lable at ScienceDirectJournal of Psychiatric Research 45 (.docx

lable at ScienceDirect
Journal of Psychiatric Research 45 (2011) 1511e1517
Contents lists avai
Journal of Psychiatric Research
journal homepage: www.elsevier.com/locate/psychires
Efficacy of a prevention program for postpartum
obsessiveecompulsive
symptoms
Kiara R. Timpano a,*, Jonathan S. Abramowitz b, Brittain L.
Mahaffey b, Melissa A. Mitchell c,
Norman B. Schmidt c
a University of Miami, FL, United States
b University of North Carolina, Chapel Hill, NC, United States
c Florida State University, FL, United States
a r t i c l e i n f o
Article history:
Received 19 April 2011
Received in revised form
21 June 2011
Accepted 22 June 2011
Keywords:
OCD
Postpartum
CBT
Prevention
* Corresponding author. University of Miami, Dep

Ponce de Leon Blvd, Coral Gables, FL 33146, United S
E-mail address: [email protected] (K.R. Tim
0022-3956/$ e see front matter � 2011 Elsevier Ltd.
doi:10.1016/j.jpsychires.2011.06.015
a b s t r a c t
ObsessiveeCompulsive Disorder (OCD) has emerged as a
common and impairing postpartum condition.
Prospective studies have identified psychological vulnerabilities
for the emergence of postpartum
obsessiveecompulsive symptoms (OCS), including general
anxiety symptoms, pre-existing OCS, and
specific cognitive distortions. The identification of these factors
makes feasible the development of
prevention programs that could reduce the impact of postpartum
OCS. The present investigation
examined a cognitive-behavioral prevention program using a
randomized, double blind, controlled trial.
Expecting mothers in their 2nd or 3rd trimester with an
empirically established, malleable risk factor for
postpartum OCS received either the prevention program (N ¼
38) or a credible control program (N ¼ 33),
both of which were incorporated into traditional childbirth
education classes. Results revealed that at 1
month, 3 months, and 6 months postpartum, the prevention
program was associated with significantly
lower levels of obsessions and compulsions than was the control
condition (all p’s < 0.05). Group
differences remained significant even after controlling for
baseline OCS and depression symptoms. Those
in the prevention condition also reported decreasing levels of
cognitive distortions, in contrast to the
control condition (p’s < 0.05). Results support the potential
utility of incorporating a CBT-based OCS

prevention program into childbirth education classes.
� 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Although much research has focused on maternal depression
during the perinatal (pregnancyand the first postpartumyear)
period
(Godfrey, 2005; Logsdon et al., 2006; Robertson et al., 2004),
our
understanding of perinatal anxiety disorders is still in the
nascent
stages. This is unfortunate given that anxiety disorders as a
group are
the most prevalent of all psychiatric conditions (Kessler et al.,
2005),
and perinatal maternal anxiety in particular is associated with
a number of adverse maternal and developmental repercussions
including: over-activation of the maternal endocrine system,
low
birth weight (Lou et al., 1994; Mulder et al., 2002), postpartum
depression (Ahmad et al., 1994; Sutter-Dallay et al., 2004),
reduced
communication with the infant (Field et al., 2005), behavioral
inhi-
bition, and insecure mothereinfant attachment (Coplan et al.,
2005;
Manassis et al.,1995). These adverse consequences, considered
along
artment of Psychology, 5665
tates.
pano).
All rights reserved.
with the high prevalence of perinatal maternal anxiety problems
(Heron et al., 2004), attest to the need for developing and

testing
effective prevention and treatment programs.
Obsessiveecompulsive disorder (OCD) is among the most
common postpartum anxiety disorders (Abramowitz et al.,
2003b;
Ross and McLean, 2006), and as such, is a desirable target for
prevention or treatment. OCD occurs in approximately 2% of
the
general population (American Psychiatric Association, 1994)
and
has been identified as one of the top ten leading causes of
disability
world-wide (Dupont, 1993; Lopez and Murray, 1998). Although
much less is known about OCD during the perinatal period,
there is
a growing consensus among researchers and clinicians that post-
partum obsessiveecompulsive symptoms (OCS) represent a
serious
problem that is under-identified, with many women not
receiving
needed services (Forray et al., 2010). Research to-date has
revealed
that OCS frequently occur during the perinatal period and can
range
in severity from mild to extremely impairing (Abramowitz et
al.,
2003a; Zambaldi et al., 2009). In addition, both pregnancy and
childbirth have been found to trigger onset of OCD or
exacerbation
of OCS (Abramowitz et al., 2003b; Forray et al., 2010). Up to
40% and
mailto:[email protected]
www.sciencedirect.com/science/journal/00223956
http://www.elsevier.com/locate/psychires

http://dx.doi.org/10.1016/j.jpsychires.2011.06.015
K.R. Timpano et al. / Journal of Psychiatric Research 45 (2011)
1511e15171512
30% of females with OCD report symptom onset during the
peri-
natal period (Williams and Koran, 1997) and the postpartum
(Buttolph and Hollander, 1990; Labad et al., 2005),
respectively. A
recent study in a small sample of women at 1 month postpartum
found that 31% reported subclinical OCS (Chaudron and Nirodi,
2010). Relatively little empirical work has focused on the
down-
stream consequences of OCS for both the parent and child;
however, in addition to causing the mother extreme distress,
postpartum OCS can also influence the type of care an infant
receives, family relationships and interactions, as well as
increase
the risk for developing further psychiatric disorders such as
depression (Abramowitz et al., 2003b).
Given these data, there is a need to substantially reduce the
personal, developmental and financial impact of OCD,
particularly if
vulnerable females can be identified during pregnancy and if an
effective OCD prevention program can be developed and imple-
mented. Longitudinal studies have identified antenatal factors
that
prospectively predict increases in postpartum OCS, including
(a)
higher levels of anxiety, (b) higher levels of pre-existing (non-
clinical) OCS, and (c) the presence of certain cognitive
distortions

(e.g., the belief that “bad” thoughts are equivalent to bad
actions)
(Abramowitz et al., 2003a, 2006). These variables represent
iden-
tifiable psychological vulnerabilities to perinatal OCD and raise
the
possibility that at-risk individuals can be identified. The
identifi-
cation of potentially malleable vulnerability factors is necessary
for
the development of a prevention program (Kraemer et al., 2001)
that could in turn demonstrably reduce the impact of postpartum
OCD and OCS.
Cognitive-behavioral models of OCD (Salkovskis et al.,1998)
and
postpartum OCS (Fairbrother and Abramowitz, 2007) posit that
symptoms arise when normally occurring, negative intrusive
thoughts (e.g., unwanted thoughts of harm befalling the infant)
are
misinterpreted as highly significant and threatening based on
faulty
beliefs about the importance of such thoughts (e.g., “If I think
this, it
is likely to happen”) and the need to control them (e.g., “I
should
never have any bad thoughts about the infant) (Larsen et al.,
2006).
Compulsive rituals (e.g., checking on the infant) subsequently
develop with the function of reducing anxiety and controlling
the
unwanted thoughts, but instead maintain the obsessional fear by
preventing the natural extinction of obsessional anxiety.
Research
indicates that about 70% of postpartum women report unwanted
infant-related intrusive thoughts (Abramowitz et al., 2003a),

and
two prospective studies found that the presence of antenatal
dysfunctional beliefs predicts more severe obsessions and
compulsions in the postpartum, over and above depression,
general
anxiety, and pre-existing OCS (Abramowitz et al., 2006, 2007).
Cognitive behavior therapy (CBT) is one efficacious treatment
for OCD, and includes techniques such as exposure, response
prevention, and cognitive therapy (Foa et al., 1998). CBT is
thought
to work by correcting cognitive distortions and weakening urges
to
perform anxiety-reducing rituals (McLean et al., 2001). The
knowledge of empirically supported interventions for reducing
both symptoms and vulnerabilities, in combination with the
identification of malleable risk factors, presents the opportunity
for
considering targeted prevention programs (Feldner et al., 2004).
The aim of the present investigation was to develop and test the
efficacy of a prevention program based on CBT principles for
postpartum OCS. Specifically, expecting mothers with an
empiri-
cally established, malleable risk factor for postpartum
OCSdobsessive beliefsdreceived either a CBT prevention
program
or a credible control program, both of which were incorporated
into
traditional childbirth education classes. We hypothesized that
compared to the control condition, the prevention program
would
be associated with lower OCS and lower scores on obsessive
beliefs
(e.g., importance of thoughts and the need to control them) at
one
month postpartum. We also hypothesized that symptom

reduction
would be maintained in the prevention condition across 6
months
postpartum, and that these findings would be specific to OCS by
statistically controlling for baseline depression symptoms.
2. Methods
2.1. Participants
The sample consisted of 71 expecting mothers who were fol-
lowed from their 2nd or 3rd trimester of pregnancy into the
post-
partum at two sites: Florida State University in Tallahassee, FL
(FSU
n ¼ 33) and the University of North Carolina at Chapel Hill
(UNC
n ¼ 38). Participants were recruited from local OB/Gyn clinics,
as
well as print and online advertisements in the respective
communities. In total, 306 women were screened: 217 did not
meet
inclusion/exclusion criteria and 18 declined to participate in the
classes. The primary reasons for declining to participate were
(1)
not being able to attend the groups due to scheduling
difficulties
and (2) choosing a program that was either recommended by, or
affiliated with the participant’s OB/Gyn. All participants met
the
following criteria: (a) identified as psychologically vulnerable
to
OCD, defined by a score of 139 or greater (1.25 SD above the
community mean) on the Obsessive Beliefs Questionnaire, (b)
age
18e65; (c) married or living with a partner; and (d) expecting

their
first child. Participants were excluded if they met criteria for
past or
current OCD, psychotic disorders, bipolar disorder, and/or
current
alcohol/substance abuse or dependence. Diagnoses were deter-
mined using the Structured Clinical Interview for DSM-IV
(SCID).
The sample’s mean age was 27.3 years (SD ¼ 4.2), and 47
(66%) of
the mothers were in their 2nd trimester. Participants endorsed
the
following ethnicities: Caucasian (74.6%), AfricaneAmerican
(16.9%),
mixed/other (8.4%). Four participants completed the prenatal
childbirth education classes with a female partner; three with
a female relative due to the father’s absence, and one with her
same-sex partner. The remainder of the sample completed the
classes with the father of the child. The mean household income
was $64,000.
2.2. Procedures and prevention program
The study design was a prospective, double (participant and
evaluator) blinded, randomized controlled trial. Participants and
their partners were randomly assigned to either the prevention
program (n ¼ 38) or control (n ¼ 33) condition, both of which
were
incorporated into a traditional childbirth education (CBE)
program.
The CBE þ prevention (prevention) and CBE þ control (control)
programs were delivered in 6 weekly group sessions prior to
childbirth. Each site enrolled 6 group cohorts. Participants were
followed for 6 months after the birth of their child to assess the
long-term effects of the prevention program. Assessments were

conducted by trained, doctoral level graduate students before
the
first class (i.e., diagnostic interview), at the conclusion of the
group,
1 month postpartum, 3 months postpartum and 6 months post-
partum. All postpartum interviews were conducted over the tele-
phone so as to reduce burden on the participants.
The basic CBE program consisted of 6 weekly 1.5 h group
meetings and covered a range of topics, including the stages of
labor, newborn characteristics, and birthing techniques. At each
site, a registered nurse certified in CBE delivered this portion
of the
classes. At the conclusion of each weekly CBE topic, a 30 min
session was added that addressed material relevant to either the
prevention or control condition. This latter portion of the class
was
delivered by trained study personnel (e.g., a psychology
graduate
student) from either the FSU or UNC site. The control condition
Table 1
Baseline scores on psychological variables.
M (SD) Prevention (n ¼ 38) Control (n ¼ 33) t Statistic p Value
Age 26.90 (3.6) 27.80 (4.8) �0.88 0.37
OBQ 172.70 (32.5) 166.64 (21.29) 0.92 0.36
EPDS 7.30 (5.10) 6.34 (4.63) 0.83 0.41
DOCS 9.30 (9.11) 6.60 (6.01) 1.46 0.15
Note. Prevention ¼ Childbirth education plus prevention
program; con-

