Filetype PDF | Posted on 28 Sep 2022 | 4 years ago
Authentication
digital resources
276x Filetype PDF File size 0.58 MB Source: academic.oup.com
MILITARYMEDICINE,177,12:1486,2012
GroupExposureTherapyTreatmentforPost-TraumaticStress
Disorder in Female Veterans
Diane T. Castillo, PhD*; Janet C’de Baca, PhD*; Clifford Qualls, PhD†; Marina A. Bornovalova, PhD‡
ABSTRACT Objectives: The purpose of this study was to examine the application of a group exposure therapy
model, the content of which consisted solely of repeated imaginal exposure during sessions, in a clinical sample of
female veterans with post-traumatic stress disorder (PTSD). Establishing group delivery of exposure therapy will
expand options, increase efficiency, and introduce group curative factors. Methods: Eighty-eight female veterans with Downloaded from https://academic.oup.com/milmed/article/177/12/1486/4336765 by guest on 28 September 2022
PTSD completed a six-session exposure group, three participants per group, as a component of a larger treatment
program. The PTSD symptom checklist (PCL) was used as the outcome measure and administered in each session.
Results: Pre/post-paired t-tests showed significant improvement in PTSD on the PCL, with 40% of completers showing at
least a 10-point drop in the PCL scores. In addition, a repeated measures analysis of variance showed a significant main
effect and a significant quadratic equation, with expected initial increases in the PCL followed by a decrease below
baseline at session 6. Conclusions: The group exposure treatment protocol showed positive outcomes on PTSD
symptoms in a real-world clinical sample of female veterans. The implications include an expansion of exposure
treatment choices for veterans with PTSD and increased options for therapists.
INTRODUCTION interventions9 and minimal examination of exposure therapy
Exposure therapy, a treatment for post-traumatic stress dis- in a group format.
order (PTSD), has consistently proven to be efficacious1 and The primary challenge of providing exposure therapy in
effective2 in reducing PTSD symptoms and is one of the a group format comparable to the robust individual PE
two therapies recommended as the first line of treatment by model is logistic. In a typical 8-member, 90-minute group,
3
VA/DoD guidelines. Early research examining the utility of in-session imaginal exposures cannot be conducted for
exposure therapy addressed PTSD among civilian rape vic- each member in every session. Two studies10,11included
4 5 6
tims and male combat veterans. Recently, effectiveness was two in-session imaginal exposures per participant among
established for exposure therapy over present-centered therapy other treatments in a group format and found PTSD
in reducing PTSD symptoms in a sample of female veterans. improvement after 30 to 36 sessions. The two in-session
The most developed model for the delivery of exposure ther- imaginal exposures were supplemented by daily listening
7
apy is prolonged exposure (PE), which consists primarily of to audio recordings of the trauma narrative between sessions
in-session, repeated imaginal exposures to a traumatic memory to achieve desensitization. Problematic with this approach
and out-of-session in vivo exposures to avoided situations. is controlling for dosage of exposure, as homework compli-
Minor elements of PE are education about PTSD symptoms, anceisvariableandexpectedtobehighinearlyses-
rationale for treatment, and breathing retraining. sions. Another methodological limitation was determining
ThePEmodelandthemajorityofexposuretherapyclinical the contribution of exposure compared to other treatments
trials have examined exposure therapy delivered in an individ- (cognitive, relapse prevention, relaxation training). The
ual format—onetherapisttoonepatient—for10to12sessions, other therapies likely contributed to PTSD improvement,
with imaginal exposure conducted in 8 to 10 of these sessions. serving to confound conclusions on the effects of the expo-
Although historically most veterans administration (VA) out- sure treatment. Despite these limitations, these two studies
patient PTSD programs have offered therapies in a group provided the necessary first steps in establishing the pros-
format,8 research on group delivery of treatments for PTSD pect of examining group exposure therapy as an expansion
is sparse and methodologically weaker than the individual to the existing individual model.
trials, with no differential effects found between treatment The primary aim of this study is to examine the effects of
group exposure therapy on PTSD with imaginal exposure
separated from other interventions, with the expectation of
*New Mexico VA Health Care System, Behavioral Health Care Line improvement in PTSD. The model for the exposure inter-
(116), 1501 San Pedro SE, Albuquerque, NM 87108.
