Dialectical Behavior Therapy for Binge Eating and Bulimia (5 page)

Why Dialectical Behavior Therapy?

without histories of an eating disorder (Corstorphine, Mountford, Tomlinson,
Waller, & Meyer, 2007).

For those wishing further information about eating disorders and obesity,
excellent overviews are available in Fairburn and Brownell (2001) and Garner and
Garfnkel (1997).

RATIONALE
FOR
D
EVELOPMENT
OF
DBT
AS
A
DAPTED
FOR
BED
AND
BN

Existing treatments can ameliorate symptoms of BED and BN. These include (1)
CBT (Fairburn, 1995; Marcus, 1997; Wilson, Fairburn, & Agras, 1997), which
focuses on normalizing disordered eating patterns (i.e., decreasing dietary restraint)
and tackling overvalued ideas regarding weight and shape; (2) IPT (Klerman &
Weissman, 1993; Wilfey et al., 1993; Wilfey et al., 2002), which aims to resolve
interpersonal problems that maintain disordered eating; and (3) BWL (Agras et
al., 1994; Marcus, Wing, & Fairburn, 1995; Munsch et al., 2007), which stresses
decreasing the chaotic eating patterns and the overconsumption of calories char—
acteristic of obese clients with BED. Although such treatments address emotions
(e.g., an IPT focus on interpersonal role disputes or grief would address negative
emotions), none directly focuses both theoretically and specifcally on the role of
negative emotions in BED and BN.

The fact that a signifcant number of clients with BED and BN continue to suf—
fer from their eating-disorder symptoms either at posttreatment with CBT, IPT,
or BWL or over the period following treatment (Wilson et al., 2007) calls for other
theoretical conceptualizations and/or treatment approaches for BED and BN. One
such model is the affect regulation model. Drawing on an extensive literature that
links negative affect and disordered eating (Abraham & Beumont, 1982; Arnow,
Kenardy, & Agras, 1992, 1995; Polivy & Herman, 1993), the affect regulation
model conceptualizes binge eating and other types of eating pathology (e.g., vomit—
ing, restrictive eating) as behavioral attempts to infuence, change, or control pain—
ful emotional states (Linehan & Chen, 2005; Waller, 2003; Wiser & Telch, 1999;
Wisniewski & Kelly, 2003). The binge episodes appear to function in both BED and
BN by providing negative reinforcement or momentary relief from these aversive
emotions (Arnow et al., 1995; Polivy & Herman, 1993; Smyth et al., 2007; Stickney,
Miltenberger, & Wolff, 1999).

As neither CBT, IPT, nor BWL is grounded in the affect regulation model, a
new treatment based on remediating the hypothesized emotion regulation def—
cits in BN and BED was developed. DBT, originally developed by Linehan (1993a,
1993b), is the most comprehensive and empirically supported affect regulation
treatment for borderline personality disorder to date (American Psychiatric Asso—
ciation, 2001). Among others, Telch (1997a, 1997b) recognized that DBT’s conceptualization of self-injury as a functional (albeit maladaptive) affect regulation
behavior in patients with borderline personality disorder might provide a helpful
model for understanding the function (albeit maladaptive) of binge eating and/or
purging as emotion regulation behaviors in patients with disordered eating. Given
that DBT is specifcally designed to teach adaptive affect regulation skills and to

DIALECTICAL BEHAVIOR THERAPY FOR BINGE EATING AND BULIMIA
target behaviors resulting from affect dysregulation, a theoretical rationale exists
for applying DBT to treat BED and BN (see also McCabe, La Via, & Marcus, 2004;
Telch et al., 2000, 2001; Wisniewski & Kelly, 2003; Wiser & Telch, 1999).

RESEARCH
E
VIDENCE
FOR
DBT
FOR
BED
AND
BN

To date, preliminary studies investigating the adaptation of DBT to target disordered eating have been promising but limited to single case reports (Safer et
al., 2001a; Telch, 1997b), uncontrolled case series (Palmer et al., 2003; Salbach—
Andrae, Bohnekamp, Pfeiffer, Lehmkuhl, & Miller, 2008), uncontrolled trials
(Salbach, Klinkowski, Pfeiffer, Lehmkuhl, & Korte, 2007; Telch et al., 2000), and
three randomized controlled trials (Safer, Robinson, & Jo, in press; Safer et al.,
2001b; Telch et al., 2001).

The treatment described in this book is currently the only adaptation of DBT
for eating disorders that is supported through randomized trials in which clients
were assigned, by chance, either to DBT as adapted for BED (Safer et al., in press;
Telch et al., 2001) or BN (Safer et al., 2001b) or to a control condition (e.g., wait list
or active nonspecifc psychotherapy). Because factors that may infuence outcome
are distributed across groups randomly, the chance of a particular bias or factor
confounding the results is minimized. Hence, randomized control trials are consid—
ered the most reliable form of scientifc evidence for the effcacy of a clinical treatment (e.g., Chambless & Hollon, 1998).

This adapted DBT treatment was originally developed for adult women (ages
18–65) who met criteria for BED, BN, or partial BN (e.g., objective binge frequency
= 1 episode/week for 3 months versus 2 episodes/week per DSM-IV-TR full criteria; American Psychiatric Association, 2000). Criteria by which individuals were
excluded from entering the trials included (1) current use of psychotropic medica—
tions (Telch et al., 2001; Safer, Telch, & Agras, 2001b) or lack of a stable psychotropic dosage for the prior 3 months (Safer et al., in press); (2) psychotic or bipolar
affective disorders diagnoses; (3) current involvement in psychotherapy or weight
loss treatments; (4) current suicidality; (5) current substance abuse or dependence;
or (6) pregnancy. Clients with borderline personality disorder were not specifcally
excluded, although only a few participants met full criteria for borderline personality disorder.