1511e1517 1513
consisted of psycho-education about general anxiety and the
specific anxiety disorders. Participants were provided with a
brief
overview of symptoms, prevalence data, and associated demo-
graphics for the DSM anxiety disorders. Because these presenta-
tions were shorter than those in the prevention condition,
participants were also shown a series of 6 short videos that told
the
perinatal stories of several couples. The prevention condition
was
derived from the cognitive-behavioral model of OCD (Rachman,
1997, 1998) and included the following components: (class 1)
education about postpartum anxiety and OCS (e.g., description
of
symptoms; prevalence; difference between OCS and postpartum
depression or psychosis); (class 2) education about the
cognitive
model of emotion and how OCS fit in this model, with a focus
on the
repercussion of misinterpreting intrusive thoughts; (classes 3 &
4)
instruction in cognitive restructuring of dysfunctional
“obsessive”
beliefs (e.g., identification of beliefs, process of challenging
these
beliefs, and how to modify them into more realistic beliefs);
(class
5) instruction in using behavioral experiments and exposure
techniques as a way of testing and modifying faulty beliefs, and
(class 6) review and wrap-up. Particular emphasis was placed on
addressing beliefs about the importance of, and need to control,
intrusive thoughts about the infant. All of the class modules
were
manualized (Timpano and Abramowitz, 2011),1 and all classes
were

video-taped and 25% were reviewed by the principal
investigators
to ensure adherence.
2.3. Assessment of depression, OCD related beliefs, and OCS
Depressive symptoms were assessed using the Edinburgh
Postnatal Depression Sale (EPDS), a reliable self-report scale
that
validly assesses the presence and severity of postnatal
depressive
symptoms (Cox et al., 1987). The EPDS was completed at
baseline,
post-class, and all three postpartum assessments.
The Obsessive Beliefs Questionnaire (OBQ) was used to assess
dysfunctional beliefs associated with OCD and OCS. The OBQ
is
a 44-item self-report instrument and has been found to possess
good validity and reliability (OCCWG, 2005; Tolin et al.,
2003). In
addition to a total score, three factor analytically derived
subscales
are used, including (1) overestimates of responsibility for, and
threat of harm (responsibility/threat), (2) importance and
control of
intrusive thoughts (importance/control), and (3) perfectionism
and
the need for certainty (perfectionism/certainty). Given that the
CBE-P condition specifically targeted beliefs in the importance/
control domain, OBQ analyses were completed using the
subscale
scores. The OBQ was administered at baseline, post-class, and
the 3
and 6 month postpartum assessments.

Baseline OCS were assessed using the Dimensional Obsessive
eCompulsive Scale (DOCS). The DOCS (Abramowitz et al.,
2010a) is
a 20-item reliable and valid self-report scale that measures the
severity of four empirically established dimensions of OCS:
contam-
ination, responsibility for harm and mistakes, unacceptable
thoughts, and symmetry/completeness in clinical and non-
clinical
samples.
The presence and severity of postpartum OCS was examined
using a semi-structured clinical interview, which included the
Postpartum Thoughts and Behaviors Checklist (PTBC) and the
Yale-
Brown ObsessiveeCompulsive Scale (YBOCS). The PTBC
(Abramowitz et al., 2006) is a modified version of the YBOCS
symptom checklist, and consists of 32 common postpartum
intru-
sive thoughts or obsessions (e.g., thoughts about SIDS) and 14
behavioral and mental compulsions (e.g., frequently checking
on
the baby). The symptoms are grouped into 8 categories and
1 A copy of the prevention program manual is available from
the authors.
participants were scored as either endorsing a category or not.
The
severity of these symptoms was then rated using the 10-item
YBOCS (Goodman et al., 1989a,b), which yields a total score,
in
addition to two subscale scores for obsessions and compulsions.
The PTBC and YBOCS were administered at all three
postpartum
assessments by trained evaluators blind to treatment condition.
2.4. Statistical analyses

All analyses were conducted using the SPSS 16.0 software
package. A 2-tailed significance level of 0.05 was chosen a
priori.
Demographic and pregnancy-related variables between groups
were compared using t-tests and chi-square tests. To examine
the
effects of group on continuous outcome variables (OBQ, EPDS,
and
YBOCS) we analyzed between-group differences using analysis
of
variance (ANOVA). For the primary outcome measure
(YBOCS), we
also used analysis of covariance (ANCOVA) to control for
baseline
OCS and depression symptoms. Within-group differences across
the different time-points were examined using paired-sample t-
tests. Differences in the PTBC were assessed using the chi-
square
test. Primary analyses were conducted with completers;
however,
an additional set of intent-to-treat analyses was conducted to
rule-
out potential drop-out effects. Prior to the primary data
analyses,
data screening was performed. This included descriptive
statistics
to check for data-entry errors, and identification of outliers. No
gross violations were detected.
3. Results
3.1. Study population and baseline values
All 71 participants completed the prevention or control
programs;

however, only 58 (prevention ¼ 33 and control ¼ 25) completed
the 1
month postpartum assessment, 50 (prevention ¼ 31 and control
¼ 19)
the 3 month postpartum assessment, and 49 (prevention ¼ 29
and
control ¼ 20) the 6 month postpartum assessment. Comparisons
of
completers and non-completers revealed no differences on any
demographic or baseline psychological factors considered (all
p’s > 0.10). Participants in the two conditions did not differ
from one
another on any demographic variables considered, including
age,
ethnicity, education, and marital status (all p’s > 0.10). Baseline
levels
of psychological variables are summarized in Table 1.
Participants in
the prevention group did not differ from those in the control
group in
pre-intervention levels of obsessive beliefs, depressive
symptoms, or
OCS (all p’s > 0.10).
3.2. Pregnancy and delivery-related factors
In the overall sample, 80% reported that this was their first
pregnancy, and 10% endorsed having had fertility issues. In
total, 72
infants were born, 30 of which were female. The two groups did
not
differ from one another on any delivery-related variables,
trol ¼ Childbirth education plus control; OBQ ¼ Obsessive
beliefs questionnaire;
EPDS ¼ Edinburgh postpartum depression scale; DOCS ¼
Dimensional obsessive

ecompulsive scale.
Table 2
Percent endorsement of PTBC categories across 3 postpartum
assessments.
PTBC category mo pp Prevention Control X2 p value
Suffocation 1 70.6% 76.0% 0.21 0.64
3 71.0% 52.6% 1.72 0.19
6 36.7% 57.1% 2.09 0.15
Violence 1 51.4% 68.0% 1.65 0.20
3 48.4% 36.8% 0.64 0.43
6 26.7% 30.0% 0.07 0.80
Accidents 1 82.4% 80.0% 0.05 0.82
3 74.2% 84.2% 0.69 0.41
6 76.7% 75.0% 0.02 0.89
Losing the baby 1 44.1% 36.0% 0.39 0.53
3 45.2% 47.4% 0.02 0.88
6 56.7% 45.0% 0.65 0.42
Contamination 1 55.9% 64.0% 0.39 0.53
3 48.4% 57.9% 0.43 0.51
6 60.0% 52.4% 0.29 0.59
Sexual 1 11.8% 12.0% 0.00 0.98
3 9.7% 10.5% 0.01 0.92
6 6.7% 5.0% 0.06 0.81
Illness 1 26.5% 23.1% 0.09 0.76
3 16.1% 15.0% 0.01 0.91

6 20.0% 20.0% 0.00 1.00
Compulsions 1 91.4% 82.1% 1.21 0.27
3 77.4% 85.0% 0.44 0.51
6 83.3% 90.0% 0.44 0.51
Note: PTBC ¼ Postpartum thoughts and behaviors checklist; mo
pp ¼ month(s)
postpartum; prevention ¼ Childbirth education plus prevention
program; con-
trol ¼ Childbirth education plus control.
1511e15171514
including: obstetric complications, mode of delivery (c-section
or
vaginal delivery), use of pain management, sex of baby, or
prema-
ture delivery (all p’s > 0.10). At 1 month postpartum 79% of the
prevention group and 76% of the control group reported breast-
feeding their infants (X2 ¼ 5.08, p < 0.08).
3.3. Change in depressive symptoms and obsessive beliefs
The two groups did not differ significantly at any postpartum
time point on levels of depressive symptoms (all ANOVA p’s >
0.10).
Within-group analyses also revealed that there were no changes
in
depressive symptoms from baseline to 6 months postpartum (all
p’s > 0.10). Similarly, the two groups did not demonstrate any
between or within-group differences in the responsibility/threat
or
perfectionism/certainty subscales of the OBQ. In contrast,
analyses
(ANOVAs) of the importance/control OBQ subscale revealed

group
differences (Fig. 1). Specifically, the prevention group had
signifi-
cantly lower scores at every post-class and postpartum
assessment
than the control group (all p’s < 0.01). Within-group analyses
demonstrated that while the prevention group maintained an
initial
decrease in scores following the classes, the control condition
evi-
denced a modest, yet significant increase following the birth of
the
baby (Fig. 1).
3.4. Assessment of postpartum OCS
Table 2 summarizes the percentage of each group that endorsed
any given PTBC category at each of the three postpartum
assess-
ments. Among both groups, thoughts about accidents (e.g.,
“thoughts about a neighbor’s dog attacking the baby”) were the
most frequently reported types of postpartum intrusive thoughts.
The least frequently endorsed category of intrusive thoughts
was
sexual thoughts (e.g., “a thought about the baby’s genitals”).
Chi-
square analyses revealed that the two groups did not differ from
one another on endorsement rate in any of the categories, at any
of
the three assessment time-points (all p’s > 0.10).
3.5. Assessment of postpartum OCS severity
The prevention group had significantly lower YBOCS total
scores
than the control group at all three postpartum assessments,

controlling for baseline OCS and depression scores (Fig. 2).
Exam-
ination of Cohen’s d revealed a small-medium intervention
effect at
1 month postpartum, and a medium-large intervention effect at 3
and 6 months postpartum (Table 3). At all three time-points, the
control group reported mild, yet clinically significant levels of
OCS
(YBOCS > 8; Goodman et al., 1989a). In contrast, the
prevention
Fig. 1. Mean levels of the importance and control of thoughts
OC beliefs across
assessments.
condition had subclinical levels across the postpartum assess-
ments. Within-group analyses revealed that although the control
group did not show any changes across time, the prevention
group
had significantly lower YBOCS scores at 6 months postpartum
compared to 1 month postpartum (t ¼ 2.77, p < 0.01). A similar
pattern of results emerged when the YBOCS obsessions and
compulsions subscale scores were considered separately (Table
3).
We next conducted a 2 (condition: prevention and control
group) � 4 (time: baseline, 1 month postpartum, 3 month post-
partum, 6 month postpartum) mixed model ANOVA to examine
changes in OCS severity from baseline across the postpartum
period. Because the YBOCS was not administered at baseline,
we
standardized the baseline DOCS scores and the three YBOCS
post-
partum scores, and used the standardized scores in our analysis.
Findings revealed a significant time � condition interaction
(Fig. 3),
Fig. 2. Mean levels of OCS severity across postpartum
assessments, controlling for

baseline OCS and depression symptoms. Both between and
within-group differences
are depicted.
Table 3
Mean OCS across postpartum assessments and magnitude of
treatment effect.
1 mo pp 3 mo pp 6 mo pp
Mean (SD) d Mean (SD) d Mean (SD) d
YBOCS 0.41 0.68 0.73
Prevention 7.27 (4.87) 5.90 (5.00) 4.45 (4.16)
Control 9.68 (6.74) 10.42 (8.03) 8.65 (6.93)
YBOCS-obs 0.39 0.62 0.70
Prevention 3.29 (2.42) 2.84 (2.27) 2.17 (1.76)
Control 4.36 (3.03) 4.68 (3.46) 4.00 (3.24)
YBOCS-comp 0.31 0.67 0.71
Prevention 4.21 (3.09) 3.06 (3.08) 2.34 (2.64)
Control 5.32 (4.02) 5.74 (4.71) 4.65 (3.80)
Note. mo pp ¼ month(s) postpartum; YBOCS ¼ Yale-brown
scale; obs ¼ obsessions subscale score; comp ¼ compulsions
subscale score; pre-
vention ¼ Childbirth education plus prevention program;
control ¼ Childbirth
education plus control.
1511e1517 1515