†Clinical & Translational Science Center (CTSC), 1 University of New vention was developed as part of a larger protocol for PTSD
Mexico, Albuquerque, NM 87131-0001. treatment by the first author and is described in detail else-
‡Department of Psychology, University of South Florida, 4202 East where.12 A second aim of the study is to describe the course
Fowler Avenue, PCD 4118G, Tampa, FL 33606. of PTSD symptoms across sessions. It is expected that PTSD
This article was presented by Castillo, DT, Bornovalova, MA, and symptoms will increase slightly before decreasing at the
LeBow, S in poster format at the International Society for Traumatic Stress
Studies, Baltimore, MD, November 2006. termination of therapy. Ultimately, our goal is to expand
doi: 10.7205/MILMED-D-12-00186 options for exposure therapy utilization.
1486 MILITARYMEDICINE,Vol.177,December2012
Group Exposure Therapy Treatment for PTSD in Female Veterans
each session. The structure of the exposure group consisted
METHODS
Participants of a combination/modification of exposure techniques by
Keane et al,17,18 Foa et al,7 and Resick and Schnicke.19 The
The sample consisted of 88 female veterans with current and/ first session included an orientation to the group, rationale for
or lifetime PTSD, treated in 33 groups in a southwest VA exposure treatment, selection of an index trauma, and direc-
outpatient women’s PTSD clinic between 1995 and 2011. tion to write a detailed trauma description for the following
Participants diagnosed with a comorbid psychotic disorder session. In-session imaginal exposure was conducted in ses-
wereexcludedfromtheexposuregroup.Individualswithother sions 2 through 5, and was structured such that each partici-
acute comorbid psychiatric disorders (e.g., substance use dis- pantfirstreadaloudtheirtraumanarrative,wasguidedthrough
order and depression) were stabilized before treatment. an imaginal exposure of the trauma memory, and then
processed the experience. Approximately 30 of the 90 minutes Downloaded from https://academic.oup.com/milmed/article/177/12/1486/4336765 by guest on 28 September 2022
Measures were spent with each participant on imaginal exposure for a
The assessment consisted of a semistructured interview and total of four in-session imaginal exposures. Participants were
interview administration of the clinician administered PTSD instructed to write and rewrite the trauma narrative each week
13 following sessions 1, 2, and 3 to allow for developing a com-
Scale (CAPS ). The CAPS is a structured interview adminis-
tered by a trained clinician to assess for PTSD and is considered plete narrative. Participants were asked to read the completed
the gold standard in diagnosing PTSD. Each of the 17 symp- narrative daily between sessions 4 and 6, for 2 weeks.
tomsareassessedforfrequencyandintensityinthepastmonth Measurement of homework compliance was not conducted;
and lifetime. The CAPS has shown internal consistency for however, participants were strongly encouraged to write the
the three-symptom categories of PTSD—reexperiencing, trauma narrative each week and were required to verbally
avoidance/numbing, and hyperarousal—with alpha coeffi- complete the imaginal exposure regardless of writing compli-
cients that have ranged from 0.73 to 0.85; convergent validity ance. It was not unusual for participants to write sparse
has been found between the CAPS and other measures of descriptions or not write at all in first exposure sessions, but
14 compliance increased in subsequent sessions. Treatment was
PTSD. Internal consistency using Cronbach’s a was com-
puted on the 17 CAPS symptom scores in the clinic sample provided by licensed clinicians, including four staff psychol-
and revealed an overall a = 0.85 with item correlations >0.40 ogists, two social workers, and two clinical nurse specialists,
for all items (symptoms) except symptom 8 (psychogenic all trained by the first author in the 6-session protocol.
amnesia), which had a correlation of 0.12. The PTSD symp- Psychology interns and postdoctoral fellows cofacilitated
15 groups with a licensed clinician. Training consisted of didac-
tom checklist (PCL) was administered at each of the six-
exposure treatment sessions to document changes in PTSD tics on theory and procedure followed by cofacilitation with
symptoms during treatment. The PCL is a 17-item, five-point the first author, before conducting the groups independently.
