0.99 (Linehan &
used specific treatment protocols, less than Korslund 2003).
half these studies reported therapist training, Of the RCTs for BPD reviewed here, all
and only 13% documented therapist compe- but Turner (2000a) included measures of ad-
tence (Luborsky et al. 1997). Including rat- herence monitoring using the scale developed
ings of adherence in intervention research is by Linehan and Korslund. No studies report
important, however, because adherence pro- competence ratings as this type of rating scale
vides information about the purity and dose has yet to be developed. Obtaining ratings of
of a treatment that is received. Additionally, adherence can be an expensive proposition,
adherence ratings allow researchers to exam- as it requires a reliably trained therapist ex-
ine interventions that are specific to a par- pert in DBT to watch and code a session in
ticular treatment modality and those that are real time. Consequently, it is recommended
common to multiple treatment modalities. In that researchers budget and plan for how they
DBT-specific treatment, protocols have been will obtain reliable ratings of adherence dur-
developed and validated, and rating systems ing the early stages of study development.
have been established to examine adherence In an effort to maximize the number of sites
to treatment protocols (Linehan & Korslund that have adherence rating ability, the Line-
2003). In contrast to adherence, competence han team recently trained a small group of
ratings provide a qualitative assessment of DBT researchers from institutions outside
therapist skill in providing the prescribed ele- the University of Washington (UW) in the
ments of the treatment, and this type of rating DBT adherence scale. There are currently
is typically employed by experts using video- nine reliable coders: four at the UW, one in
taped sessions (Miller & Binder 2002). How- private practice, one in New York City, one
ever, measures of competence also should take in Canada, one in the Netherlands, and one
into account contextual issues, such as the in Spain (K. Korslund, personal communica-
stage in therapy, patient difficulty, and pre- tion). All coding is coordinated through the
senting problems (Waltz et al. 1993). UW regardless of the coder’s physical loca-
The DBT rating instrument generates a tion. Several other groups of coders are in
single item index of DBT adherence and sub- training presently (both in the United States
scale scores for the 12 DBT strategy domains. and abroad). However, more raters need to
The rating scale comprises 66 items reflec- be trained to facilitate DBT research. That
tive of the major DBT strategies, each oper- said, a significant strength of the existing ap-
ationalized with behaviorally defined anchor proach is that a reliable measure has been
points in the corresponding DBT adherence developed, and expert raters using the mea-
strategy manual (Linehan & Korslund 2003). sure are tested to reliability on an ongoing
186 Lynch et al.

basis. Given this measure’s strength, it is un- size and limited ability to control for nonspe-
likely that the field would benefit from the cific factors in the comparison treatment such
development of additional or alternative ad- as the intensity, stability, and affordability of
TAU: treatment as
herence measures. On the contrary, use therapy and the amount of training and su- usual
of multiple scales would make comparisons pervision received by therapists. In addition,
SI: suicidal ideation
across studies more difficult. Thus, at least for two subjects assigned to the DBT condition
now, it is probably best to consider the UW were not included in some of the final sta-
scale as the gold standard. tistical analyses because they dropped out af-
ter four or fewer sessions. Although this is
methodologically appropriate in many cases,
RANDOMIZED CONTROLLED particularly where statistical power is limited
TRIALS OF DIALECTICAL by small sample size, it is relatively less infor-
BEHAVIOR THERAPY FOR mative than the gold-standard intent-to-treat
BORDERLINE PERSONALITY (ITT) analysis, in which all subjects who are

DISORDER randomized to a treatment condition are in-
cluded in all analyses regardless of whether
Linehan et al. 1991, 1993, 1994 they actually received the treatment.
Linehan and colleagues at the UW under- Secondary analyses of the data in which
took the first major RCT of DBT, which re- many of these issues are statistically controlled
sulted in three published manuscripts exam- for have been reported and indicate that the
ining different aspects of the data (Linehan advantages found for DBT are maintained
et al. 1991, 1993, 1994). As this study has even when these factors are taken into account
been reviewed in detail elsewhere (Robins & (see Linehan et al. 1991, 1993, 1994). How-
Chapman 2004, Scheel 2000), we only dis- ever, although reassuring, such post-hoc anal-
cuss it briefly here (see Table 2 for a sum- yses do not provide a true substitute for a pri-
mary of RCTs). The study involved 44 sub- ori experimental control. Fortunately, since
jects with BPD and a history of recent and the completion of this original study, six ad-
repeated intentional self-injury and/or suicide ditional and more methodologically refined
attempts who were randomized to either DBT RCTs have been published that examine the
(N = 22) or treatment as usual (TAU) in the use of DBT for treatment of BPD or BPD co-
community (N = 22). The results indicated morbid with substance abuse and substantially
several statistically and clinically significant address these issues.
advantages for DBT over TAU. These in-
cluded substantially greater reductions in in-
tentional self-injury rate and associated med- Koons et al. 2001
ical risk, total psychiatric inpatient hospital An independent research team at Duke Uni-
days, treatment dropout, self-rated anger, and versity (Koons et al. 2001) compared outpa-
greater improvements in global and social tient DBT with TAU for borderline women
role functioning among DBT clients (Line- veterans being treated at the Durham Vet-
han et al. 1991, 1993, 1994). Both groups erans Affairs Medical Center. This study at-
improved similarly on measures of suicidal tempted to replicate Linehan’s original find-
ideation (SI) and depression. ings as well as examine DBTs efficacy with a
Certainly, this original study represented a less severely afflicted group of patients. The
seminal achievement. However, it was an early researchers hypothesized that lower symp-
efficacy study of an as-yet unproven treat- tomatic acuity would allow a shift in treat-
ment. As such, it does suffer from some of ment focus from imminently life-threatening
the methodological limitations that are typi- behaviors to treatment targets lower on the
cal of such studies. These include small sample therapeutic hierarchy, including depression,

Table 2 Summary of randomized controlled trials of dialectical behavior therapy (DBT)

Treatments (number of
patients) Inclusion criteria Length Main effects Reference(s)
Trials of DBT for BPD
DBT (N = 24) versus BPD + suicide attempt in 1 year ISI frequency and medical Linehan et al. 1991,
community mental past 8 weeks + one other risk, treatment retention, 1993, 1994
health TAU (N = 22) in past 5 years emergency/inpatient
Female treatment, anger, social and
global adjustment
DBT (N = 12) versus BPD + current drug 1 year Illicit drug use, social and Linehan et al. 1999
TAU (N = 16) dependence global adjustment,
Female treatment retention
DBT + LAAM (N = 11) Females with BPD + 1 year Opiate use Linehan et al. 2002
versus CVT + 12-step + current opiate

LAAM (N = 12) dependence
DBT-oriented (N = 12) BPD + referral from 1 year ISI/suicide attempts, Turner 2000a
versus CCT (N = 12)
emergency services for impulsiveness, anger,

suicide attempt depression, global
adjustment, inpatient
treatment
DBT (N = 10) versus BPD 6 months ISI/suicide attempts (trend), Koons et al. 2001
Veterans Administration Female hopelessness, suicidal
TAU (N = 10) ideation, depression, anger
expression
DBT (N = 31) versus BPD 1 year ISI/suicide attempts (trend), van den Bosch et al.
TAU (N = 33) Female treatment retention, 2002, 2005;
self-damaging impulsivity Verheul et al. 2003
DBT (N = 52) versus BPD + recent and 1 year Suicide attempts, treatment Linehan et al. 2006b
CTBE (N = 51) recurrent self-injury retention, emergency and
Female inpatient treatment
Trials of DBT modifications for non-BPD diagnoses
DBT + MED (N = 17) Age ≥60 28 weeks Remission at 6-month Lynch et al. 2003
versus MED alone Current major depression follow-up
(N = 17)
DBT + MED (N = 21) Age ≥55 + personality 24 weeks Interpersonal sensitivity, Lynch et al. 2006b
versus MED alone disorder interpersonal aggression,
(N = 14) Current depressive depression (trend)
symptoms
Nonresponse to MED
trial
Modified DBT skills Females age 18–65 20 weeks Binge episodes, binge days, Safer et al. 2001
training (N = 14) versus Binge/purge at least once eating in response to
wait list (N = 15) per week for 12 weeks aversive emotions (trend)
Modified DBT skills Females age 18–65 20 weeks Binge episodes; binge days; Telch et al. 2001
training (N = 22) versus Binge eating disorder anger; concerns about
wait list (N = 22) weight, body shape, and
eating
BPD, borderline personality disorder; CCT, client-centered therapy; CTBE, community treatment by psychotherapy experts in suicide and BPD;
CVT, comprehensive validation therapy; ISI, intentional self-injury; LAAM, levo-alpha-acetylmethadol; MED, antidepressant pharmacotherapy;
TAU, treatment as usual.
188 Lynch et al.