F(1, 34) ¼ 4.83, p < 0.01. Simple effect analyses were
conducted to
examine the form of this interaction. ANOVA analyses revealed
that
the groups did not differ at baseline; however, significant group
differences emerged during the postpartum period (Fig. 3).
Specifically, we found that the prevention group endorsed fewer
OCS during the postpartum, compared to the control group.
3.6. Intent-to-treat analysis
We conducted follow-up intent-to-treat analyses using the last
observation carried forward (Mazumdar et al.,1999). The pattern
of
effects was identical to those observed in our completer
analyses.
4. Discussion
The current investigation represents the first example of
a prevention program designed specifically to target postpartum
OCS. Results provide support for the utility of incorporating a
CBT-
based prevention program into a traditional CBE curriculum.
Our
intervention falls in-line with a recent psycho-educational
program
developed by Fisher et al. (2010), which was found helpful in
pre-
venting general mood, anxiety and adjustment disorders in the
postpartum. That program did not target specific anxiety symp-
toms, but the findings speak to the general amenability of the
perinatal population for prevention efforts. Our investigation
suggests that a more targeted prevention program packaged into
a CBE course is both feasible and practical: two important
consid-
erations for any prevention work (Dozois, 2004).
Fig. 3. A 2 (condition) � 4 (time) mixed ANOVA to examine

changes in OCS severity
across assessments.
We used an indicated prevention program design (Feldner et al.,
2004), meaning that our intervention focused on a subsyndromal
risk factor (i.e., obsessive beliefs) for obsessions and
compulsions.
Results demonstrated that relative to a credible control
condition,
our prevention program reduced levels of this risk factor,
specifi-
cally with respect to the primary target; that is, beliefs
regarding
the importance of thoughts and the need to control them. Exami-
nation of our OCS outcome measure supported our hypothesis
that
the prevention program would reduce the degree to which OCS
developed during the postpartum period. Although, as expected,
both groups endorsed comparable levels of the incidence of
common postpartum obsessions and compulsions as measured
by
the PTBC, the two groups differed significantly in the severity
of
these symptoms. The YBOCS severity indices capture time
spent on
symptoms, the distress and impairment associated with them,
and
the ability to control obsessions/compulsions. We can therefore
conclude that despite experiencing postpartum intrusive
thoughts,
participants in the prevention group were not as distressed by
them; perhaps as a result of the differential reduction in
obsessive
beliefs. Finally, we found that these effects were specific to
OCS,
since between-group differences remained significant after
controlling for baseline depression symptoms.

Although some authors have reported marked decreases in OCS
following childbirth (Uguz et al., 2007), this was not the case
for our
control group. While the prevention group evidenced continued
decreases in the severity of OCS, participants in the control
group
reported relatively stable OCS across the 6 month postpartum
period with YBOCS scores consistently within the mild yet
clinically
significant range of symptoms (Goodman et al., 1989a). One
explanation for the difference in findings may be cultural. The
investigation by Uguz et al. (2007) was conducted in Turkey,
and
they hypothesized that the decrease in symptoms might be
attributed to increased levels of social support and decreased
levels
of stressors following the birth of the baby. Other research has
suggested that pregnancy and the postpartum period is
associated
with an adverse impact on the new mother’s physical and
emotional health (Miller and Sollie, 1980; Schytt and
Hildingsson,
2011), and in the United States it is not always typical for
extended family to assist as extensively during the postpartum
period as in other cultures.
Our approach to prevention in the present research focused
exclusively on psychological factors. Yet given that
dysfunctional
cognitions only partially predict postpartum OCS (e.g.,
Abramowitz
et al., 2006), additional work is needed to elucidate other
factors
involved in the development of this problem. Certain biological
factors, for example, have been proposed given that pregnancy

and
the postpartum are marked by fluctuations in hormones (Steiner
et al., 2003), brain morphology (Kim et al., 2010), and certain
genetic polymorphisms (Costas et al., 2010). If biological
markers of
postpartum OCS could be identified, this might allow for better
detection of candidates for prevention programs such as that
examined in our investigation.
The present study used a prospective, randomized, controlled-
trial design, which was meant to stringently evaluate the effects
of our prevention program. One limitation, however, was that
there
was no assessment of the degree to which participants imple-
mented the techniques included in the prevention program. We
are
therefore not able to speak to whether the reduction in
symptoms
stems from the use of cognitive restructuring, exposure
techniques,
or the OCS specific psycho-education. The intervention may
also
have been effective by reducing levels of obsessive beliefs. We
found that although the prevention condition maintained
lowered
OBQ levels, the control condition experienced a slight increase
in
obsessive beliefs following the birth of the baby. Future
research
should examine specific mechanisms by which change may have
1511e15171516
occurred. Another limitation is that symptoms of depression

were
ascertained by self-report. A closer inspection of these
symptoms is
warranted, given the overlap between OCS and depression
symp-
toms. Multiple reports have pointed to the possibility that these
two constructs may represent a spectrum of perinatal psychiatric
symptoms (Abramowitz et al., 2010b; Austin et al., 2007; Miller
et al., 2006). In fact, OCD was the only anxiety disorder
predictive
of false-positives for depression in one recent study (Mauri et
al.,
2010). A final limitation is that we did not assess factors
relevant
to social support and the relationship between parents. An
inves-
tigation by Iles et al. (2011) demonstrated that both perceived
support and attachment patterns play a role in posttraumatic
stress
symptoms and depression during the postpartum period. The
same
processes may be important in the generation of OCS,
particularly
as family members are often implicated in the completion of
compulsions or are impacted by excessive reassurance seeking
on
the part of the affected relative (Albert et al., 2010).
Given the significant personal and societal burden associated
with OCD (Lopez and Murray, 1998), and in particular,
perinatal
OCD (Fairbrother and Abramowitz, 2007), it is striking that
very
little attention has been given to the prevention of these
symptoms.
The present investigation represents a seminal step in our

efforts to
enhance traditional therapies and develop new avenues to
prevent
OCD. The identification of vulnerable mothers during the
perinatal
period offers a unique opportunity to directly target specific
and,
importantly, modifiable risk factors. Future research should
consider additional risk factors that could be targeted, as well
as the
beneficial effects prevention programs may have on the global
functioning of mother and child. The prevention program
described
in the current report is relatively inexpensive and viable, and
lowering OCS using a cognitive and behavioral intervention
may
reduce the need for psychotropic medications during the
perinatal
period (Oberlander et al., 2006). Future research may consider
more stream-lined methods for identifying at-risk mothers, and
also the possibility of training CBE instructors to deliver the
CBT
curriculum.
Conflict of interest
Authors Timpano, Schmidt, Mitchell, Abramowitz, & Mahaffey
have no conflicts of interest.
Role of funding source
Funding for this study was provided by a grant from the Inter-
national Obsessive Compulsive Disorder Foundation (IOCDF)
to K.R.
Timpano (PI). The IOCDF had no further role in study design,
data
collection, analysis and interpretation of data, in the writing of

the
report, and in the decision to submit the paper for publication.
Contributors
Authors Timpano, Abramowitz, and Schmidt designed the study
and wrote the protocol. Authors Mahaffey and Mitchell helped
with
data collection efforts and acted as project coordinators. Author
Timpano wrote the first draft of the manuscript. All authors
contributed to and have approved the final manuscript.
Acknowledgments
We thank Heidi Chavers and Elissa Brody, who delivered the
childbirth education programs at the FSU and UNC sites,
respectively.
References
Abramowitz JS, Deacon BJ, Olatunji BO, Wheaton MG, Berman
NC, Losardo D, et al.
Assessment of obsessiveecompulsive symptom dimensions:
development and
evaluation of the dimensional obsessiveecompulsive scale.
Psychological
Assessment 2010a;22:180e98.
Abramowitz JS, Khandker M, Nelson CA, Deacon BJ, Rygwall
R. The role of cognitive
factors in the pathogenesis of obsessiveecompulsive symptoms:
a prospective
study. Behaviour Research and Therapy 2006;44:1361e74.
Abramowitz JS, Meltzer-Brody S, Leserman J, Killenberg S,
Rinaldi K, Mahaffey BL,
et al. Obsessional thoughts and compulsive behaviors in a
sample of women
with postpartum mood symptoms. Archives of Women’s Mental

Health 2010b;
13:523e30.
Abramowitz JS, Nelson CA, Rygwall R, Khandker M. The
cognitive mediation of
obsessiveecompulsive symptoms: a longitudinal study. Journal
of Anxiety
Disorders 2007;21:91e104.
Abramowitz JS, Schwartz SA, Moore KM. Obsessional thoughts
in postpartum
females and their partners: content, severity and relationship
with depression.
Journal of Clinical Psychology in Medical Settings
2003a;10:157e64.
Abramowitz JS, Schwartz SA, Moore KM, Luenzmann KR.
Obsessiveecompulsive
symptoms in pregnancy and the puerperium: a review of the
literature. Journal
of Anxiety Disorders 2003b;17:461e78.
Ahmad TFE, Suleiman R, Shougah D. Mixed anxiety depression
in pregnancy: (the
influence of misconceptions). Arab Journal of Psychiatry
1994;5:103e13.
Albert U, Bogetto F, Maina G, Saracco P, Brunatto C, Mataix-
Cols D. Family
accommodation in obsessiveecompulsive disorder: relation to
symptom
dimensions, clinical and family characteristics. Psychiatry
Research 2010;
179:204e11.
American Psychiatric Association. Diagnostic and statistical

manual of mental
disorder. 4th ed.. Washington, DC: American Psychiatric Press,
Inc.; 1994.
Austin M, Tully L, Parker G. Examining the relationship
between antenatal anxiety
and postnatal depression. Journal of Affective Disorders
2007;101:169e74.
Buttolph ML, Hollander AD. Obsessiveecompulsive disorder in
pregnancy and
childbirth. In: Jenike MA, Baer L, Minichiello WE, editors.
Obsessi-
veecompulsive disorder: theory and management. Chicago: Year
Book Medical
Publishers; 1990.
Chaudron LH, Nirodi N. The obsessiveecompulsive spectrum in
the perinatal
period: a prospective pilot study. Archives of Women’s Mental
Health 2010;13:
403e10.
Coplan RJ, O’Neil K, Arbeau KA. Maternal anxiety during and
after pregnancy and
infant temperament at three months of age. Journal of Prenatal
& Perinatal
Psychology & Health 2005;19:199e215.
Costas J, Gratacòs M, Escaramís G, Martín-Santos R, de Diego
Y, Baca-García E, et al.
Association study of 44 candidate genes with depressive and
anxiety symptoms
in post-partum women. Journal of Psychiatric Research
2010;44:717e24.