Likert scale with each PTSD symptom anchoredfrom1(notat Fidelity was not systematically checked; however, training
all) to 5 (extremely). The PCL is frequently used in clinical tapes for the exposure groups developed by the first author
settings and has a high correlation (0.93) with the CAPS, were reviewed periodically by clinicians after the initial
high internal consistency (Chronbach’s a = 0.94), with a sen- training. All data were collected through archival record
sitivity of 0.78, specificity of 0.86, and diagnostic efficiency review and approved by local VA and University of New
16 Mexico Institutional Review Boards.
of 0.83.
RESULTS
Procedure Anintent-to-treat analysis was conducted with all 88 partici-
Female veterans diagnosed with PTSD were offered a variety pants, imputing the last PCL value forward in a paired t-test
of group treatments developed by the first author12 including and was found significant (p < 0.03). Eleven participants
an unstructured psychoeducation group, followed by four (12.5%) were defined noncompleters, as they attended less
possible structured, topic-specific groups, including cognitive than 4 of the 6 sessions. The final completer sample consisted
restructuring, behavioral interventions (assertiveness and of 77 participants in 32 groups, with 59% (n = 45) attending
relaxation training, nightmare therapy), sexual functioning, all 6 sessions, 31% (n = 24) 5 sessions, and 10% (n = 8)
and exposure therapy, the latter of which is presented in this 4sessions. The characteristics of the 77 subjects reflected entry
article. Selection of groups was optional, most attended expo- level PTSD scores on the CAPS similar to other clinical
sure after other treatments, and most with the same cohort. populations (current: mean (M) = 67.55; lifetime: M = 104.50;
13
Someindividuals subthreshold for PTSD elected to attend the total: M = 170.28). Trauma characteristics of the sample
exposure group. were sexual trauma alone (57%), combat and other nonsexual
Theexposure therapy group consisted of 6 weekly sessions traumas (4%), and multiple traumas including sexual trauma
of 90-minutes each focusing exclusively on repeated imagi- (39%). Eighty-eight percent reported more than one trauma,
nal exposure. Three members participated in each group with with 18% childhood only (under age 18), 34% adult only, and
twotherapists. The PCL was administered at the beginning of 48% both childhood and adulthood trauma. Medical record
MILITARYMEDICINE,Vol.177,December2012 1487
Group Exposure Therapy Treatment for PTSD in Female Veterans
TABLEI. Demographics, Trauma Characteristics, and CAPS in The examination of PTSD improvement with group expo-
Female Veterans (n = 77) sure treatment was addressed through paired t-test analysis of
Type of Trauma n % the pre/post-total PCL scores. Paired t-tests were also computed
on the three pre/post-PTSD symptom categories within the
Sexual 44 57.1 PCL—reexperiencing, avoidance/numbing, and hyperarousal.
Combat 2 2.6 TherewasasignificantdecreaseinthetotalPCL(p < 0.01,
Other 1 1.3
Combination (Sexual + ) 30 39.0 effect size = 0.26) and avoidance/numbing symptom cate-
Combat Exposure 8 10.4 gory scores (p < 0.001, effect size = 0.40), suggesting the
PTSD+OtherDiagnosis 56 73 group exposure treatment was effective in lowering PTSD
Ethnicity and avoidance/numbing PTSD symptoms in the sample. The
Non-Hispanic White 50 65
Hispanic 18 23 means, SDs, and within-group effect sizes for total and symp- Downloaded from https://academic.oup.com/milmed/article/177/12/1486/4336765 by guest on 28 September 2022
Other 9 12 tom categories for the PCL are presented in Table II. Follow-
AgeatTimeofTrauma n % 6
upanalyses examined PTSD improvement in three additional
Childhood 14 18.1 ways:(1)thepercentage/numberofsubjectsnolongermeeting
Adulthood 26 33.8 criteria for the disorder, defined as the PCL scores 50 or below,
Both 37 48.1
NumberofTraumas (2) the percentage/number improving by 10 PCL points, and
1 9 11.7 (3) the percentage/number in complete remission, defined as
>1 68 88.3 the PCL scores decreasing 20 or more points. Twenty-two of
Marital Status the total 77 participants, with baseline PCL scores below 50
Married 24 31 were removed, as they were not clinically significant on
Divorced 27 35
Never Married 25 33 treatment entry. In the remaining sample of 55 participants,
Widowed 1 1 22% (n = 12) no longer met clinical significance for PTSD
Current 67.6 20.9 (PCL < 50), 40% (n = 22) dropped 10 or more PCL points,
Lifetime 104.5 19.1 and 13%(n=7)wereincompleteremissionwithPCLscores
Total 170.3 38.2 dropping 20 points or more at the completion of the group
CAPSa M SD
Current 67.6 20.9 exposure therapy.