hopelessness, anxiety, anger, and dissocia- han’s original study (e.g., small sample size,
tion. Twenty-five patients were randomized no ITT analysis, multiple end points increas-
[13 DBT, 12 TAU with 20 completing the ing risk of type I error), it also incorporated
CCT:
study (10 in each group)]. All five dropouts several methodological improvements, most client-centered
cited transportation difficulties as the rea- notably in regard to the TAU comparison therapy
son for discontinuing treatment and were re- condition. Specifically, the TAU and DBT
moved from all analyses except for treatment conditions were essentially identical in terms
retention. Alterations were made to Linehan’s of treatment setting, treatment affordability
original inclusion criteria such that a history and availability, availability of expert ther-
of intentional self-injury was not required and apist supervision, and institutional prestige.
subjects with antisocial personality disorder Concerns regarding multiple end points with
were excluded. All subjects met DSM-III- such a small sample are valid but are perhaps
R criteria for BPD at study entry, and 40% partly allayed when we consider that DBT was
had a recent history of self-injurious behav- significantly superior to TAU on more than
ior. Reflecting the lower acuity of these sub- one-third of the outcomes assessed despite
jects, treatment duration was reduced from the small sample size and was numerically
12 months to 6 months, and weekly group superior to TAU on over two-thirds of the
skills-training sessions were reduced from outcomes. In addition, the majority of DBT
180 minutes to 90 minutes. Aside from these subjects met criteria for clinically meaningful
modifications, standard DBT was delivered improvement on six of the seven variables for
according to Linehan’s original (1993a,b) pro- which this could be defined a priori, whereas
tocol. The TAU comparison condition was this was true for TAU subjects on only two of
actually a somewhat enhanced version of usual the seven.
treatment at the Veterans Affairs Medical
Center and involved weekly 60-min individ-
ual psychotherapy sessions, referral to various Turner 2000a
psychosocial groups as appropriate, and reg- A second independent research team led by
ular medication-management visits. Turner compared a modified version of DBT
Despite the small sample size, results with client-centered therapy (CCT) in the
pointed strongly to the superiority of DBT treatment of BPD. The DBT treatment reg-
across a number of treatment outcomes. DBT imen included several substantial modifica-
was statistically superior to TAU on four out- tions to standard DBT. First, psychodynamic
comes (hopelessness, depression, anger ex- principles were introduced into the treatment
pression, and SI) as measured by group mul- to “conceptualize patients’ behavioral, emo-
tiplied by time interaction over the treatment tional, and cognitive relationship schema”
period. On four other variables, the DBT (Turner 2000a, p. 415). Second, to minimize
group showed either significant improvement between-group differences in total hours of
(dissociation and unexpressed anger) or a therapeutic contact, the DBT condition in-
strong trend toward improvement (hospital- cluded no formal skills-training group. In-
izations, intentional self-injury), whereas the stead, DBT skills were taught during indi-
TAU group did not. Both groups showed vidual therapy sessions. Despite the treatment
similar significant reductions in interviewer- modifications, we chose to include this study
rated depression scores and SCID-II bor- in our review as it met the definition of a DBT
derline symptoms, whereas interviewer-rated trial according to our criteria as outlined in the
anxiety symptoms were unimproved by either introduction. In addition, it represents an im-
intervention. portant contribution to the DBT treatment
Although this study suffered from some literature because it was the first RCT ex-
of the same limitations described for Line- amining DBT in a nonacademic setting and

the first to employ a structured, theory-driven control for effects of the DBT training course.
comparison treatment. Each therapist attended two weekly group su-
A local emergency room referred poten- pervision sessions: a DBT supervision group
tial subjects after they were seen for a suicide led by the author and a CCT supervision
attempt. All subjects met full DSM-III crite- group led by the senior clinic therapist. Group
ria for BPD according to standardized inter- supervision focused on improving treatment
view. Subjects with psychosis, bipolar disor- adherence by reviewing videotaped sessions.
der, mental retardation, or an organic mental No formal adherence ratings were reported.
disorder were excluded; however, unlike most Although subjects in both treatments
other DBT RCTs, both men and women were tended to improve, the results strongly fa-
included. Thirty-three of 62 potential subjects vored DBT over CCT. Significant between-
met study criteria and consented to partici- group differences were found for all three
pate. Of these, 7 dropped out prior to ran- primary outcomes, representing composite
domization and an additional 2 dropped out measures of suicidality, affective dysregula-

after learning treatment assignment, leaving a tion, and global mental health functioning.
modified ITT sample of 24 (12 in each group; Secondary analyses of individual measures re-
19 women and 5 men). The subjects tended vealed significant between-group differences
to be young (average age 22, with a range of favoring DBT on rates of suicide and inten-
18 to 27), and 83% carried a diagnosis of non- tional self-injury, depression, SI, hospitaliza-
nicotine substance abuse. tion days, a global score on a brief psychiatric
The active comparator, CCT, was based rating scale, impulsiveness, and anger.
on a treatment model developed by Carkuff. As noted above, the Turner study has sev-
CCT emphasizes “empathic understanding of eral characteristics that weaken its status as an
the patient’s sense of aloneness and provid- authentic DBT study. Perhaps most contro-
ing a supportive atmosphere for individua- versial in this respect is the incorporation of
tion” (Turner 2000a, p. 416). Therapeutic in- some psychodynamic principles into the DBT
terpretation and confrontation were generally condition. In addition, although the mention
proscribed in CCT. Whenever possible, CCT of weekly supervision to promote adherence is
clients were seen twice weekly (three times somewhat reassuring, there is no mention of
weekly in case of crisis). Frequency of individ- what sort of formal training the DBT supervi-
ual therapy for DBT was not specifically men- sor (Turner) had obtained, and no formal ad-
tioned. Both treatments lasted for 12 months, herence rating was used. Although these con-
and a total of six group sessions were of- cerns are valid and might justifiably exclude
fered to all participants over the course of the this study from a formal meta-analysis of DBT
year. The group format was focused loosely RCTs, they may also be viewed as evidence
around traditional DBT skills training, but that therapists who accept DBT assumptions
there was no mention of the specific skills and make a sincere effort to apply DBT can
addressed. achieve beneficial results.
The same four clinicians delivered both
treatments. The therapists had an average
of 22 years experience and reported back- Van den Bosch et al. 2002, 2005;
grounds in family systems, client-centered, Verheul et al. 2003
and psychodynamic treatments. Instruction in A third independent research team led by Ver-
DBT consisted of five lectures and 12 90- heul, van den Bosch, and colleagues compared
minute training sessions over three months. standard DBT with TAU for the treatment of
Although the therapists were generally famil- women with BPD who were attending com-
iar with CCT, each attended a weekly edu- munity psychiatry and substance-abuse clin-
cational seminar for 12 weeks in an effort to ics in Amsterdam (van den Bosch et al. 2002,
190 Lynch et al.