Cox JL, Holden JM, Sagovsky R. Detection of postnatal
depression. Development of
the 10-item Edinburgh postnatal depression scale. British
Journal of Psychiatry
1987;150:782e6.
Dozois DJ. Prevention of anxiety psychopathology: conceptual,
methodological, and
practical issues. Clinical Psychology: Science and Practice
2004;11:425e9.
Dupont RL. “Town Hall” panel presentation. The anxiety
disorders of America
conference. Charleston, SC; 1993.
Fairbrother N, Abramowitz JS. New parenthood as a risk factor
for the development
of obsessional problems. Behaviour Research and Therapy
2007;45:2155e63.
Feldner MT, Zvolensky MJ, Schmidt NB. Prevention of anxiety
psychopathology:
a critical review of the empirical literature. Clinical
Psychology: Science and
Practice 2004;11:405e24.
Field T, Hernandez-Reif M, Vera Y, Gil K, Diego M, Bendell
D, et al. Anxiety and anger
effects on depressed mothereinfant spontaneous and imitative
interactions.
Infant Behavior & Development 2005;28:1e9.
Fisher JRW, Wynter KH, Rowe HJ. Innovative psycho-
educational program to
prevent common postpartum mental disorders in primiparous
women:

a before and after controlled study. BMC Public Health
2010;10:432.
Foa EB, Franklin ME, Kozak MJ. Psychosocial treatments for
disorder: literature review. In: Swinson RP, Antony MM,
Rachman S,
Richter MA, editors. Obsessiveecompulsive disorder: theory,
research and
treatment. New York, NY: Guilford Press; 1998.
Forray A, Focseneanu M, Pittman B, McDougle CJ, Epperson
CN. Onset and exac-
erbation of obsessiveecompulsive disorder in pregnancy and the
postpartum
period. The Journal of Clinical Psychiatry 2010;71:1061e8.
Godfrey JR. Toward optimal health: Donna E, Stewart, M.D.
discusses perinatal
depression. Journal of Women’s Health 2005;14:803e7.
Goodman WK, Price LH, Rasmussen SA, Mazure C. The Yale-
Brown obsessive
compulsive scale: I. Development, use, and reliability. Archives
of General
Psychiatry 1989a;46:1006e11.
Goodman WK, Price LH, Rasmussen SA, Mazure C, Delgado P,
Heninger GR, et al. The
Yale-Brown obsessive compulsive scale. II. Validity. Archives
of General
Psychiatry 1989b;46:1012e6.
Heron J, O’Connor TG, Evans J, Golding J, Glover V. The
course of anxiety and
depression through pregnancy and the postpartum in a

community sample.
Journal of Affective Disorders 2004;80:65e73.
Iles J, Slade P, Spiby H. Posttraumatic stress symptoms and
postpartum depression
in couples after childbirth: the role of partner support and
attachment. Journal
of Anxiety Disorders; 2011.
1511e1517 1517
Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR,
Walters EE. Lifetime prevalence
and age-of-onset distributions of DSM-IV disorders in the
national comorbidity
survey replication. Archives of General Psychiatry
2005;62:593e602.
Kim P, Leckman JF, Mayes LC, Feldman R, Wang X, Swain JE.
The plasticity of human
maternal brain: longitudinal changes in brain anatomy during
the early post-
partum period. Behavioral Neuroscience 2010;124:695e700.
Kraemer HC, Stice E, Kazdin A, Offord D, Kupfer D. How do
risk factors work
together? Mediators, moderators, and independent, overlapping,
and proxy risk
factors. American Journal of Psychiatry 2001;158:848e56.
Labad J, Menchon JM, Alonso P, Segalas C, Jimenez S, Vallejo
J. Female reproductive
cycle and obsessiveecompulsive disorder. Journal of Clinical
Psychiatry 2005;

66:428e35. quiz 546.
Larsen KE, Schwarts SA, Whiteside SP, Khandker M, Moore
KM, Abramowitz JS.
Thought control strategies used by parents reporting postpartum
obsessions.
Journal of Cognitive Psychotherapy: An International Quarterly
2006;20:435e45.
Logsdon MC, Wisner K, Billings DM, Shanaham B. Raising the
awareness of primary
care providers about postpartum depression. Issues in Mental
Health Nursing
2006;27:59e73.
Lopez AD, Murray CC. The global burden of disease,
1990e2020. Nature Medicine
1998;4:1241e3.
Lou HC, Hansen D, Nordentoft M, Pryds O, Jensen F, Nim J, et
al. Prenatal stressors of
human life affect fetal brain development. Developmental
Medicine and Child
Neurology 1994;36:826e32.
Manassis K, Bradley S, Goldberg S, Hood J, Swinson RP.
Behavioural inhibition,
attachment and anxiety in children of mothers with anxiety
disorders. Cana-
dian Journal of Psychiatry 1995;40:87e92.
Mauri M, Oppo A, Montagnani MS, Borri C, Banti S, Camilleri
V, et al. Beyond
“postpartum depressions”: specific anxiety diagnoses during
pregnancy predict
different outcomes. Results from PND-ReScU. Journal of

Affective Disorders
2010;127:177e84.
Mazumdar S, Liu KS, Houck PR, Reynolds CF. Intent-to-treat
analysis for longitudinal
clinical trials: coping with the challenge of missing values.
Journal of Psychiatric
Research 1999;33:87e95.
McLean PD, Whittal ML, Thordarson DS, Taylor S, Sochting I,
Koch WJ, et al. Cognitive
versus behavior therapy in the group treatment of
disorder. Journal of Consulting and Clinical Psychology
2001;69:205e14.
Miller BC, Sollie DL. Normal stresses during the transition to
parenthood. Family
Relations 1980;29:459e65.
Miller RL, Pallant JF, Negri LM. Anxiety and stress in the
postpartum: is there more
to postnatal distress than depression? BMC Psychiatry
2006;6:12.
Mulder EJ, Robles de Medina PG, Huizink AC, Van den Bergh
BR, Buitelaar JK,
Visser GH. Prenatal maternal stress: effects on pregnancy and
the (unborn)
child. Early Human Development 2002;70:3e14.
Oberlander TF, Warburton W, Misri S, Aghajanian J, Hertzman
C. Neonatal outcomes
after prenatal exposure to selective serotonin reuptake inhibitor
antidepres-
sants and maternal depression using population-based linked
health data.

Archives of General Psychiatry 2006;63:898e906.
Obsessive Compulsive Cognitions Working Group.
Psychometric validation of the
obsessive belief questionnaire and interpretation of intrusions
inventory e Part
2: Factor analyses and testing of a brief version. Behaviour
Research and
Therapy 2005;43:1527e42.
Rachman S. A cognitive theory of obsessions. Behaviour
Research and Therapy
1997;35:793e802.
Rachman S. A cognitive theory of obsessions: elaborations.
Behaviour Research and
Therapy 1998;36:385e401.
Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk
factors for post-
partum depression: a synthesis of recent literature. General
Hospital Psychiatry
2004;26:289e95.
Ross LE, McLean LM. Anxiety disorders during pregnancy and
the postpartum
period: a systematic review. Journal of Clinical Psychiatry
2006;67:1285e98.
Salkovskis PM, Forrester E, Richards C. Cognitive-behavioural
approach to under-
standing obsessional thinking. British Journal of Psychiatry
Supplement;
1998:53e63.
Schytt E, Hildingsson I. Physical and emotional self-rated

health among Swedish
women and men during pregnancy and the first year of
parenthood. Sexual &
Reproductive Healthcare 2011;2:57e64.
Steiner M, Dunn E, Born L. Hormones and mood: from
menarche to menopause and
beyond. Journal of Affective Disorders 2003;74:67e83.
Sutter-Dallay AL, Giaconne-Marcesche V, Glatigny-Dallay E,
Verdoux H. Women
with anxiety disorders during pregnancy are at increased risk of
intense
postnatal depressive symptoms: a prospective survey of the
MATQUID cohort.
European Psychiatry 2004;19:459e63.
Timpano KR, Abramowitz JS. BabyPREP: a postpartum OCS
prevention program.
Unpublished manuscript; 2011.
Tolin DF, Woods C, Abramowitz JS. Relationship between
obsessional beliefs and
obsessiveecompulsive symptoms. Cognitive Therapy and
Research 2003;27:
657e69.
Uguz F, Gezginc K, Zeytinci I, Karatayli S, Askin R, Guler O,
et al. Course of obses-
siveecompulsive disorder during early postpartum period: a
prospective
analysis of 16 cases. Comprehensive Psychiatry
2007;48:558e61.
Williams KE, Koran LM. Obsessiveecompulsive disorder in
pregnancy, the puer-

perium, and the premenstruum. Journal of Clinical Psychiatry
1997;58:330e4.
quiz 335e6.
Zambaldi CF, Cantilino A, Montenegro AC, Paes JA, de
Albuquerque TLC, Sougey EB.
Postpartum obsessiveecompulsive disorder: prevalence and
clinical charac-
teristics. Comprehensive Psychiatry 2009;50:503e9.
Efficacy of a prevention program for postpartum obsessive–
compulsive symptoms1 Introduction2 Methods2.1
Participants2.2 Procedures and prevention program2.3
Assessment of depression, OCD related beliefs, and OCS2.4
Statistical analyses3 Results3.1 Study population and baseline
values3.2 Pregnancy and delivery-related factors3.3 Change in
depressive symptoms and obsessive beliefs3.4 Assessment of
postpartum OCS3.5 Assessment of postpartum OCS severity3.6
Intent-to-treat analysis4 Discussion Conflict of interest Role of
funding source Contributors Acknowledgments References
Brief Report
Beating Bipolar: exploratory trial of a novel
internet-based psychoeducational treatment
for bipolar disorder
Psychoeducation has emerged recently as an
effective intervention for preventing relapse in the
long-term management of bipolar disorder. The
goals of psychoeducation include providing
patients and their families with accurate and reliable
information about the diagnosis, causes and treat-
ment of bipolar disorder, as well as teaching

patients a range of self-management skills, such
as effective mood monitoring, early relapse recog-
nition and pragmatic relapse prevention strategies
(1). Many of the trials of psychoeducational
Smith DJ, Griffiths E, Poole R, di Florio A, Barnes E, Kelly
MJ,
Craddock N, Hood K, Simpson S. Beating Bipolar: exploratory
trial of a
novel internet-based psychoeducational treatment for bipolar
disorder.
Bipolar Disord 2011: 13: 571–577. ª 2011 The Authors.
Journal compilation ª 2011 John Wiley & Sons A ⁄ S.
Objectives: Psychoeducational approaches are promising
interventions
for the long-term management of bipolar disorder. In
consultation with
professionals, patients, and their families we have developed a
novel
web-based psychoeducational intervention for bipolar disorder
called
Beating Bipolar. We undertook a preliminary exploratory
randomized
trial to examine efficacy, feasibility and acceptability.
Methods: This was an exploratory randomized controlled trial of
Beating Bipolar (current controlled trials registration number:
ISRCTN81375447). The control arm was treatment-as-usual and
the a
priori primary outcome measure was quality of life [measured
by the brief
World Health Organization Quality of Life (WHOQOL–BREF)
scale].
Secondary outcomes included psychosocial functioning, insight,
depressive and manic symptoms and relapse, and use of

healthcare
resources. Fifty participants were randomized to either the
Beating
Bipolar intervention plus treatment-as-usual or just treatment-
as-usual.
The intervention was delivered over a four-month period and
outcomes
were assessed six months later.
Results: There was no significant difference between the
intervention
and control groups on the primary outcome measure (total
WHOQOL–
BREF score) but there was a modest improvement within the
psychological subsection of the WHOQOL–BREF for the
intervention
group relative to the control group. There were no significant
differences
between the groups on any of the secondary outcome measures.
Conclusions: Beating Bipolar is potentially a safe and engaging
intervention which can be delivered remotely to large numbers
of
patients with bipolar disorder at relatively low cost. It may have
a
modest effect on psychological quality of life. Further work is
required to
establish the impact of this intervention on insight, knowledge,
treatment
adherence, self-efficacy and self-management skills.
Daniel J Smitha, Emily Griffithsa, Ria
Poolea, Arianna di Florioa, Emma
Barnesa, Mark J Kellyb, Nick
Craddocka, Kerenza Hoodb

and Sharon Simpsonb
aDepartment of Psychological Medicine and
Neurology, Cardiff University School of Medicine,
Cardiff, bSouth East Wales Trials Unit, Department
of Primary Care and Public Health, School of
Medicine, Cardiff University, Heath Park, Cardiff,
UK
doi: 10.1111/j.1399-5618.2011.00949.x
Key words: bipolar disorder – clinical trial –
internet – patient education – quality of life
Received 25 November 2010, revised and
accepted for publication 2 August 2011
Corresponding author:
Dr. Daniel J. Smith
Department of Psychological Medicine and
Neurology
Cardiff University School of Medicine
Monmouth House, Heath Park