In the frst randomized controlled trial of group DBT for BED, 89% (16 of 18)
who completed DBT were abstinent from binge eating (e.g., had no objective binge
episodes within the prior 4 weeks) by the end of the 20-week group treatment,
compared with 12.5% (2 of 16) of individuals randomized to a wait list (Telch et al.,
2001). The dropout rate was low. Of the original 22 assigned to DBT, only 9% (2 of
22) of those who attended at least the frst session dropped out. At posttreatment,
clients in DBT reported signifcantly improved weight and shape concerns, eating
concerns, and, on the Emotional Eating Scale (Arnow et al., 1995), signifcantly
reduced urges to eat when angry. At the 3-month and 6-month follow-up, 67% (12 of
18) and 56% (10 of 18), respectively, of the participants in DBT were abstinent from
binge eating. DBT clients also reported practicing on average 3.6 different skills
per week for an average of 4 days per week at the fnal assessment.

Why Dialectical Behavior Therapy?

The high abstinence rates were consistent with those of a smaller uncontrolled
trial of DBT for BED in which 82% of the participants were abstinent from binge
eating after 20 group sessions, with none dropping out after commencing treatment (Telch et al., 2000). Similar fndings were reported as part of a replication/
extension study of DBT for BED in which the client population was expanded to
include both men and women and individuals on stable psychotropic medications
(Safer et al., in press). Using a conservative statistical analysis that involved all
participants, including those who dropped out from treatment (i.e., the intent-to—
treat sample), the binge abstinence results for those receiving DBT for BED were
64% after 20 sessions, which was maintained at the 12-month follow-up (Safer et
al., in press). These rates are similar to abstinence rates reported for CBT and IPT
for BED by Wilfey
et al.
(1993, 2002). Long-term comparative studies are needed
to clarify the relative response rates of particular participants to different treatment approaches.

In the randomized controlled trial of group DBT for BN, 20 weeks of individually delivered DBT for bulimic symptoms was compared with a wait-list control.
Abstinence from binge-eating and purging behaviors at the end of 20 weeks of
treatment was 28.6% (4 of 14) for DBT and 0% (0 of 15) for the wait-list control
(Safer et al., 2001b). These fndings were similar to posttreatment abstinence rates
from the largest multisite CBT for BN trial (Agras, Walsh, Fairburn, Wilson, &
Kraemer, 2000). Importantly, the effect sizes (which denote the magnitude of the
effect of a treatment) showed that DBT resulted in moderate to large effect sizes
for several of the emotion regulation measures, thus providing support for the role
of decreasing affect dysregulation as a potential mechanism of DBT for BN. For
example, on the Emotional Eating Scale (EES; Arnow et al., 1995), participants
reported reduced urges to eat when angry, anxious, or depressed. In addition, the
Positive and Negative Affect Schedule (Watson, Clark, & Tellegen, 1988) showed
signifcant decreases in participants’ experiences of negative affect. At posttreatment the dropout rate in DBT was 0%.

Such initial positive results provided the impetus, propelled by repeated
requests from the clinical and research community, to provide a detailed descrip—
tion of how DBT was adapted to treat BED and the symptoms of BN. The result
is this book, based on Telch’s original manual (1997a). For those unfamiliar with
standard DBT, the following chapter offers an orientation to standard DBT and to
the adapted version for BED and BN.

CHAPTER
2

Orientation for Therapists

T
his chapter provides an orientation for therapists before they begin to imple—
ment treatment with clients. To provide context, we frst describe the impetus
behind this adapted DBT treatment’s development. This is followed by a brief
review of the empirical evidence for standard DBT. We then provide an introduc—
tion to this treatment model and assumptions and rationale that underlie this
therapy, as well as its goals and targets. The fnal section of the chapter focuses
on the delivery of this treatment, including basic therapist strategies and specifcs
regarding how sessions are structured.

IMPETUS FOR DEVELOPMENT OF THIS BOOK

The impetus for developing the treatment described in this book originated from
years spent by one of us (C. F. T.) treating clients with eating disorders and con—
ducting clinical research in this area. Part of this work included an ongoing search
for more effective treatments for eating disorders. As discussed in Chapter 1, a siz—
able number of individuals with BED and BN do not receive maximum beneft from
currently available psychotherapy treatments (e.g., CBT, IPT, BWL).

I reasoned that one potential explanation for this suboptimal treatment
response may be a failure of such treatments to directly target the emotional
aspects of binge eating. In other words, despite the considerable descriptive and
experimental research supporting the relationship between emotional distress and
disordered eating, neither CBT, IPT, nor BWL is based on an affect regulation
model for binge eating.

In seeking more effcacious treatments for binge eating, I discovered a treatment developed by Marsha Linehan for individuals with borderline personality
disorder: DBT. Standard DBT is based on the assumption that borderline personality disorder is best conceptualized as a dysfunction of the emotion regulation sys—
tem such that many impulsive behaviors (e.g., suicidal behavior and nonsuicidal

Orientation for Therapists

self-injury) are maladaptive attempts to regulate painful affects. As I investigated
this treatment and received formal training in DBT, I became more convinced
that the treatment model, principles, and strategies could be usefully adapted to
treating individuals with eating disorders; thus the development and research that
underlie this treatment manual.