2005; Verheul et al. 2003). The study was the sion, and an additional two subjects in the
first large-scale RCT of standard DBT un- DBT group dropped out after attending only
dertaken in a nonacademic setting. Subjects one session. Analyses were performed on a
were mostly referred from psychiatric and ad- modified ITT sample excluding these six sub-
diction treatment centers, and each referring jects (27 DBT, 31 TAU). Despite high treat-
provider was required to sign a letter stating ment dropout from the TAU condition, 78%
that he or she was willing to treat the patient of all assessments were completed with no
for 12 months if the patient was assigned to difference between groups. Significant differ-
the control condition. Subjects referred by a ences in outcomes between groups were ev-
general practitioner or self-referred were re- ident. Subjects assigned to DBT had signif-
quired to obtain a similar letter from a psychi- icantly greater reductions in self-mutilating
atrist or psychologist prior to acceptance into and self-damaging impulsive behaviors and
the study. were significantly more likely to stay in treat-
The comparison condition was a true TAU ment than TAU subjects. Fewer DBT sub-
condition. The setting was similar between jects attempted suicide (2 out of 27 versus 8
groups, but the treatments differed in several out of 31), but this difference was not statis-
other ways. Treatment intensity was substan- tically significant (Fisher’s exact p = 0.0871).
tially higher in the DBT condition and con- Post-hoc analyses employing a severity factor
sisted of the full DBT program as described (defined by a median split on lifetime number
by Linehan et al. (1993). Subjects assigned of parasuicidal acts) found that DBT’s advan-
to TAU, by contrast, “attended generally no tage over TAU for treating suicidal and self-
more than two sessions per month” (Verheul mutilating behavior was most pronounced
et al. 2003, p. 136). Between-group differ- among severely afflicted subjects. DBT was
ences in clinician characteristics were not ex- not associated with a differential reduction
plicitly addressed but may have been impor- in prescription psychotropic medication use.
tant as well. DBT therapists (4 psychiatrists Unfortunately, the study did not present data
and 12 clinical psychologists) may have had regarding the use of crisis services, depres-
more overall education than TAU therapists sion/anxiety ratings, SI, or global function-
(a mix of psychiatrists, psychologists, and so- ing. A follow-up assessment six months after
cial workers) and were probably more enthu- treatment ended found that the superior gains
siastic about the treatment given that the DBT associated with DBT were maintained, al-
therapists were volunteers, whereas the TAU though DBT’s advantage was less pronounced
therapists were the same clinicians who had than it was immediately post-treatment.
originally referred subjects for potential treat- A secondary objective of this study was
ment elsewhere. DBT therapists also received to examine the efficacy of DBT among bor-
intensive training and supervision that were derline subjects with active substance abuse
not provided to TAU therapists. Although and dependence diagnoses. The study found
these differences are notable, they are also DBT to be equally effective for subjects with
typical for a comparison of a new treatment and without substance dependence in terms
with TAU and demonstrate that DBT can of reducing target behaviors (i.e., intentional
be applied in nonacademic community set- self-injury, self-mutilation). However, gener-
tings. In light of significant limitations, how- alization of improvements to nontarget be-
ever, it is probably best to interpret the results haviors (i.e., substance abuse) appeared to be
conservatively. limited. Analysis of 10 variables reflecting sub-
A total of 94 subjects were referred, of stance misuse found only one significant dif-
whom 64 were randomized to either DBT ference between groups (group multiplied by
(N = 31) or TAU (N = 33). Two subjects in time treatment effect for alcohol use) (van den
each group dropped out prior to the first ses- Bosch et al. 2002, 2005).

Recognizing the importance of treating DBT versus 8 out of 11 TAU). The primary
substance-abuse issues among borderline pa- drug-use outcome was based on structured in-
tients, Linehan’s group undertook the first terviews at baseline and at 4, 8, 12, and 16
CVT:
comprehensive significant modification of standard DBT in months. ITT analyses using one-tailed t-tests
validation therapy the mid-1990s. Specific modifications were indicated significant advantages for DBT on
based largely on early clinical experiences this measure for the treatment year overall as
with substance-abusing clients and included well as at the 4- and 16-month assessments.
(a) more aggressively targeting treatment Treatment-effect size estimates at all time
dropout by introducing a set of attachment points were in the moderate to large range
strategies and increasing the positive emo- per Cohen’s (1988) recommendations, possi-
tional valence of therapy, (b) encouraging pa- bly indicating that the study was underpow-
tients with opiate and stimulant addictions ered to detect a clinically significant differ-
to use replacement pharmacotherapy, and (c) ence between treatments. Urine drug screens
providing targeted case management to ad- were only performed on a maximum of six oc-
dress issues related to housing, finances, and casions (once at each assessment and once at
the legal system. Their work culminated in the random during the study). Urine-drug-screen
creation of a treatment manual for comorbid results tended to favor DBT at all time points
BPD and substance disorder in 1997 (Linehan according to ITT analysis, nearly reaching
& Dimeff 1997). statistical significance at 4 and 16 months. Es-
timated between-groups effect sizes for this
outcome were generally small to moderate.
Linehan et al. 1999 No between-groups differences emerged for
Results of the first randomized trial of the intentional self-injury, anger, global adjust-
modified treatment were published in 1999 ment, or social adjustment during treatment,
(Linehan et al. 1999). This trial compared but significant differences favoring DBT were
DBT with community TAU and involved 28 found for social adjustment and global ad-
women with comorbid BPD and substance- justment at 4 months post-treatment. Inter-
use disorder. Using a minimization proce- estingly, better adherence to DBT protocol
dure to match for age, severity of dependence, may have produced better results, as indicated
readiness to change, and Global Assessment by a post-hoc analysis revealing that adherent
Functioning score, the investigators assigned DBT therapist-client dyads had a higher pro-
12 subjects to DBT and 16 to TAU. They portion of negative urinalyses throughout the
ran analyses on both the ITT sample and the study.
treated sample, defined as those subjects in
either group who attended more than six ses-
sions and for whom outcome assessments be- Linehan et al. 2002
yond pretreatment were available (N = 18; 7 Building on the results of the 1999 study, a
DBT, 11 TAU). follow-up study compared DBT with a highly
Results from the ITT analyses indicated structured control condition for the treat-
an advantage for DBT in terms of treatment ment of comorbid BPD and opiate depen-
retention (7 out of 12 DBT versus 3 out of dence (Linehan et al. 2002). Women who met
16 TAU; Fisher’s exact p = 0.0497). Included DSM-IV criteria for both disorders were ran-
among DBT dropouts was one subject who domly assigned to DBT or a combination of
died during the study, apparently as the re- comprehensive validation therapy and a struc-
sult of an accidental overdose. Among subjects tured 12-step program (CVT + 12-step). The
who attended at least one session (11 DBT, CVT + 12-step control condition was de-
11 TAU), a nonsignificant trend favoring bet- signed to more thoroughly control for the in-
ter retention in DBT was seen (4 out of 11 fluence of nonspecific treatment variables on
192 Lynch et al.

treatment outcome. Essentially, CVT repre- The results of this study reinforce the
sents only the acceptance side of the accep- importance of controlling for nonspecific
tance/change dialectic that underlies DBT. As treatment factors in therapy trials, and they
CTBE: community
such, CVT therapists employ all of the valida- perhaps also speak to concerns regarding the treatment by experts
tion techniques used in DBT, but they do not empirical rigor with which DBT has been in suicide and
use cognitive-behavioral change techniques, evaluated. The control treatment here was not borderline
give overt advice, or actively direct the ther- a so-called paper tiger designed to empha- personality disorder
apy session beyond insisting that drug use be size DBT’s effectiveness while providing some
brought up at least once. To control for group- semblance of experimental control. Instead,
therapy hours, the study required that CVT CVT + 12-step was both efficacious and skill-
subjects attend a weekly Narcotics Anony- fully executed, as evidenced by the remarkable
mous “12 and 12” meeting conducted by the 100% retention rate of clients who have his-
two CVT therapists who were also recovering torically been difficult to keep in treatment.
addicts. As part of the 12-step program, sub- The increasing sophistication of the control
jects in this condition were also encouraged conditions used in DBT trials also reflects a
to meet with an NA sponsor weekly. guiding principle that has been embraced by
The DBT protocol was essentially identi- Linehan and others examining DBT, which
cal to the one used in the 1999 study, except holds that subject safety and well-being must
that an additional 30-minute individual skills- be of paramount concern.
training session was offered to DBT clients to
control for the weekly 12-step sponsor meet-
ings in the control group. Women in both Linehan et al. 2006b
groups received opiate replacement therapy In the largest and most rigorously controlled
throughout the study. Twenty-four women RCT of DBT to date, Linehan et al. (2006b)
were randomized (12 to each arm). One sub- compared standard DBT with community
ject was subsequently dropped from the DBT treatment by experts (CTBE). This study was
condition after it was discovered that she did designed to replicate the results of the origi-
not meet inclusion criteria, leaving a modi- nal study while controlling for a wide range
fied ITT sample of 23 for analyses. Significant of potential confounds not specifically ad-
decreases in opiate use were evident in both dressed in that study. Subjects in the two
treatment arms, and primary outcomes analy- groups were matched according to total num-
ses found no significant differences between ber of lifetime suicide attempts and nonsui-
groups on the main measures of drug use cidal self-injuries combined, number of psy-
and parasuicidal behavior. Secondary analy- chiatric hospitalizations, history of bona fide
ses revealed a significantly lower proportion suicide attempts versus nonsuicidal self-injury
of opiate-positive drug screens among DBT only, age, and presence of negative prognos-
clients over the course of the treatment, pri- tic factors (severe depression and severely
marily resulting from some rebound in drug impaired interviewer-assessed global func-
use among CVT clients over the last four tioning). The comparator condition (CTBE)
months of treatment. An additional finding was carefully designed to control for a vari-
of interest was that self-report of drug use was ety of nonspecific treatment effects, includ-
significantly more accurate in the DBT group ing treatment availability; ease of obtaining
as corroborated by thrice weekly urine drug and traveling to the first appointment; hours
screens throughout treatment. Remarkably, of individual psychotherapy offered; insti-
not a single subject dropped out of the CVT tutional prestige associated with treatment;
group over the entire 12 months of treatment. and therapist factors including gender, alle-
This was attributed to the supportive and val- giance to treatment offered, formal education
idating environment fostered by CVT. (i.e., doctoral versus master’s degree), clinical