Cardiff CF14 4DW, UK
Fax: 44-029-2074-3143
E-mail: [email protected]
The authors of this paper do not have any commercial
associations
that might pose a conflict of interest in connection with this
manu-
script.
Bipolar Disorders 2011: 13: 571–577 ª 2011 John Wiley and
Sons A/S
BIPOLAR DISORDERS
571
interventions carried out so far have focused on
symptom reduction and relapse prevention as
primary outcomes. There has been very little
systematic assessment of the impact of psychoed-
ucation on quality of life, psychosocial functioning,
treatment adherence, personal insight and self-
management skills.
Psychoeducational treatments can be delivered
in a number of formats, including individually (2),
in groups (3), to patients� families, and ⁄ or carers
(4, 5), and as part of systematic care programmes
(6). To date, there is evidence that group-based
psychoeducation is effective for preventing relapse
in some (7, 8) but not all (4) clinical trials. Potential

disadvantages of group approaches include the
considerable cost in terms of therapist time, the
likelihood that some patients will not be comfort-
able in group settings and the need for patients to
travel to group sessions. We therefore set out to
develop and test a novel internet-based psychoed-
ucational treatment for bipolar disorder, called
Beating Bipolar, which translates the content of
group psychoeducation and, to some degree, the
experience of peer-group support, for use on the
internet. Based on feedback from individuals with
bipolar disorder, their families and mental health
professionals, we decided that our primary focus in
terms of outcome should be the degree to which
this new treatment improved quality of life (9). The
development process for Beating Bipolar and the
protocol for its evaluation have been described in
detail elsewhere (9, 10). Here we report the findings
from a phase II exploratory trial of the Beating
Bipolar intervention.
Patients and methods
This was a phase II randomized controlled trial
carried out between March 2009 and September
2010 (current controlled trials registration number:
ISRCTN81375447) (10). It was approved by the
South East Wales NHS Research Ethics Commit-
tee. In total, 80 participants with bipolar disorder
were screened for inclusion in this study and 50
were randomized to either the Beating Bipolar
intervention plus treatment-as-usual (TAU) or just
TAU. Outcome was assessed six months after the
end of the intervention.
Inclusion and exclusion criteria

The main inclusion criteria for this study were age
between 18 and 65 and a diagnosis of DSM-IV
bipolar disorder [including type I and type II ⁄ not
otherwise specified (NOS)] currently in clinical
remission. Diagnosis was confirmed using the Mini
International Neuropsychiatric Interview (MINI)
(11) and clinical remission was defined as not
fulfilling diagnostic criteria for a depressive, manic
or mixed affective episode during the preceding
three-month period, plus a current Montgomery–
Åsberg Depression Rating Scale (MADRS) (12)
score of £ 10 and a Young Mania Rating Scale
(YMRS) (13) score of £ 8. These MADRS and
YMRS threshold scores are widely accepted cor-
relates of symptomatic remission in bipolar dis-
order. The exclusion criterion was an inability to
engage fully in the psychoeducational programme,
for example, because of cognitive impairment or
not having English language ability of sufficient
level. Given that this was an exploratory trial, no
other exclusion criteria were specified.
Recruitment
Recruitment was from multiple sources across
South Wales, including Primary Care Practices
and Community Mental Health Teams (CMHTs).
Potential participants in this study were identified
from case files by Clinical Studies Officers from the
Mental Health Research Network Cymru (Wales)
and invitations to take part were sent to patients
on behalf of the research team from general
practitioners and consultant psychiatrists. Mem-
bers of the Manic Depression Fellowship in Wales

were also invited to take part by their local group
co-ordinators. Written informed consent was
obtained from all participants.
Intervention
Beating Bipolar was developed in a three-stage
process (9). In stage 1, a literature search was
conducted to identify data and other information
which could help to inform the design and content of
a web-based psychoeducational programme for
bipolar disorder (1). This included searching for
information related to the design of e-learning
environments, learner engagement, interactivity,
presentation and instructional design. In stage 2, a
multidisciplinary team including a psychiatrist, two
psychologists (one with expertise in designing online
learning programmes), and an educational web
designer convened to draft ideas for content, deliv-
ery and interactivity based on the available literature
and professional experience. In stage 3 we convened
three focus groups with the goal of iteratively
developing the content and format of the interven-
tion. The draft ideas from stage 2 acted as an initial
stimulus for discussions. The focus groups were
made up of a combination of service users, carers
and mental health professionals in order to achieve a
Smith et al.
572
balance between users� subjective needs and recom-
mendations based on clinical experience. The final-

ized Beating Bipolar intervention involves a
blending of different delivery mechanisms, with
initial face-to-face delivery, followed by written and
web-based interactive delivery of factual content
and ongoing support via a web forum (9).
The key areas covered in the package are: (i) the
accurate diagnosis of bipolar disorder; (ii) the
causes of bipolar disorder; (iii) the role of medica-
tion; (iv) the role of lifestyle changes; (v) relapse
prevention and early intervention; (vi) psycholog-
ical approaches; (vii) gender-specific consider-
ations, and (viii) advice for family and carers.
This content has a similar focus to the content
within Phase I of Bauer and McBride�s Life Goals
Program (which also includes information on the
nature of bipolar disorder, triggers and early
symptoms of relapse, and self-management strat-
egies for relapse) (14). It also has some similarities
with Colom and Vieta�s group psychoeducation
intervention for bipolar disorder (15), although in
greatly abbreviated form given that the Colom and
Vieta intervention comprises 21 sessions. The
programme was not designed in such a way as to
be able to address individual differences (for
example, patients who tend to experience more
depression than mania or vice versa).
The eight modules were delivered online on a
fortnightly basis over a four-month period. There
was an initial face-to-face introductory meeting led
by a consultant psychiatrist (DJS) to demonstrate
how to use the programme. Thereafter, partici-
pants logged onto the website and completed the
modules. Within each of the modules there was
approximately a 50:50 mix of didactic video-based

delivery of information and interactive exercises
for participants to complete (for example, com-
pleting an online life-chart). In order for partici-
pants to progress through each of the modules it
was necessary for them to complete each of the
subsections in turn. Throughout the trial there was
an opportunity for participants in the intervention
group to discuss the content of the material with
each other within a secure discussion forum
moderated by DJS. A reminder email inviting
participants to access the content was sent by the
moderator (DJS) one week before each of the eight
modules was made available.
Participants who were randomized to the inter-
vention also continued to receive TAU whereas
those not randomized to the intervention received
just TAU. TAU for all participants comprised
usual care delivered in a collaborative model
between general practitioners and local multidisci-
plinary community mental health teams.
Randomization
Using dynamic block allocation, participants were
randomized remotely using computer-generated
number lists (16). The balanced variables were
age, gender and bipolar disorder subtype.
Outcome measures
Outcomewas assessedsix monthsaftertheendof the
intervention by two members of the research team
(RP and AdF) who were blinded as to whether
participantshadreceivedtheinterventionornot.The
primary outcome was improvement in quality of life

as measured by the World Health Organization
Quality of Life, Brief version (WHOQOL–BREF)
questionnaire (17). Quality of life was chosen as the
primary outcome measure in response to feedback
from individuals with bipolar disorder, their families
and mental health professionals (9). These groups
identified a need for new psychoeducational inter-
ventions to impact on broad areas of everyday
functioning (particularly quality of life) which go
beyond a focus on symptoms of depression and ⁄ or
mania or episodes of illness. The WHOQOL–BREF
is comprised of scores within four separate domains
(physical health, psychological health, social rela-
tionships and environment). It is a reliable, valid and
widely used measure of quality of life in psychiatric
out-patient settings (18). When this study began
there was no specific quality of life measure for use in
bipolar disorder but the WHOQOL has been recom-
mended as suitable for use in this population (19).
Secondary outcomes included the Global Assess-
ment of Functioning (GAF) scale (20), the Func-
tioning Assessment Short Test (FAST) (21), and
insight measured using a modified Schedule for
Assessment of Insight (SAI) (22). Current depres-
sive symptoms according to the MADRS (23) and
current manic symptoms according to the YMRS
(13) were also compared between the two groups.
Using the MINI (11), the number and severity of
depressive and manic symptoms and number and
timing of episodes of depression and mania or
hypomania experienced during the 10-month
period since the beginning of the trial were
compared between groups.
Statistical analyses

The primary analysis was an intention-to-treat
analysis comparing the WHOQOL–BREF scores
between the intervention and control groups while
controlling for baseline WHOQOL–BREF scores
using analysis of covariance (ANCOVA). Second-
ary outcome analyses were performed similarly,
Internet-based psychoeducation for bipolar disorder
573
also controlling for baseline scores. Categorical
and non-normally distributed data were analysed
using the chi-squared and Mann–Whitney U tests,
respectively, and the Cohen�s d effect size was
calculated for the intervention group (24). An
exploratory analysis excluding those who did not
comply with the intervention was also carried out.
Results
Eighty potential participants were assessed for
inclusion but only 50 satisfied the inclusion and
exclusion criteria (Fig. 1). Randomization resulted
in 24 participants in the intervention group and 26
in the control TAU group. Full outcome data were
available on 17 participants from the intervention
arm and 20 from the control arm. Baseline charac-
teristics of trial participants are detailed in Table 1.
Randomized participants were well matched in
terms of baseline sociodemographic and clinical
characteristics and there were no differences be-
tween groups on current medication use (Table 1).

Primary outcome measure
In terms of changes from baseline on the total
WHOQOL–BREF score, there was no significant
difference between the intervention and control
groups (Table 2). Although there were no differ-
ences between groups on the physical, social
relationships, and environment subsections of the
WHOQOL–BREF, within the psychological sub-
section, there was a marginally significant differ-
ence without correction for multiple testing: an
increase of 8.1 units from 52.7 at baseline to 60.8 at
follow-up within the intervention group compared
to a decrease of 5.0 units from 61.9 at baseline to
56.9 at outcome within the control group (p =
0.05; 95% confidence interval 0.24 to 22.6). This
represents a medium level (Cohen�s d) effect size for
the treatment group of 0.43. The psychological
quality of life subsection of the WHOQOL–BREF
assesses several areas, including: body image ⁄
appearance; negative feelings; positive feelings;
self-esteem; spirituality ⁄ religion ⁄ personal beliefs and
thinking; and learning, memory and concentration.
Secondary outcome measures
There were no significant differences between
groups on any of the secondary outcome measures
(Table 2).
Compliance with the Beating Bipolar intervention
Figure 2 illustrates the use of the Beating Bipolar
programme by each of the 24 participants within

the intervention group. We were able to collect
data on whether participants had completed
(rather than simply accessed) the subsections
within each of the eight modules. All but one of
the modules had 6 subsections (the remaining
module had 5) so that the total number of
subsections available for completion was 47.
Compliance with the intervention is defined as
the proportion of these 47 subsections completed
by participants (Fig. 2). Although three partici-
pants did not access any of the programme during
the trial, 16 ⁄ 24 (66.6%) completed at least 75% of
the programme. Usage of the discussion forum was
variable. Only 13 ⁄ 24 (54.2%) of participants
posted at least one message on the board and four
individuals accounted for 92 out of a total of 127
messages posted (72.4%). Having said this, many
of the messages posted were very positive regarding
the content and format of Beating Bipolar and in
general participants posted comments to each
other which were supportive and constructive.
As an exploratory analysis, we excluded the
three participants in the intervention arm who did
not access any of the programme (and therefore
did not receive the intervention) and re-analysed
the outcome data. This analysis did not identify
any differences between the groups in primary or
secondary outcomes.Fig. 1. CONSORT diagram.
Smith et al.
574

Discussion
To our knowledge, this is the first exploratory
randomized controlled trial of a web-based psy-
choeducational treatment for bipolar disorder. The
primary outcome measure (total WHOQOL–
BREF score) was not significantly different
between the intervention and control groups. There
was a marginally statistically significant improve-
ment (p = 0.05) within the psychological quality
of life subsection of the WHOQOL–BREF. The
intervention group increased by 8.1 points whereas
the control group decreased by 5.0 points, with a
Cohen�s d effect size within the medium range at
0.43. It should be noted that this finding is not
statistically significant when corrected for multiple
testing. On balance, it is possible that the Beating
Bipolar intervention has a modest positive impact
on the quality of life of individuals with bipolar
disorder, particularly within the domain of psy-
chological quality of life.
Although there were no statistically significant
improvements between groups on the secondary
outcome measures, it is likely that the small sample
size (n = 50) and the relatively brief follow-up
period of six months make it difficult to draw any
firm conclusions about the likely impact of the
intervention on these domains, which included rates
of relapse into depression, hypomania and mania,
measures of psychosocial functioning, insight, and
contact with health and social care services.
Over two-thirds of those in the intervention
group completed at least 75% of the programme,

suggesting that this was an acceptable and engag-
ing experience for the majority of participants,
although only half of participants posted messages
on the discussion forum.
Strengths and limitations
This was a small study of 50 randomized patients
with bipolar disorder. The majority of participants
had bipolar I disorder (n = 43, 86%) (Table 1).
Possible limitations include a very high proportion
of participants of Caucasian ethnicity (98%) and a
relatively high proportion of participants who were
currently in employment (60%). As a phase II
exploratory trial, strengths of this study include a
clear a priori protocol for conducting the study and
analysing the findings (10) and the recruitment of
participants from real-world NHS settings in the
UK. Ideally, a longer follow-up period than six
Table 1. Baseline characteristics
Treatment group (n = 24) Control group (n = 26) p-value
Diagnosis, n (%)
BP-I 19 (79.2) 24 (92.3) 0.34a
BP-II 4 (16.7) 2 (7.7)
BP-NOS 1 (4.2) 0 (0)
Age, years, mean (SD) 42.7 (11.4) 44.7 (9.9) 0.51b
Males, n (%) 11 (45.8) 8 (30.8) 0.27a
Caucasian ethnicity, n (%) 23 (95.8) 26 (100) 0.29a

Educated to A-level or above, n (%) 12 (50.0) 12 (46.2) 0.79a
Currently employed, n (%) 14 (58.3) 16 (61.5) 0.82a
Current medication, n (%)
Antidepressant 13 (54.2) 10 (38.5) 0.27a
Mood stabilizer 16 (66.7) 20 (76.9) 0.42a
Antipsychotic 10 (41.7) 10 (38.5) 0.82a
Other 11 (47.8) 13 (52.0) 0.77a
MADRS score, mean (SD) 4.0 (3.0) 3.5 (2.8) 0.51b
YMRS score, mean (SD) 1.4 (2.0) 0.9 (1.7) 0.39b
WHOQOL–BREF total, mean (SD) 228.7 (75.0) 254.7 (81.3)
0.25b
Physical 56.1 (19.2) 59.5 (16.3) 0.50b
Psychological 52.7 (22.9) 61.9 (21.6) 0.15b
Social relationships 53.6 (27.8) 61.5 (31.1) 0.35b
Environment 66.3 (19.6) 71.9 (23.1) 0.37b
GAF score, mean (SD) 68.3 (13.3) 73.4 (15.1) 0.21b
FAST score, mean (SD) 27.5 (15.0) 21.0 (16.3) 0.15b
SAI score, mean (SD) 5.6 (2.2) 5.2 (1.7) 0.44b
BP-I = bipolar I disorder; BP-II = bipolar II disorder; BP-NOS
= bipolar disorder not otherwise specified; MADRS =

Montgomery–Åsberg
Depression Rating Scale; YMRS = Young Mania Rating Scale;
WHOQOL–BREF = World Health Organization Quality of Life,
Brief version;
GAF = Global Assessment of Functioning; FAST = Functioning
Assessment Short Test; SAI = Schedule for Assessment of
Insight; SD =
standard deviation.
aChi-squared test.
bt-test.
575
months would have been preferable in order to
assess any sustained benefit in the medium to
longer term. Given that this was an exploratory
study we did not correct for multiple testing.
Clinical implications
Our findings are in keeping with a growing body
of work which suggests that psychoeducational
approaches are acceptable and effective interven-
tions in the long-term management of bipolar
disorder (1). This study suggests that Beating
Bipolar is potentially a safe and engaging inter-
vention which can be delivered remotely to large
numbers of patients with bipolar disorder and
which may have a modest effect on psychological
quality of life (including areas such as body
image ⁄ appearance, the experience of negative

and positive emotions, self-esteem, spirituality
and learning ⁄ concentration). It is unclear at pres-
ent which of these areas is most improved by
Beating Bipolar. Further work is required to
Table 2. Primary and secondary outcomes
Treatment group
(n = 17)
Control group
(n = 20) F-statistic Chi-squared Mann–Whitney U p-value
Primary outcome, mean (SD)
WHOQOL–BREF total 256.6 (52.7) 259.2 (63.2) 1.29 – – 0.27
Physical 62.3 (19.0) 62.6 (16.5) 0.36 – – 0.56
Psychological 60.8 (17.8) 56.9 (16.4) 4.31 – – 0.05
Social relationships 55.5 (20.9) 62.7 (25.4) 0.01 – – 0.93
Environment 78.1 (14.9) 77.0 (19.8) 1.75 – – 0.19
Secondary outcomes
MADRS score, mean (SD) 9.1 (8.4) 11.1 (13.6) 0.46 – – 0.50
YMRS score, mean (SD) 2.4 (2.9) 3.9 (7.7) 1.03 – – 0.32
GAF score, mean (SD) 70.8 (14.8) 65.9 (21.8) 0.95 – – 0.34
FAST total score, mean (SD) 22.8 (12.3) 19.4 (13.6) 0.08 – –
0.78
Autonomy 2.0 (2.3) 1.9 (2.6) 0.001 – – 0.97
Occupational functioning 8.2 (4.8) 6.2 (5.8) 0.20 – – 0.66
Cognitive functioning 4.5 (4.0) 4.8 (3.1) 0.44 – – 0.51
Finances 1.5 (1.6) 1.5 (1.9) 0.08 – – 0.78
Relationships 4.4 (4.2) 3.6 (3.6) 0.05 – – 0.82
Leisure 2.2 (1.6) 1.7 (1.5) 0.05 – – 0.83
SAI score, mean (SD) 5.4 (2.1) 5.7 (1.9) 3.07 – – 0.09

Depressive episode during
study period, n (%)
10 (55.6) 9 (45.0) – 0.42 – 0.52
Episodes of depression,
median [range]
1 [0–4] 0.5 [0–4] – – 180.0 1.00
Total number of months with
depression, median [range]
0.25 [0–6] 0.25 [0–10] – – 174.5 0.87
Hypomanic episode during
study period, n (%)
3 (17.6) 6 (30.0) – 0.76 – 0.38
Episodes of hypomania,
median [range]
0 [0–8] 0 [0–5] – – 150.0 0.42
hypomania, median [range]
0 [0–2.5] 0 [0–9] – – 150.5 0.43
Manic episode during study
period, n (%)
4 (22.2) 5 (25.0) – 0.04 – 0.84
Episodes of mania during study

period, median [range]
0 [0–1] 0 [0–10] – – 144.5 0.44
hypomania, median [range]
0 [0–2.5] 0 [0–9] – – 150.5 0.43
Contacts with psychiatric
services, median [range]
3 [0–30] 5 [0–52] – – 150.0 0.54
Contacts with primary care
services, median [range]
6 [0–11] 7 [4–41] – – 132.0 0.25
Contacts with social services,
median [range]
0 [0–12] 0 [0–0] – – 150.0 0.12
Contacts with other services
(e.g., A & E), median [range]
1 [0–10] 0 [0–14] – – 110.0 0.06
WHOQOL–BREF = World Health Organization Quality of Life,
Brief version; MADRS = Montgomery–Åsberg Depression
Rating Scale;
YMRS = Young Mania Rating Scale; GAF = Global Assessment
of Functioning; FAST = Functioning Assessment Short Test;
SAI =
Schedule for Assessment of Insight; SD = standard deviation

Smith et al.
576
establish the impact of this intervention on areas
such as insight, knowledge, self-efficacy and self-
management skills. This is likely to inform the
design of a formal randomized controlled trial and
particularly the most appropriate choice of outcome
measures. The delivery of Beating Bipolar via the
internet represents a potentially cost-effective means
of providing high-quality psychoeducational mate-
rial to large numbers of individuals at relatively low
cost.
Acknowledgements
DJS was funded by a postdoctoral fellowship from the
National Institute of Health Research (UK) and the Welsh
Assembly Government. Development of the Beating Bipolar
programme was partly funded by the Big Lottery Fund�s
Mental Health Matters Programme in Wales.
We would like to thank all of those individuals who
volunteered to take part in this study, the Manic Depression
Fellowship in Wales (MDF-Cymru), and the Healthcare
Learning Company PLC.
References
1. Smith DJ, Jones I, Simpson S. Psychoeducation for bipolar
disorder. Adv Psychiatr Treat 2010; 16: 147–154.

2. Perry A, Tarrier N, Morriss R, McCarthy E, Limb K.
Randomised controlled trial of efficacy of teaching patients
with bipolar disorder to identify early symptoms of relapse
and obtain treatment. BMJ 1999; 318: 149–153.
3. Colom F, Lam D. Psychoeducation: improving outcomes in
bipolar disorder. European Psychiatry 2005; 20: 359–364.
4. Miklowitz DJ, George E, Richards JA, Simoneau TL,
Suddath RL. A randomized study of family-focused
psychoeducation and pharmacotherapy in the outpatient
management of bipolar disorder. Arch Gen Psychiatry
2003; 60: 904–912.
5. Miklowitz DJ, Axelson DA, Birmaher B et al. Family-
focused treatment for adolescents With bipolar disorder:
results of a 2-year randomized trial. Arch Gen Psychiatry
2008; 65: 1053–1061.
6. Simon GE, Ludman EJ, Bauer MS, Unutzer J, Operskalski
B. Long-term effectiveness and cost of a systematic care
program for bipolar disorder. Arch Gen Psychiatry 2006;
63: 500–508.
7. Colom F, Vieta E, Martinez-Aran A et al. A randomized
trial on the efficacy of group psychoeducation in the
prophylaxis of recurrences in bipolar patients whose
disease Is in remission. Arch Gen Psychiatry 2003; 60:
402–407.
8. Colom F, Vieta E, Sanchez-Moreno J et al. Group
psychoeducation for stabilised bipolar disorders: 5-year
outcome of a randomised clinical trial. Br J Psychiatry
2009; 194: 260–265.
9. Barnes E, Hood K, Simpson S, Smith DJ. Development of

an online psychoeducational treatment for bipolar disor-
der. J Ment Health 2011; 20: 21–31.
10. Simpson S, Barnes E, Griffiths E et al. The Bipolar
Interactive Psychoeducation (BIPED) study: trial design
and protocol. BMC Psychiatry 2009; 9: 50.
11. Sheehan DV, Lecrubier Y, Harnett-Sheehan K et al. The
Mini International Neuropsychitric Interview (MINI): the
development and validation of a structured diagnostic
psychiatric interview. J Clin Psychiatry 1998; 59(Suppl.
20): 22–33.
12. Snaith RP, Harrop FM, Newby D.A. Grade scores of the
Montgomery-Asberg depression and the clinical anxiety
scales. Br J Psychiatry 1986; 148: 599–601.
13. Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating
scale for mania: reliability, validity and sensitivity. Br J
Psychiatry 1978; 133: 429–435.
14. Bauer M, McBride L. Structured Group Psychotherapy for
Bipoar Disorder: The Life Goals Program. 2nd ed. New
York: Springer, 2003.
15. Colom F, Vieta E. Psychoeducation Manual for Bipolar
Disorder. Cambridge: Cambridge University Press, 2006.
16. Carter BR, Hood K. Balance algorithm for cluster
randomized trials. BMC Med Res Methodol 2008; 8:
65.
17. WHO. Development of the World Health Organization
WHOQOL-BREF Quality of Life Assessment. Psychol
Med 1998; 28: 551–558.