experience, and availability of supervision and posure, and 5 were either graduate students
group clinical consultation. or postdoctorates. DBT training consisted of
Subjects were women between the ages of 45 hours of training followed by supervised
18 and 45 who met DSM-IV diagnostic cri- practice, and therapists were hired after being
teria for BPD and who had attempted suicide rated to adherence on six out of eight consec-
and/or self-injured at least once in the past utive training case sessions. The two groups
eight weeks and twice in the past five years. of therapists were matched according to ed-
Potential subjects were excluded if they had ucation (i.e., doctoral versus master’s degree)
a psychotic disorder, bipolar disorder, mental and gender. However, therapists in the CTBE
retardation, a seizure disorder requiring med- group had significantly more experience than
ication, or if treatment was mandated. One- did DBT therapists on average.
hundred and eleven subjects were randomized Although subjects in both conditions
to either DBT (60) or CTBE (51). Eight DBT showed substantial improvements, the DBT
training cases and two CTBE pilot subjects group generally exhibited better treatment re-
were not included in the analyses, leaving a sponse, particularly on outcomes related to
final ITT sample of 101 (52 DBT, 49 CTBE). behaviors specifically targeted by treatment.
CTBE therapists were nominated by local Subjects assigned to DBT were half as likely
mental health leaders based on reputation for to attempt suicide as those assigned to CTBE
expertise with especially difficult and chron- (23.1% with at least one suicide attempt in
ically suicidal clients. Of 94 therapists nomi- DBT versus 46% in CTBE; p = 0.01). A sim-
nated, 38 were selected for the study and 25 ilar, although not statistically significant, ad-
accepted at least one study client. To avoid vantage was seen when considering only non-
cross-contamination of treatment techniques, ambivalent suicide attempts (5.8% in DBT
only therapists who described their treatment versus 13.3% in CTBE; p = 0.18). Among
approach as nonbehavioral or mostly psycho- subjects who did engage in self-injurious or
dynamic were selected for the CTBE condi- suicidal behaviors, ratings of medical risk as-
tion. To optimize therapist allegiance to the sociated with these behaviors were signifi-
delivered treatment, CTBE therapists were cantly lower in the DBT group. Although no
instructed to provide the dose and type of significant difference was found for nonsuici-
therapy that they felt was most appropriate dal self-injury between groups, a greater re-
for the client, with the single requirement that duction was documented for the DBT group
individual therapy be offered at least once per as indicated by an estimated between-group
week. To control for both the effects of the treatment effect size of 0.49 (“moderate” ef-
DBT therapist consultation team and client fect per Cohen 1988). Subjects receiving DBT
expectations linked to institutional prestige, also used significantly fewer crisis services
all CTBE therapists were encouraged to at- (e.g., psychiatric emergency room visits and
tend a weekly group supervision session led by inpatient admissions) than subjects assigned
the training director of the Seattle Psychoan- to CTBE. Although a significant difference
alytic Society. To ensure optimal treatment af- was seen for all psychiatric emergency room
fordability and availability in both conditions, visits and admissions in general, it was espe-
CTBE therapists were paid with study funds, cially evident when considering only emer-
and the study coordinator helped clients con- gency room visits and admissions due to
tact therapists and arrange transportation to SI. During the treatment year, CTBE sub-
the first meeting. jects were twice as likely as DBT subjects
Standard DBT was administered by 16 to visit the emergency room for SI (33.3%
therapists who were nominated by colleagues CTBE versus 15.6% DBT) and three times
based on their potential to be good DBT ther- as likely to be admitted for SI (35.6% CTBE
apists. Of the 16, 8 had no prior DBT ex- versus 9.8% DBT) (Linehan et al. 2006b).
194 Lynch et al.

Consistent with prior studies, subjects in DBT criteria. Furthermore, at a six-month follow-
were also significantly less likely to change up evaluation, clients in the DBT condition
therapists or drop out of treatment. Both had significantly higher remission rates (75%)
groups improved significantly and similarly than those in the medication-only condition
on measures of depression, hopelessness, sui- (31%). Clients in the DBT condition also had
cidality, and reasons for living. No measured significant improvement on measures of adap-
outcomes favored CTBE. tive coping and dependency, whereas those in
the medication-alone condition did not. The
authors hypothesize that the improvements in
RANDOMIZED CONTROLLED these areas reduce vulnerability to depression.
TRIALS OF DIALECTICAL The main objective of this first study
BEHAVIOR THERAPY FOR (Lynch et al. 2003) was to determine the feasi-
CLIENTS WITH OTHER bility of a group intervention with a skills ori-
DIAGNOSES entation for older adults. Encouraged by these

findings, a second randomized clinical trial
Dialectical Behavior Therapy for was conducted to apply standard DBT (both
Depression and Other Personality group and individual) to older adults with ma-
Disorders jor depression and personality disorder with
Several RCTs have examined applications of the goal of modifying the DBT specifically
DBT for populations other than individuals for this population (Lynch et al. 2006b).
with BPD. In one such pilot study, Lynch In this second study, 35 adults over the age
et al. (2003) randomly assigned 34 adults of 55 with personality disorders and comorbid
over the age of 60 in a current major de- depressive symptoms were randomly assigned
pressive episode to either an antidepressant to either 24 weeks of medication manage-
medication alone condition or an antidepres- ment alone or 24 weeks of medication man-
sant medication plus a modified form of DBT agement plus standard DBT. The DBT con-
condition. The modified form of DBT condition included in-person weekly individual
sisted of 28 weeks of a skills-training group sessions and group skills training. Those in the
as well as six months of weekly 30-minute DBT condition showed significantly greater
phone contact with an individual therapist, decreases in interpersonal sensitivity and in-
followed by three months of once every two terpersonal aggression compared with medi-
weeks and three months of once every three cation alone. Additionally, assessment at the
weeks 30-minute phone contact. Phone con- time point corresponding with the end of the
tacts in the first six months focused on re- DBT skills group indicated that 71% of clients
view of diary cards and problem-solving dif- in the DBT condition met criteria for remis-
ficulties with applying skills, whereas phone sion of depression, whereas only 50% of those
contacts in the second six months focused on in the medication-alone condition met crite-
use of skills to prevent depression relapse. ria for remission. Clients in both conditions
Those in the DBT condition showed sig- showed significant reductions on standard-
nificantly greater improvements than those ized, clinician-administered ratings of depres-
in the medication-alone condition in areas sion, with a nonsignificant difference favoring
including self-rated depression at treatment DBT seen at end of treatment and follow-up
end and interviewer-rated depression scores assessments. Nine clients in the DBT condi-
at six-month follow-up. Post-treatment in- tion and seven clients in the medication-alone
terviewer ratings of depression indicated that condition no longer met diagnostic criteria
71% of clients in the DBT condition met for personality disorder after treatment com-
criteria for remission, whereas only 47% of pletion. Moderate effect sizes on several vari-
clients on medication alone met remission ables suggest that this study may have been

underpowered to detect significant differ- 89%, respectively. The DBT condition had a
ences between conditions. dropout rate of 0%. Although not statistically
As mentioned above, a major goal of this significant, moderate effect sizes favoring
second study was to modify standard DBT, DBT were found for negative affect overall
and a major impetus for the modifications and eating due to anger/frustration, anxiety,
came from the clinical observation that older and depression. In a second study, Telch et al.
adults were generally fairly adept at emotion (2001) randomly assigned 44 women meeting
regulation but had substantial difficulty em- full DSM-IV research criteria for binge eating
bracing the new perspectives and behavioral disorder either to a wait-list control condition
changes important for healing. These obser- or to a modified 20-week DBT skills group.
vations sparked development of a new form of The group met weekly for two hours and in-
the biosocial theory for this population that cluded units on core mindfulness skills, dis-
retains the basic tenets of the original theory tress tolerance, and emotion-regulation skills.
(e.g., maladaptive behavior produced through At the end of the 20 weeks, significantly more
interaction of biological predisposition and of those in the DBT condition than in the con-
environmental shaping) but recasts the fun- trol condition (89% versus 12.5%) were ab-
damental dialectic as being between rigidity stinent from binge eating episodes for at least
(fixed mind) and openness to experience (fresh four weeks. Additionally, those in the DBT
mind). In addition to those skills taught in condition had significantly fewer binge days,
standard DBT, new skills in this modification binge episodes, weight concerns, shape con-
emphasize reducing rigid mindsets, increasing cerns, eating concerns, and lower urges to eat
openness to new experience, and reconciling when angry than those in the control con-
events over the life course through reviewing dition. More studies with larger numbers of
past events and generating forgiveness. subjects are needed to further examine the ef-
fectiveness of these adaptations of DBT for
treating problematic eating behaviors.
Dialectical Behavior Therapy for
Eating Disorders
Two RCTs have also examined DBT for eat- Quasi-Experimental Studies
ing disorders. Safer et al. (2001) randomly A number of nonrandomized, quasi-
assigned 29 women with at least one binge- experimental studies have examined DBT
and-purge incident per week over the pre- (see Table 3). Two controlled but nonran-
vious three months to either 20 weeks of domized studies show promise for adaptations
wait-list control or to a modified form of of DBT with suicidal adolescents with BPD
DBT. The adapted DBT condition conceptu- symptoms in both outclient settings (Rathus
alized binge-and-purge behaviors as attempts & Miller 2002) and inpatient settings (Katz
at emotion regulation and therefore involved et al. 2004). Other quasi-experimental studies
weekly individual therapy sessions focusing have suggested beneficial effects of DBT
on teaching alternative emotion-regulation in adult inpatient units (Barley et al. 1993,
skills. Clients in the DBT condition had sig- Bohus et al. 2000), in forensic settings
nificantly greater reductions in binge episodes (McCann et al. 2000, Trupin et al. 2002),
and purge episodes than those in the control and in females with binge eating disorder
condition. Significantly more subjects in the (Telch et al. 2000). However, although these
DBT condition than in the control condition studies may indicate interesting directions
were abstinent from binge-and-purge behav- for future research, conclusions that can
iors at the end of 20 weeks (28.6% versus 0%). be drawn from them are limited. RCTs are
Five additional DBT subjects reduced their needed to establish these adaptations of DBT
bingeing and purging episodes by 88% and as empirically supported therapies.
196 Lynch et al.

Table 3 Summary of quasi-experimental studies of dialectical behavior therapy (DBT)

Treatment(s) Participants Length Main effects Reference
Adapted inpatient Women with BPD 12 weeks Group comparisons: depression, Bohus et al. 2004
DBT (N = 31) versus anxiety, interpersonal functioning,
waiting list with social adjustment, global
community TAU psychopathology, and
(N = 19) self-mutilation
Adapted DBT Adolescent inpatients in 12 weeks Group comparisons: treatment Rathus & Miller
(N = 29) versus depression/suicide retention, psychiatric 2002
supportive-dynamic program; those with hospitalization
therapy + family suicide attempt and ≥3 Pre-/post-DBT comparison:
therapy (N = 82) BPD criteria assigned to suicidal ideation, general
DBT psychiatric symptoms, BPD
symptoms
Adapted DBT Adolescent inpatients 2 weeks Group comparisons: problem Katz et al. 2004
(N = 32) versus TAU with suicide attempt or behavior on the ward
(N = 30) ideation
Pre-/postincorporation Consecutive admissions Average stay Pre-/postincorporation of DBT: Barley et al. 1993
of DBT onto adult during transition from 106 days mean monthly self-harm rate on
inpatient unit psychodynamic to DBT the unit
(N = 130) + psychodynamic
treatment
Adapted adult inpatient Females with BPD + ≥2 Average stay Pre-/postincorporation of DBT: Bohus et al. 2000
DBT (N = 24) suicide attempts and/or 94 days self-harm behaviors, depression,
ISI incidents in 2 years dissociation, anxiety, global stress
Adapted DBT Inpatients on adult 20 months Depressed and hostile mood, McCann et al.
(N = 21) versus TAU forensic unit with at paranoia, psychotic behaviors, 2000
(N = 14) least three BPD criteria maladaptive coping, adaptive
coping, staff burn-out (trend)
Adapted DBT Inpatient adolescent Variable Pre-/postincorporation of DBT: Trupin et al. 2002
females on a forensic behavioral problems; staff use of
unit restrictive punishments;
participation in therapeutic,
educational, and vocational
services
Adapted DBT for Females age 18–65 with 20 weeks Pre-/postincorporation of DBT: Telch et al. 2000
binge eating disorder binge eating disorder binge episodes, binge days
(N = 11)
BPD, borderline personality disorder; ISI, intentional self-injury; TAU, treatment as usual.
GENERAL ISSUES relevant. To date five independent research

One issue that has reverberated throughout labs have conducted DBT RCTs showing
the BPD and personality disorder research positive between-group effects (Koons et al.
field has been a concern as to whether DBT 2001; Lynch et al. 2003, 2006b; Safer et al.
can be successfully translated to the commu- 2001; Telch et al. 2001; Turner 2000a; Verheul
nity settings that serve many of the individuals et al. 2003), suggesting clearly that the efficacy
it was designed to treat. Recent developments of the treatment is not dependent on specific
immediately suggest that this issue may be less people or organizations. Two of these RCTs

(i.e., Turner 2000a, Verheul et al. 2003) were DBT with CTBE (Linehan et al. 2006b). In
conducted in nonacademic community clin- this study, the CTBE therapists were nom-
ics, indicating that dissemination of DBT to inated by community mental health leaders
community settings is both feasible and ef- and were considered experts in treating dif-
fective. Further support for DBT’s general- ficult clients. The content of the treatment
izability comes from multiple published pre- provided by them was not prescribed by the
/postdesign and nonrandomized controlled research study or interfered with, and insti-
studies (Brassington & Krawitz 2006, Bohus tutional prestige was controlled for by having
et al. 2004, Comtois et al. 2007, Katz et al. the base of operations for CTBE at the Seattle
2004, McCann et al. 2000, Rathus & Miller Psychoanalytic Society and Institute. In addi-
2002, Trupin et al. 2002) and from unpub- tion, therapists were told to provide the treat-
lished data from community clinics across the ment they had the greatest allegiance to (i.e.,
country that have been compiled by Behav- the treatment they thought would work best),
ioral Tech, LLC, a training company focused and there were no differences between condi-
on the dissemination of evidence-based treat- tions in expectancies. Thus, it is reasonable to
ments (L. Dimeff, personal communication). conclude that allegiance was high in CTBE.
Additional issues that may impact gener-

alizability include at least three primary char-
acteristics that can distinguish between aca- CURRENT DEVELOPMENTS
demic research and treatment programs in AND FUTURE DIRECTIONS
which treatments are developed and the com- The accumulated data clearly indicate that
munity treatment programs in which they are DBT is an effective treatment for BPD.
disseminated: to clients, counselors, and set- Across studies, DBT has resulted in reduc-
tings. A major strength of DBT is that it is tions in several problems associated with
explicitly designed to treat clients who typ- BPD, including self-injurious behavior, sui-
ically present for treatment in community cide attempts, SI, hopelessness, depression,
treatment programs (e.g., highly symptomatic and bulimic behavior. Nonetheless, further
individuals with a high degree of psychi- advances in the treatment of this complex dis-
atric comorbidity). Accordingly, participants order are needed. The question now involves
in the RCTs described above included het- how best to move the field forward.
erogeneous samples of individuals with high
axis I and II comorbidity. Overall, additional
research in applying DBT in real-world set- Are Direct Comparisons with Other
tings is needed, and this includes an evaluation Borderline Personality Disorder
of the effects training has on changing thera- Treatments Needed?
pist behavior. In our opinion, a so-called horse-race study in
An additional concern regarding DBT re- which another multicomponent treatment is
search has been the influence of allegiance systematically compared with DBT does not
effects on outcome (Westen 2000). Despite appear warranted at this point. To date there
criticisms regarding the importance of alle- have been only two other RCTs for treatment
giance to therapy outcome (e.g., Chambless of BPD. The first was Bateman & Fonagy’s
2002), it is reasonable to conclude that al- (1999) study of a psychodynamic partial hos-
legiance effects on psychotherapy outcomes pital program, but this has not been replicated
may influence what treatment wins (Luborsky either by the authors or in a second indepen-
et al. 1999). How well has DBT dealt with dent lab. Interested readers may refer to a spe-
this issue? The most representative outcome cial issue of the Journal of Personality Disor-
study that systematically controlled for alle- ders (volume 16, issue 2) that was devoted to
giance effects has been a study comparing the American Psychiatric Association’s (2001)
198 Lynch et al.