18. Trompenaars FJ, Masthoff ED, Van Heck GL, Hodia-
mont PP, De Vries J. Content validity, construct validity,
and reliability of the WHOQOL-BREF in a population of
Dutch adult psychiatric outpatients. Qual Life Res 2005;
14: 151–160.
19. Michalak E, Yatham L, Lam R. Quality of life in bipolar
disorder: A review of the literature. Health Qual Life
Outcomes 2005; 3: 72.
20. Spitzer RL, Gibbon M, Williams JBW. Global Assessment
of Functioning (GAF) Scale. In: Sederer LI, Dickey B, eds.
Outcome Assessment in Clinical Practice. Baltimore:
Williams and Wilkins, 1996: 76–77.
21. Rosa A, Sanchez-Moreno J, Martinez-Aran A et al.
Validity and reliability of the Functioning Assessment
Short Test (FAST) in bipolar disorder. Clin Pract Epi-
demiol Ment Health 2007; 3: 5.
22. Kemp R, David A. Insight and compliance. In: Blackwell
B ed. Insight and Compliance. Newark, NJ: Gordon and
Breach Publishing Group, 1996: 61–84.
23. Williams JBW, Kobak KA. Development and reliability of
a structured interview guide for the Montgomery–Asberg
Depression Rating Scale (SIGMA). Br J Psychiatry 2008;
192: 52–58.
24. Cohen J. Statistical Power Analysis for the Behavioral
Sciences (2nd edn). Hillsdale, NJ: Erlbaum, 1988.
Fig. 2. Compliance of participants with the Beating Bipolar
intervention.

577
Copyright of Bipolar Disorders is the property of Wiley-
Blackwell and its content may not be copied or emailed
to multiple sites or posted to a listserv without the copyright
holder's express written permission. However,
users may print, download, or email articles for individual use.
Journal of Traumatic Stress, Vol. 24, No. 5, October 2011, pp.
591–595 ( C© 2011)
B R I E F R E P O R T
A Resilience-Oriented Treatment for Posttraumatic
Stress Disorder: Results of a Preliminary Randomized
Clinical Trial
Martha Kent
Phoenix VA Healthcare System and Banner Behavioral Health
Hospital, Scottsdale
Mary C. Davis and Shannon L. Stark
Arizona State University
Laura A. Stewart
Southwest Behavioral Health Services, Phoenix
This preliminary randomized trial examined the effect of a

resilience-oriented intervention for posttraumatic
stress disorder (PTSD) versus a waitlist control on anxiety and
depressive symptoms, positive emotional health,
and cognitive performance in 39 veterans with a variety of
traumatic exposures. From pre- to posttreatment, the
intervention but not the control group showed improvements
that were large in magnitude for affective symptoms
and positive emotional health (ds = 0.73–1.18), moderate in
magnitude for memory (ds = 0.50–0.54),
and small-to-moderate in magnitude for executive function (ds
= 0.30–0.35). Findings suggest that treatment
explicitly targeting resilience resources (e.g., positive
emotional engagement, social connectedness) may provide
broad benefits, including alleviation of anxiety and depressive
symptoms and improved positive emotional and
cognitive function.
Individuals are motivated to approach rewards and avoid
painful conditions. Attention to both motivational systems is
highly adaptive. In the case of posttraumatic stress disorder
(PTSD), however, individuals are oriented more toward avoid-
ance of threat and attend less to the pursuit of positive, val-
ued experiences (Kashdan, Breen, & Julian, 2010). At a neural
level, the circuitry of fear and avoidance remains hyperactive
long
past trauma exposure and is relatively unmodulated by higher
level cortical activity (Aupperle & Paulus, 2010; Walter,
Palmieri,
& Gunstad, 2010). Not surprisingly, most current psychological
PTSD treatments focus primarily on extinguishing the fear re-
sponse, most typically through exposure (Foa, Keane, Friedman,
& Cohen, 2008).
Martha Kent, Phoenix VA Healthcare System and Banner
Behavioral Health Hospital, Scotts-
dale; Mary C. Davis and Shannon L. Stark, Department of

Psychology, Arizona State University,
Tempe; Laura A. Stewart, Southwest Behavioral Health
Services, Phoenix.
This work was supported by the Phoenix VA Health Care
System and the Institute for Mental
Health Research to Martha Kent. Contents do not represent
views of the Department of
Veterans Affairs or U. S. Government.
Correspondence concerning this article should be addressed to
Martha Kent, Research De-
partment, R151, Phoenix VA Health Care System, 650 E. Indian
School Road, Phoenix, AZ
85012. E-mail: [email protected]
C© 2011 International Society for Traumatic Stress Studies.
View this article online at
wileyonlinelibrary.com DOI: 10.1002/jts.20685
In this study, we tested an alternative approach to PTSD treat-
ment, one that emphasizes reengagement of the approach moti-
vational system to create a positive hub of regulatory activity.
A
capacity-building approach is grounded in evidence from
disparate
literatures pointing to the value of positive resilience resources
in
ameliorating responses to trauma or stress. Strong social
relation-
ships and self-efficacy are protective factors for children facing
adversity (Masten, 2001). Likewise, positive emotions
accelerate
physiological recovery following stress-induced autonomic
arousal
(Tugade & Fredrickson, 2004), and even build intellectual and
cognitive resources (Fredrickson, 2004).

This preliminary randomized trial of a resilience-oriented treat-
ment attempted to bolster positive emotional engagement and
so-
cial bonds prior to revisiting past traumatic experiences in
veterans
with PTSD. The treatment was expected to have broad effects,
improving symptoms, positive emotional health, and cognitive
function compared to a waitlist control.
M E T H O D
Participants and Procedure
Eligibility criteria included (a) being United States veterans
from
the Vietnam war era up through the Gulf war, and (b) scoring >
591
592 Kent et al.
40 on the Clinician-Administered PTSD Scale (CAPS; Blake et
al.,
1998). Exclusion criteria included (a) active suicidality, (b)
active
alcohol/substance abuse, (c) psychosis, and (d) life-threatening
illness.
Of 47 individuals assessed for eligibility, 39 completed the
preassessment and were randomly assigned stratifying by sex (n
=
20 intervention, n = 19 control). The sample was primarily male
(67%), Caucasian (75.8 %), and unemployed (66%), and had
attended at least some college (73%). Mean age was 54 years

(SD =
8.34, range = 34–66), and modal household income was
$21,000–
$24,999. The traumas indexed by the CAPS were combat (31%),
childhood sexual abuse (21%), childhood physical abuse (18%),
violent unexpected death of another (14%), sexual assault (6%),
physical assault (5%), and accident (5%). The PTSD symptom
duration averaged 12 years (range = 1–41).
Study procedures were approved by the institutional review
boards of the Phoenix Veterans Affairs Health Care System
(VA)
and Arizona State University and conducted at the VA.
Individuals
were recruited via posted flyers at the VA and local veteran
centers,
screened by phone, consented, and administered the CAPS by a
psychologist (MK) or a master’s-level clinician trained in the
use
of the CAPS according to the guidelines of the National Center
for PTSD.
Pre and postassessment included completion of standardized
questionnaires assessing mental health symptoms and positive
emotional health, and neuropsychological testing. Because it
was
a group intervention, randomization was in waves. Participants
in
both conditions within a treatment wave completed
questionnaire
postassessments 1 week after and neuropsychological
assessments
within 5 weeks of the final treatment session in meetings
conducted
at the VA.

The manualized intervention, delivered in 12 weekly 90-minute
group sessions, initially fosters resilience resources (e.g.,
awareness
of positive emotions; social connectedness) that individuals are
en-
couraged to draw on in later sessions when they revisit
traumatic
episodes or experience PTSD symptoms. Week 1 provides an
intro-
duction to the concept of resilience and the topics to be covered
in
the remainder of the treatment. Weeks 2 and 3 focus on
increasing
attention to bodily sensations as sources of vitality and engage-
ment. Weeks 4–7 seek to build positive emotional experiences
and
social bonds, first by drawing on childhood experiences and
then
expanding the focus to include current experiences. Weeks 8–10
involve revisiting stressors and traumas, tapping into the
positive
emotional resources and strong social bonds developed in
earlier
sessions. Finally, Weeks 11 and 12 focus on developing a plan
to
sustain change.
The intervention was implemented in two waves (Wave 1 was
10 intervention, 9 controls; Wave 2 was 10 intervention, 10
con-
trols), with all treatment delivered by the same clinician (MK).
Average attendance was 9.75 group sessions (SD = 2.24, range
=
2–12).
Measures

The CAPS (Blake et al., 1998) provided a PTSD diagnostic and
symptom score at enrollment. Pre- to-post change in mental
health
symptoms were assessed via the Posttraumatic Stress Diagnostic
Scale (PDS; Foa, 1995) for PTSD symptoms; Beck Depression
Inventory-II (BDI-II; Beck, Steer, & Brown, 1996) for
depressive
symptoms, and the State-Trait Anxiety Inventory (STAI; Spiel-
berger, Gorsuch, & Lushene, 1970) for anxiety levels. Internal
consistencies at pre- and post-treatment were good for measures
of PTSD (pre = .83; post = .96); depression (pre = .93; post =
.95); anxiety (pre = .90; post = .95).
Pre- to-post change in positive emotional health was assessed
with the Vitality and Social Functioning subscales of the RAND
36-item Health Survey-1.0 (Ware, Kosinski, & Keller, 1994),
and by summing the Purpose in Life (e.g., “I have a sense of
direction and purpose in life”), Positive Relations with Oth-
ers (e.g., “Most people see me as loving and affectionate”), and
Personal Growth (e.g., “Life has been a continuous process of
learning, changing, and growth”) subscales of the Psychological
Well-Being Scale (Ryff, 1989) to provide a rating of well-being.
In-
ternal consistency was adequate to good at pre- and post-
treatment
for Vitality (pre = .84; post = .75), Social Functioning (pre =
.80; post = .68), and Well-being (pre = .90; post = .90).
Executive function was assessed with the Word Generation sub-
test of the Neuropsychological Assessment Battery (Stern &
White,
2003), and the Category Fluency, Category Switching, and
Color-
Word Switching subtests of the Delis-Kaplan Executive
Function
System (Delis, Kaplan, & Kramer, 2001). Repeatable Battery

for
the Assessment of Neuropsychological Status (Randolph, 1998)
subtests assessed working memory (List Learning) and episodic
memory (Story Memory, Story Recall, List Recall). Alternate
ver-
sions of all cognitive tests were employed at pre- and
postassess-
ment. Test-retest reliability from pre- to postasssessment ranged
from .50 for Category Switching to .74 for Word Generation.
Data Analysis
Both t -test and chi-square analyses were conducted to compare
groups on demographic characteristics and pretreatment levels
of
all outcome measures. Pretreatment scores of three individuals
who dropped out prior to study completion (one intervention,
two control participants) were carried forward to posttreatment
(i.e., intent-to-treat; see Figure 1).
To evaluate intervention effects, 2 (Group) × 2 (Time) re-
peated measures analyses of variance (ANOVA) were performed
for symptom, positive emotional health, executive function, and
memory measures. Significant Group × Time effects were fol-
lowed by within-group t -tests. Cohen’s d values (Cohen, 1988)
estimated the magnitude of within-group pre- to-post change,
and
Journal of Traumatic Stress DOI 10.1002/jts. Published on
behalf of the International Society for Traumatic Stress Studies.
Resilience-Oriented Treatment for PTSD 593
Assessed for eligibility (n = 47)
Excluded (n = 8)

Not meeting inclusion criteria (n = 2)
Declined to participate (n = 2)
Other reasons (n = 4)
Analyzed (n = 20)
Excluded from analysis (n = 0);
Pretreatment values carried forward to
posttest for 1 drop out
Discontinued intervention and lost to
posttest (n = 1)
Allocated to intervention (n = 20)
Received allocated intervention (n = 19)
Did not receive allocated intervention (n
= 1; due to increased family
obligations)
Discontinued intervention and lost to
posttest (n = 2)
Allocated to waitlist (n = 19)
Received allocated waitlist (n = 17)
Did not receive allocated waitlist (n = 2; 1
due to worsening mental health, 1 due
to increased family obligations)
Analyzed (n = 19)
Excluded from analysis (n = 0);
Pretreatment values carried forward to
posttest for 2 dropouts
Allocation