practice guidelines for the treatment of BPD used in recent DBT trials constitute a first step
and addresses weaknesses of research exam- toward identifying these factors. The find-
ining psychoanalytically informed treatments ings thus far, particularly from the compar-
(e.g., Sanderson et al. 2002). The second RCT ison of DBT with CTBE, suggest that su-
was the Giesen-Bloo et al. (2006) study of perior results obtained with DBT cannot be
schema-focused therapy versus transference- solely attributed to therapist expertise, expe-
focused psychotherapy, but again this has not rience, gender, and allegiance; institutional
been replicated. Thus, there is currently no prestige; availability of supervision and af-
other well-established treatment for BPD that fordable treatment; assistance to connect with
would allow comparison with DBT, and the therapist; hours of individual therapy; or other
treatments referenced above were of much nonspecific factors. A slightly different ap-
longer duration than standard DBT, which proach to this issue involves the systematic
mitigates direct comparisons. Assuming that rating of specific therapist behaviors in in-
the efficacy of another treatment for BPD dividual sessions. Such a study is currently
can be firmly established, it is still not en- under way in conjunction with a multisite
tirely clear that a direct comparison with DBT study (Duke University, Principal Investiga-
would represent a wise use of resources. This tor T.R. Lynch; University of Washington,
is not to say that such a comparison might Principal Investigator M.M. Linehan) com-
not provide useful information. Rather, the paring DBT with an established, manualized
argument against a direct comparison study treatment for substance abuse (individual and
in this case rests more on whether it is prac- group drug counseling) in the treatment of co-
tical or even logistically feasible owing to the morbid BPD and heroin dependence. How-
large sample size that would be required, and ever, although this line of inquiry can tell us
the inherently complex nature of the disorder which factors do not account for DBT’s effi-
and, thus, of the treatments being compared. cacy, it is relatively less informative regarding
The likely incremental gains from such a study which elements of DBT do make it exception-
would not offset, in our opinion, the tremen- ally effective.
dous costs. Efforts to further define the relative im-
portance of different aspects of DBT are
underway. One current approach involves
Dismantling Studies and Testing comparing standard DBT with modified or
Mechanisms of Change dismantled forms in which one or more treat-
From our perspective, a more appropriate and ment elements are missing. For example, the
cost-effective approach would involve deter- individual impact of skills training has been
mining what makes effective treatments work assessed in two published RCTs and one un-
and using that information to improve patient published report. Both published studies were
outcomes. Accordingly, an important goal of adaptations of DBT that included skills train-
current DBT research is to identify the essen- ing as the primary intervention with some
tial features of the treatment to improve its ef- features of individual therapy built into the
ficacy, efficiency, and generalizability. The re- treatment (Lynch et al. 2003, Telch et al.
cent work of Linehan and others (Lynch et al. 2001). Results suggested that skills training
2006a) reflects commitment to this goal. One alone with minimal individual therapy con-
approach to this problem involves assessing tact may be helpful for less severe disorders
the degree to which DBT strategies are used (e.g., eating disorders, chronic depression).
in non-DBT comparison treatments to deter- However, the nonpublished report (Linehan
mine which strategies and factors are unique 1993a, p. 25) found that the outcomes for
to DBT and, thus, might explain its greater 11 clients who received DBT skills training
efficacy. The rigorous comparison conditions in addition to non-DBT individual therapy

were no better than 8 clients receiving non- multiple measurement points to determine
DBT individual therapy only. This suggests both a gradient (dosage effect) as well as a
that skills training may be an important com- time line (i.e., changes in the mechanism of
ponent of the full treatment package, espe- action precede changes in outcome). Finally,
cially for more severely affected populations. the proposed mechanism of action must stand
A related dismantling study is currently up to tests of plausibility and coherence. In
ongoing at UW (P.I. Linehan). In this study, other words, there must be a credible expla-
women with BPD and histories of suicidal and nation for how and why the mechanism results
other self-injurious behaviors are randomly in change. Theory, then, is an important over-
assigned to one of three treatment conditions: arching element in the testing of mechanisms
(a) standard DBT, (b) individual therapy plus of action. The more assessment periods that
activity support group, in which the DBT are included, the more fine grained the anal-
skills-training group is replaced by a struc- ysis of gradient and time line can be.
tured weekly group activity and DBT skills are Based on the data accumulated thus far,
not taught in individual therapy, or (c) DBT Lynch et al. (2006a) have posited several
skills training plus case management, which mechanisms of action specific to DBT that
includes no individual DBT therapy. distinguish this treatment from other behav-
Preliminary research regarding the rela- ioral interventions. For example, based on the
tive importance of change versus acceptance dialectical change theory, the authors sug-
strategies in DBT has also been undertaken. gest that a dialectical focus with a synthesis
The study comparing DBT with CVT + 12- of change and acceptance strategies may be
step for treatment of comorbid BPD and sub- an important mechanism of action in DBT.
stance dependence (Linehan et al. 2002) rep- They suggest that strategies specific to DBT
resents the first step in this direction. Results used in both individual sessions (e.g., utiliz-
of that study suggest that validation strategies ing commitment strategies, focusing on DBT
may be critical for preventing dropout among skills such as opposite action, and high ther-
subjects with comorbid BPD and substance apist self-disclosure) and group skills training
dependence (Linehan et al. 2002). We can- (e.g., mindfulness skills, emotion-regulation
not draw firm conclusions regarding the im- skills, interpersonal effectiveness skills, and
portance of change strategies from that study, self-respect effectiveness skills) may account
however, as change strategies were included for significant clinical change.
in the control condition as part of the struc- In addition to the initiatives mentioned
tured 12-step intervention. In another study, above, ongoing research includes evaluation
Shearin & Linehan (1992) examined individ- of adaptations of DBT to non-BPD diag-
ual sessions and found that a combination noses, mediator/moderator studies, and basic
of change and validation strategies was im- research examining the theoretical precepts of
portant. Specifically, sessions in which clients DBT (e.g., biosocial theory; see Linehan et al.
rated therapists as maintaining a balance be- 2006a for a review).
tween change and validation strategies were
associated with greater reductions of parasui-
cidal behavior and ideation relative to sessions CONCLUSION
during which the therapist was rated as purely The primary purpose of this review was to
accepting or change focused. conservatively scrutinize the status of DBT
Future studies must be designed in which research and evaluate the rigor with which
a strong association between the mechanism criticisms of prior research have been ad-
of action and both pretreatment variables dressed to date. Using the criteria for man-
and post-treatment variables can be demon- ualized treatments established by Chambless
strated. Additionally, future studies should use & Hollon (1998), we found that the current
200 Lynch et al.

literature quickly reveals that DBT is the RCTs. Once it is known that a treatment is
only treatment for BPD considered well es- efficacious, the next task is to improve the
tablished or efficacious and specific. How- treatment further by enhancing its efficiency
ever, despite its strong empirical foundation, and efficacy (Linehan et al. 1999). This phase
a number of gaps do remain in the DBT lit- of treatment development includes compo-
erature. These include a relative paucity of nent and process-analytic studies, dismantling
RCTs involving male or minority clients and studies, analysis of response predictors, and
little information on the relative importance large-sample effectiveness research in com-
of DBT’s different components to treatment munity settings. We hope that this review pro-
outcomes. In addition, although preliminary vides the impetus for others to expand re-
attempts to apply DBT to diagnoses other search efforts into these new domains and
than BPD have been promising, these appli- continue a tradition based on empirical ob-
cations should still generally be considered servation to maximize the likelihood that the
experimental pending further evidence from treatment helps those it was designed to help.
SUMMARY POINTS
1. DBT has been reformulated and conceptualized as a treatment for multidiagnostic
treatment-resistant populations. It has been evaluated and found to be efficacious for
the treatment of BPD in seven well-controlled RCTs conducted across four indepen-
dent research teams.
2. Treatment approaches can be distilled down into the following process: the reduction
of ineffective action tendencies linked with dysregulated emotions. However, studies
examining specific mechanisms of change need further development.
3. There is a reliable measure of treatment adherence that generates a single item index of
DBT adherence and subscale scores for the 12 DBT strategy domains. Dissemination
of the treatment may be slowed by an overreliance on the UW site for adherence
ratings. However, until a convincing argument can be made that a new adherence
scale is needed, it is probably best to consider the UW scale as the gold standard.
4. DBT has demonstrated efficacy in RCTs for chronically depressed older adults, older
depressed adults with comorbid personality disorder, and eating-disordered individu-
als. Although preliminary attempts to apply DBT to diagnoses other than BPD have
been promising, these applications should still generally be considered experimental
pending further evidence from RCTs.
5. DBT can be successfully conducted outside of UW as evidenced by the positive out-
comes from independent research teams, adherence ratings of therapists, and com-
munity treatment involvement. Allegiance effects have been recently controlled for
in a rigorous randomized trial (Linehan et al. 2006a).
FUTURE ISSUES
1. Future research should focus on component and process-analytic studies, dismantling
studies, and studies designed to analyze response predictors.
2. New adaptations of DBT require further testing using evidence from RCTs.

3. Research should emphasize inclusion of males and minority populations in future

studies to enhance treatment generalizability.
4. Future research should examine factors that enhance translation of the treatment into
community settings, and large-sample effectiveness research in community settings
should be conducted to test this.
ACKNOWLEDGMENTS
The authors thank Linda Dimeff and Kathryn Korslund for their extremely helpful editorial
comments and assistance, as well as Sammy Banawan for his contributions to the literature
search and manuscript organization.

LITERATURE CITED
Am. Psychiatr. Assoc. 2001. Practice Guideline for the Treatment of Patients With Borderline Per-
sonality Disorder. Washington, DC: Am. Psychiatr. Assoc.
Barley WD, Buie SE, Peterson EW, Hollingsworth AS, Griva M, et al. 1993. Development of
an inpatient cognitive-behavioral treatment program for borderline personality disorder.
J. Personal. Disord. 7:232–40
Bateman A, Fonagy P. 1999. Effectiveness of partial hospitalization in the treatment of bor-
derline personality disorder: a randomized controlled trial. Am. J. Psychiatry 156:1563–
69
Bohus M, Haaf B, Simms T, Limberger M, Schmahl C, et al. 2004. Effectiveness of inpatient
dialectical behavioral therapy for borderline personality disorder: a controlled trial. Behav.
Res. Ther. 42:487–99
Brassington J, Krawitz R. 2006. Australasian dialectical behavior therapy pilot outcome study:
effectiveness, utility and feasibility. Australas. Psychiatry. In press
Campbell DT. 1957. Factors relevant to the validity of experiments in social settings. Psychol.
Bull. 54:297–312
Chambless DL. 2002. Beware the dodo bird: the dangers of overgeneralization. Clin. Psychol.
9:13–16
Chambless DL, Hollon SD. 1998. Defining empirically supported therapies. J. Consult. Clin.
Psychol. 66:7–18
Chambless DL, Ollendick TH. 2001. Empirically supported psychological interventions: con-
troversies and evidence. Annu. Rev. Psychol. 52:685–716
Chapman AL, Linehan MM. 2005. Dialectical behavior therapy for borderline personality
disorder. In Borderline Personality Disorder, ed. M. Zanarini, pp. 211–42. Boca Raton, FL:
Taylor & Francis
Cohen J. 1988. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Erlbaum. Rev.
ed. 474 pp.
Comtois KA, Elwood L, Holdcraft LC, Simpson TL, Smith WR. 2007. Effectiveness of di-
alectical behavior therapy in a community mental health center. Cogn. Behav. Pract. In
press
Giesen-Bloo J, van Dyck R, Spinhoven P, van Tilburg W, Dirkson C, et al. 2006.
Outpatient psychotherapy for borderline personality disorder: randomized trial of
202 Lynch et al.

schema-focused therapy vs transference-focused psychotherapy. Arch. Gen. Psychiatry

63:649–58
Katz LY, Cox BJ, Gunasekara S, Miller AL. 2004. Feasibility of dialectical behavior ther-
apy for suicidal adolescent inpatients. J. Am. Acad. Child Adolesc. Psychiatry 43:276– Presents findings
from the first major
82
RCT of DBT for
Koons CR, Robins CJ, Tweed J, Lynch TR, Gonzalez AM, et al. 2001. Efficacy of dialectical borderline
behavior therapy in women veterans with borderline personality disorder. Behav. Therapy personality
32:371–90 disorder;
demonstrated
Levendusky PG. 2000. Dialectical behavior therapy: so far so soon. Clin. Psychol. 7:99–
DBT’s efficacy for
100 reducing
Linehan M, Armstrong HE, Suarez A, Allmon D, Heard HL. 1991. Cognitive-behavioral intentional
treatment of chronically parasuicidal borderline patients. Arch. Gen. Psychiatry self-injury.
48:1060–64
Linehan MM. 1993a. Cognitive-Behavioral Treatment of Borderline Personality Disorder. Describes the
New York: Guilford development and
theory of DBT,
Linehan MM. 1993b. Skills Training Manual For Treating Borderline Personality Disorder.
with detailed
New York: Guilford review of
Linehan MM, Bohus M, Lynch TR. 2006a. Dialectical behavior therapy for pervasive techniques used
emotion dysregulation: theoretical and practical underpinnings. In Handbook of and tips for clinical
application.
Emotion Regulation, ed. J Gross. New York: Guilford. In press
Linehan MM, Comtois KA, Murray AM, Brown MZ, Gallop RJ, et al. 2006b. Two-
year randomized controlled trial and follow-up of dialectical behavior therapy vs Includes specific
therapy by experts for suicidal behaviors and borderline personality disorder. Arch. worksheets and
exercises for group
Gen. Psychiatry 62:1–10
skills training;
Linehan MM, Dimeff LA. 1997. Dialectical Behavior Therapy Manual of Treatment Interventions applicable to both
for Drug Abusers with Borderline Personality Disorder. Seattle: Univ. Wash. individual and
Linehan MM, Dimeff LA, Reynolds SK, Comtois KA, Welch SS, et al. 2002. Dialectical group sessions.
behavior therapy versus comprehensive validation therapy plus 12-step for the treatment
of opioid dependent women meeting criteria for borderline personality disorder. Drug
Discusses DBT
Alcohol Depend. 67:13–26 techniques aimed
Linehan MM, Heard HL, Armstrong HE. 1993. Naturalistic follow-up of a behavioral at improving
treatment for chronically parasuicidal borderline patients. Arch. Gen. Psychiatry 50:971– emotion regulation
and the importance
74
of emotion
Linehan MM, Korslund KE. 2003. Dialectical Behavior Therapy Adherence Manual. Seattle: Univ. regulation in BPD
Wash. and other
Linehan MM, Schmidt H, Dimeff LA, Craft JC, Kanter J, Comtois KA. 1999. Dialectical disorders.
behavior therapy for patients with borderline personality disorder and drug dependence.
Am. J. Addict. 8:279–92 Describes results of
Linehan MM, Tutek DA, Heard HL, Armstrong HE. 1994. Interpersonal outcome of cog- study designed to
nitive behavioral treatment for chronically suicidal borderline patients. Am. J. Psychiatry replicate/extend
findings of
151:1771–76
landmark 1991
Luborsky L, Barber JP, Siqueland L, McLellan ET, Woody G. 1997. Establishing a therapeutic study using larger
alliance with substance abusers. NIDA Res. Monogram 165:233–44 sample and
Luborsky L, Diguer L, Seligman DA, Rosenthal R, Krause ED, et al. 1999. The researcher’s meticulous control
of nonspecific
own therapy allegiances: a “wild card” in comparisons of treatment efficacy. Clin. Psychol.
treatment factors.
6:95–106

Lynch TR, Chapman AL, Rosenthal MZ, Kuo JR, Linehan M. 2006a. Mechanisms
Proposes
hypothesized of change in dialectical behavior therapy: theoretical and empirical observations.
mechanisms of J. Clin. Psychol. 62:459–80
change associated Lynch TR, Cheavens JS, Cukrowicz KC, Thorp SR, Bronner LL, Beyer JL. 2006b. Treatment
with DBT of older adults with co-morbid depression and personality disorder: a dialectical behavior
techniques
therapy approach. Int. J. Geriatr. Psychiatry. Publ. online Nov 10. DOI: 10.1002/gps.1703
including
mindfulness, Lynch TR, Morse JQ, Mendelson T, Robins CJ. 2003. Dialectical behavior therapy for de-
validation, opposite pressed older adults: a randomized pilot study. Am. J. Geriatr. Psychiatry 11:33–45
action, and McCann RA, Ball EM, Ivanoff A. 2000. DBT with an inpatient forensic population: the CMHIP
dialectics. forensic model. Cogn. Behav. Pract. 7:447–56
Miller SJ, Binder JL. 2002. The effects of manual-based training on treatment fidelity and
outcome: a review of the literature on adult individual psychotherapy. Psychother. Theory
Res. Pract. 39:184–98
Rathus JH, Miller AL. 2002. Dialectical behavior therapy adapted for suicidal adolescents.
Suicide Life Threat. Behav. 32:146–57
Robins CJ, Chapman AL. 2004. Dialectical behavior therapy: current status, recent develop-
ments, and future directions. J. Personal. Disord. 18:73–89