Analysis
Posttest
Randomized (n = 39)
Enrollment
Figure 1. CONSORT flow diagram.
between-group differences in pre- to-post change and were
calculated as the difference between the two group means
divided
by the pooled standard deviation of the groups.
R E S U L T S
At pretreatment, the groups were comparable on demographic
characteristics ( p s > .13) and CAPS score (intervention M =
74.79, SD = 24.07; control M = 83.42, SD = 18.13, t (37) =
1.25, ns). Groups also were comparable across dependent
measures
( p s > .43), suggesting that random assignment yielded groups
that
were equivalent at pretreatment.
Table 1 presents the pre- and posttreatment raw scores and d
val-
ues for symptoms, positive emotional health, executive
function,
and memory separately for the intervention and control groups.
Pre- to-post changes in symptom scores all varied by group,
Group
× Time p s < .003. Within-group comparisons revealed that the
intervention group improved on all measures pre- to
posttreatment
( p s < .02, d s = 0.82–1.18), whereas waitlist controls remained

unchanged ( p s > .09). Differences between groups in the mag-
nitude of symptom change pre- to posttreatment were large for
PTSD, depression, and anxiety (d s > 1.02). The proportion
scor-
ing in the severe range posttreatment was 30% for intervention
versus 58% for control participants for PTSD (PDS > 35), and
10% for intervention versus 63% for control participants for de-
pression (BDI-II > 28). The proportion of individuals scoring in
the severe range at posttreatment differed between groups only
for
depression; χ2(1, N = 39) = 11.97, p = .001.
Pre- to-post change in positive emotional functioning measures
varied between groups, Group × Time p s < .05. Within-group
comparisons indicated that the intervention group improved in
well-being, vitality, and social functioning ( p s < .05, ds = 0.42
–0.73), whereas controls showed declines in well-being ( p =
.039,
d = −0.32) and no change in vitality or social functioning.
Effect
sizes comparing group pre- to-post change were large for well-
being and vitality (d s > 0.96), and moderate-to-large for social
functioning (d = 0.68).
Pre- to-post change varied by group for three measures of ex-
ecutive functioning: Word Generation, Category Fluency, and
Color-Word Switching, Group × Time p s < .03. Within-group
comparisons for the intervention group revealed that pre-to-
post change was significant for Word Generation ( p < .04,
d = 0.35), marginally significant for Category Fluency ( p < .06,
d = 0.30), and nonsignificant for Color-Word Switching ( p >
.10). For controls, Word Generation scores declined ( p = 0.011,
d = −0.37), whereas other measures of executive function re-
mained unchanged (ps > .08). The magnitude of group
differences

594 Kent et al.
Table 1. Pre- and Posttreatment Group Means, and Standard
Deviations, and Treatment Effect Sizes
Pretreatment Posttreatment
Intervention (n = 20) Control (n = 19) Intervention (n = 20)
Control (n = 19)
Measures M SD M SD M SD M SD Cohen’s d a
PDS 35.90 9.49 37.53 9.77 23.00 12.23 36.90 9.65 1.40
BDI-II 26.05 10.04 29.16 16.16 15.65 8.71 29.90 15.16 1.25∗
STAI-State 53.05 10.64 54.53 12.84 44.05 11.32 56.63 10.58
1.02∗
Well-Being 78.95 14.62 75.95 18.95 88.50 14.85 69.95 18.55
1.30∗
SF-36 Social Role Functioning 36.25 18.54 30.92 29.28 45.63
25.42 26.32 17.63 0.68∗
SF-36 Vitality 33.31 14.28 34.80 16.66 43.44 13.49 31.97 14.83
0.96∗
Word Generation 10.00 5.21 10.32 4.80 11.75 4.82 8.63 4.45
1.10
Category Fluency 9.50 3.50 10.05 3.19 10.55 3.47 9.32 2.77

0.80∗
Category Switching 10.80 3.09 10.84 3.08 11.70 3.60 9.84 4.02
0.58
Color-Word Switching 8.60 3.93 9.58 3.78 9.80 2.98 8.74 3.11
0.77∗
List Learning 25.15 4.98 25.05 5.05 27.95 5.40 24.74 5.39
0.79∗
Story Memory 16.90 4.42 16.58 3.72 17.35 4.16 15.58 3.44 0.43
List Recall 3.75 2.15 4.21 2.10 4.95 2.24 4.37 1.98 0.67∗
Story Recall 8.25 1.83 8.53 2.01 9.35 1.84 8.42 2.34 0.66∗
Note. PDS = Posttraumatic Stress Diagnostic Scale; BDI-II =
Beck Depression Inventory-II; STAI = State-Trait Anxiety
Inventory. Cohen’s d reflects between-group
differences in magnitude of pre- to-post change.
a Significance levels based on 2 (Group) × 2 (Pre-, Post-)
analyses of variance interaction F s (1,37).
∗ p <.05.
in pre- to-post change was large for both Word Generation and
Category Fluency (ds > 0.80).
Pre- to-post change varied by group for three memory mea-
sures: List Learning, List Recall, and Story Recall, Group ×
Time
p s < .05. Within-group comparisons revealed that all three
scores
improved significantly from pre- to post-treatment in the
interven-
tion group ( p s < .03, d s = 0.54–0.60), but not in controls ( p s
>
.53, d s = −0.08–0.06). The magnitude of the difference between

groups was moderate-to-large for Story Recall,List Learning,
and
List Recall (d s > 0.66).
D I S C U S S I O N
The current findings suggest that a PTSD intervention designed
to
enhance resilience capacities yields benefits for veterans not
only
in symptoms, but also in positive emotional health and cogni-
tive functioning, compared to a waitlist control. Large
treatment-
control differences emerged for mental health symptoms,
compa-
rable to those reported in trials of established treatments of
PTSD
(Powers, Halpern, Ferenschak, Gillihan, & Foa, 2010). Thus, at-
tention to both symptoms and positive aspects of health appear
to
be warranted in efforts to promote change in PTSD.
Most novel are the findings for cognitive function. The treat-
ment group showed modest improvement whereas controls
showed
modest declines in executive function. These findings extend
those of a small, uncontrolled study that found moderate to-
post-
treatment improvement in executive function among women re-
ceiving individual psychotherapy for PTSD (Walter et al.,
2010).
The treatment group also showed moderate-to-large gains rela-
tive to controls in two memory processes: encoding and reten-
tion/retrieval. Encoding of new information is a process
engaging
prefrontal cortical areas that deliver information to the

hippocam-
pal medial temporal region, and is the most common memory
impairment in PTSD (Vasterling & Bailey, 2005). Retention and
retrieval, memory processes that engage frontal-temporal-polar
re-
gions, also showed significant improvement.
Several methodological limitations deserve comment. First, use
of a small sample of veterans, a single clinician delivering
treatment,
and variability in session attendance constrains generalizability.
Second, individuals administering posttreatment cognitive tests
were aware of treatment condition. Although the testing format
is
standardized, there remains potential bias in cognitive
outcomes.
Third, no data are available to determine whether treatment
gains
persist over time. Finally, the contribution of nonspecific
factors
to intervention effects cannot be evaluated due to use of a
waitlist
control.
Nevertheless, this study suggests that a capacity-building
approach to PTSD is a tolerable treatment (i.e., one with low
attrition) that may address multiple problems in PTSD, includ-
ing decreased positive emotional health and cognitive
dysfunction.
Resilience-Oriented Treatment for PTSD 595

Future work that attempts to replicate the current findings, elab-
orate potential treatment mechanisms, and compare a resilience
approach with more established PTSD treatments is warranted.
R E F E R E N C E S
Aupperle, R. L., & Paulus, M. P. (2010). Neural systems
underlying approach and
avoidance in anxiety disorders. Dialogues in Clinical
Neuroscience, 12, 517–
529.
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck
Depression Inventory-II
(BDI-II). San Antonio, TX: Psychological Corporation.
Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G.,
Charney, D. S., &
Keane, T. M. (1998). Clinician-Administered PTSD Scale for
DSM-IV. Boston,
MA: Boston National Center for PTSD.
Cohen, J. (1988). Statistical power analysis for the behavioral
sciences (2nd ed.).
Hillsdale, NJ: Erlbaum. doi:10.1002/0470013192.bsa109
Delis, D. C., Kaplan, E., & Kramer, J. H. (2001). Delis-Kaplan
Executive Function
System (D-KEFS). San Antonio, TX: Psychological
Corporation.
Foa, E. B. (1995). The Posttraumatic Diagnostic Scale (PDS)
manual. Minneapolis,
MN: National Computer Systems.

Foa, E. B., Keane, T .M., Friedman, M. J., & Cohen, J. (2008).
Effective treatments
for PTSD: Practice guidelines from the International Society for
Traumatic
Stress Studies (2nd ed.). New York, NY: Guilford Press.
Fredrickson, B. L. (2004). The broaden-and-build theory of
positive emotions.
Philosophical Transactions of the Royal Society B: Biological
Sciences, 359, 1367–
1378. doi:10.1098/rstb.2004.1512
Kashdan, T. B., Breen, W. E., & Julian, T. (2010). Everyday
strivings
in war veterans with posttraumatic stress disorder: Suffering
from a
hyper-focus on avoidance and emotion regulation. Behavior
Therapy, 41, 350–
363. doi:10.1016/j.beth.2009.09.003
Masten, A. S. (2001). Ordinary magic: Resilience processes in
development. Amer-
ican Psychologist, 56, 227–238. doi:10.1037//0003-
066X.56.3.227
Powers, M. B., Halpern, J. M., Ferenschak, M. P., Gillihan, S.
J., &
Foa, E. B. (2010). A meta-analytic review of prolonged
exposure for
posttraumatic stress disorder. Clinical Psychology Review, 30,
635–641.
doi:10.1016/j.cpr.2010.04.007
Randolph, C. (1998). Repeatable battery for the assessment of
neuropsychological status.

San Antonio, TX: Psychological Corporation.
Ryff, C. D. (1989). Happiness is everything, or is it?
Explorations on the meaning
of psychological well-being. Journal of Personality and Social
Psychology, 57,
1069–1081. doi:10.1037/0022-3514.57.6.1069
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970).
STAI manual for the
State-Trait Anxiety Inventory. Palo Alto, CA: Consulting
Psychologists Press.
Stern, R. A., & White, T. (2003). Neuropsychological
Assessment Battery (NAB).
Lutz, FL: Psychological Assessment Resources, Inc.
Tugade, M. M., & Fredrickson, B. L. (2004). Resilient
individuals use positive
emotions to bounce back from negative emotional experiences.
Journal of Per-
sonality and Social Psychology, 86, 320–333. doi:10.1037/0022-
3514.86.2.320
Vasterling, J. J., & Bailey, K. (2005). Neuropsychological
finding in adults with
PTSD. In J. J. Vasterling & C. R. Brewin (Eds.),
Neuropsychology of PTSD:
Biological, cognitive, and clinical perspectives (pp. 178–207).
New York, NY:
Guilford Press.
Walter, K. H., Palmieri, P. A., & Gunstad, J. (2010). More than
symptom reduction:
Changes in executive function over the course of PTSD
treatment. Journal of

lable at ScienceDirectJournal of Psychiatric Research 45 (.docx

lable at ScienceDirectJournal of Psychiatric Research 45 (.docx

Recommended

Recommended

More Related Content

Similar to lable at ScienceDirectJournal of Psychiatric Research 45 (.docx

Similar to lable at ScienceDirectJournal of Psychiatric Research 45 (.docx (18)

More from smile790243

More from smile790243 (20)

Recently uploaded

Recently uploaded (20)

lable at ScienceDirectJournal of Psychiatric Research 45 (.docx