Safer DL, Telch CF, Agras W. 2001. Dialectical behavior therapy for bulimia nervosa. Am. J.
Psychiatry 158:632–34
Sanderson C, Swenson C, Bohus M. 2002. A critique of the American psychiatric practice
guideline for the treatment of patients with borderline personality disorder. J. Personal.
Disord. 16:122–29
Scheel KR. 2000. The empirical basis of dialectical behavior therapy: summary, critique, and
implications. Clin. Psychol. 7:68–86
Shearin EN, Linehan MM. 1992. Patient-therapist ratings and relationship to progress in
dialectical behavior therapy for borderline personality disorder. Behav. Ther. 23:730–
41
Swenson CR. 2000. How can we account for DBT’s widespread popularity? Clin. Psychol. 7:87–
91
Telch CF, Agras W, Linehan MM. 2000. Group dialectical behavior therapy for binge eating
disorder: a preliminary uncontrolled trial. Behav. Ther. 31:569–82
Telch CF, Agras W, Linehan MM. 2001. Dialectical behavior therapy for binge eating disorder.
J. Consult. Clin. Psychol. 69:1061–65
Trupin EW, Stewart DG, Beach B, Boesky L. 2002. Effectiveness of dialectical behavior therapy
program for incarcerated female juvenile offenders. Child Adolesc. Ment. Health 7:121–
27
Turner RM. 2000a. Naturalistic evaluation of dialectical behavior therapy-oriented treatment
for borderline personality disorder. Cogn. Behav. Pract. 7:413–19
Turner RM. 2000b. Understanding dialectical behavior therapy. Clin. Psychol. 7:95–98
van den Bosch LMC, Koeter MWJ, Stijnen T, Verheul R, van den Brink W. 2005. Sustained
efficacy of dialectical behavior therapy for borderline personality disorder. Behav. Res.
Ther. 43:1231–41
van den Bosch LMC, Verheul R, Schippers GM, van den Brink W. 2002. Dialectical behavior
therapy of borderline patients with and without substance use problems: implementation
and long-term effects. Addict. Behav. 27:911–23
Verheul R, van den Bosch LM, Koeter MW, De Ridder MA, Stijnen T, van den Brink W. 2003.
Dialectical behavior therapy for women with borderline personality disorder: 12-month,
randomised clinical trial in The Netherlands. Br. J. Psychiatry 182:135–40
204 Lynch et al.

Waltz J, Addis ME, Koerner K, Jacobson NS. 1993. Testing the integrity of a psychother-
apy protocol: assessment of adherence and competence. J. Consult. Clin. Psychol. 61:620–
30
Westen D. 2000. The efficacy of dialectical behavior therapy for borderline personality disor-
der. Clin. Psychol. 7:92–94
Widiger TA. 2000. The science of dialectical behavior therapy. Clin. Psychol. 7:101–3

AR307-FM ARI 2 March 2007 14:4
Annual Review of
Clinical Psychology
Contents Volume 3, 2007
Mediators and Mechanisms of Change in Psychotherapy Research

Alan E. Kazdin p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Evidence-Based Assessment
John Hunsley and Eric J. Mash p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 29

Internet Methods for Delivering Behavioral and Health-Related
Interventions (eHealth)
Victor Strecher p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 53
Drug Abuse in African American and Hispanic Adolescents: Culture,
Development, and Behavior
José Szapocznik, Guillermo Prado, Ann Kathleen Burlew, Robert A. Williams,
and Daniel A. Santisteban p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 77
Depression in Mothers
Sherryl H. Goodman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p107
Prevalence, Comorbidity, and Service Utilization for Mood Disorders
in the United States at the Beginning of the Twenty-first Century
Ronald C. Kessler, Kathleen R. Merikangas, and Philip S. Wang p p p p p p p p p p p p p p p p p p p p p137
Stimulating the Development of Drug Treatments to Improve
Cognition in Schizophrenia
Michael F. Green p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p159
Dialectical Behavior Therapy for Borderline Personality Disorder
Thomas R. Lynch, William T. Trost, Nicholas Salsman,
and Marsha M. Linehan p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p181
A Meta-Analytic Review of Eating Disorder Prevention Programs:
Encouraging Findings
Eric Stice, Heather Shaw, and C. Nathan Marti p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p207
Sexual Dysfunctions in Women
Cindy M. Meston and Andrea Bradford p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p233
Relapse and Relapse Prevention
Thomas H. Brandon, Jennifer Irvin Vidrine, and Erika B. Litvin p p p p p p p p p p p p p p p p p p p257
vii
AR307-FM ARI 2 March 2007 14:4
Marital and Family Processes in the Context of Alcohol Use and

Alcohol Disorders
Kenneth E. Leonard and Rina D. Eiden p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p285
Unwarranted Assumptions about Children’s Testimonial Accuracy
Stephen J. Ceci, Sarah Kulkofsky, J. Zoe Klemfuss, Charlotte D. Sweeney,
and Maggie Bruck p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p311
Expressed Emotion and Relapse of Psychopathology
Jill M. Hooley p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p329
Sexual Orientation and Mental Health
Gregory M. Herek and Linda D. Garnets p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p353
Coping Resources, Coping Processes, and Mental Health

Shelley E. Taylor and Annette L. Stanton p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p377
Indexes
Cumulative Index of Contributing Authors, Volumes 1–3 p p p p p p p p p p p p p p p p p p p p p p p p p p p403

Cumulative Index of Chapter Titles, Volumes 1–3 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p405
Errata
An online log of corrections to Annual Review of Clinical Psychology chapters (if any)
may be found at http://clinpsy.AnnualReviews.org
viii Contents

Dialectical Behavior Therapy For Borderline Personality Disorder

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dialectical Behavior Therapy For Borderline Personality Disorder

Uploaded by

Copyright:

Available Formats

ANRV307-CP03-08 ARI 21 February 2007 15:49

Thomas R. Lynch,1 William T. Trost,2

Annu. Rev. Clin. Psychol. 2007. 3:181–205 Key Words

Behavior Therapy . . . . . . . . . . . . . 186 Eating Disorders . . . . . . . . . . . . . . 196

BEHAVIOR THERAPY FOR CURRENT DEVELOPMENTS

INTRODUCTION this review are to continue with this approach

182 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 183

capabilities and motivation treatment manuals, supervision,

184 Lynch et al.

nally, interpersonal effectiveness training is

www.annualreviews.org • Dialectical Behavior Therapy 185

186 Lynch et al.

BORDERLINE PERSONALITY (ITT) analysis, in which all subjects who are

www.annualreviews.org • Dialectical Behavior Therapy 187

Table 2 Summary of randomized controlled trials of dialectical behavior therapy (DBT)

versus CVT + 12-step + current opiate

emergency services for impulsiveness, anger,

188 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 189

domization and an additional 2 dropped out measures of suicidality, affective dysregula-

190 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 191

192 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 193

194 Lynch et al.

DIAGNOSES entation for older adults. Encouraged by these

www.annualreviews.org • Dialectical Behavior Therapy 195

196 Lynch et al.

Table 3 Summary of quasi-experimental studies of dialectical behavior therapy (DBT)

GENERAL ISSUES relevant. To date ﬁve independent research

www.annualreviews.org • Dialectical Behavior Therapy 197

Additional issues that may impact gener-

198 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 199

200 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 201

3. Research should emphasize inclusion of males and minority populations in future

search and manuscript organization.

202 Lynch et al.

schema-focused therapy vs transference-focused psychotherapy. Arch. Gen. Psychiatry

www.annualreviews.org • Dialectical Behavior Therapy 203

ments, and future directions. J. Personal. Disord. 18:73–89

204 Lynch et al.

www.annualreviews.org • Dialectical Behavior Therapy 205

Contents Volume 3, 2007

Mediators and Mechanisms of Change in Psychotherapy Research

John Hunsley and Eric J. Mash p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 29

Marital and Family Processes in the Context of Alcohol Use and

Coping Resources, Coping Processes, and Mental Health

Cumulative Index of Contributing Authors, Volumes 1–3 p p p p p p p p p p p p p p p p p p p p p p p p p p p403

You might also like