Lifetime 104.5 19.1 To address possible correlations (clustering) among indi-
Total 170.3 38.2 viduals within groups and impact on outcome results, an
aMissing = 18. analysis of variance (ANOVA) was conducted on CAPS and
baseline PCLscoreswithgroupdefinedastheunitofanalysis
(n = 32; group mean scores) rather than the individual subject
review showed 73% diagnosed with a psychiatric diagnosis in (n = 77). Significant differences were found between groups
addition to PTSD. The average age was 46.1 (SD = 9.4) and on both the PCL and CAPS scores (p < 0.001). A repeated
ethnicity reflected a primarily non-Hispanic white (65%) and measures analysis of covariance using baseline PCL scores as
Hispanic (23%) sample. The complete demographics with a covariate was computed to control for variability in PTSD
details are shown in Table I. The 77 participants completing betweengroupsandthepre/post-PCLscoresremainedsignif-
the treatment were compared to the 11 noncompleters (t-tests icant (p < 0.01), suggesting the correlation within groups did
2 not alter the original significant outcome results.
for continuous and c -tests for discrete variables) on the entry
level CAPS scores and baseline demographics (age at time The examination of the course of PTSD symptom change
of trauma, number of traumas, type of trauma, combat expo- across sessions in the group exposure was conducted by a
sure, and comorbid psychiatric diagnoses), as well as the expo- repeated measures ANOVA (RM-ANOVA) for the total PCL
sure group entry PCL scores. None of the comparisons scores across the six sessions. The RM-ANOVA resulted in a
revealed significant differences between the noncompleter and significant session main effect (p < 0.001), suggesting differ-
completer participants. ences between sessions across time. To determine the exact
TABLE II. Means, Standard Deviations, and Effect Sizes for Total PCL Scores and for Each Symptom Category (n = 77)
PTSDSymptomCategories
Total PCL Reexperiencing Avoidance/Numbing Hyperarousal
M SD M SD M SD M SD
Pre 59.69 14.23 17.10 4.77 24.54 6.68 18.08 4.59
Post 55.70* 16.36 16.42 5.51 21.78** 7.19 17.52 5.07
Effect Size 0.26 ns 0.40 ns
*p < 0.01, **p < 0.001. ns, not significant.
1488 MILITARYMEDICINE,Vol.177,December2012
Group Exposure Therapy Treatment for PTSD in Female Veterans
Downloaded from https://academic.oup.com/milmed/article/177/12/1486/4336765 by guest on 28 September 2022
FIGURE1. MeanPCLscoresacross six treatment sessions and best-fit quadratic curve.
22
type of changes occurring in PTSD across the six sessions, the PTSD symptoms also replicate finding in the use of PE
total PCL data were used in an RM-ANOVA to compute a where some participants experienced a temporary increase in
quadratic regression by including linear and quadratic terms as PTSD symptoms before improvement. The elevations in the
factors. This analysis was significant (p = 0.003; predicted PCL scores in sessions 2 through 4 in our study may appear
2
total PCL = 63.08 + 0.064 sessions − 1.18 [session–3] ). exaggerated because of the weekly PCL administration, rather
7
The actual and predicted means are plotted in Figure 1. As than bimonthly, as in typical PE protocols. Although slight
expected, the significant quadratic effect showed the total PCL increase in symptoms during initial exposure sessions is not
scores increase (PTSD symptoms worsened) before decreasing surprising, given avoidance characterizes the disorder, most
(PTSDsymptoms improved) to a value lower than baseline. important is the significant decrease in PTSD symptoms at the
completion of treatment. The significant parabolic curve lends
CONCLUSION information for further examination of the process of desensi-
This study examined exposure therapy in a group format, in an tization during exposure therapy.
applied setting, with a structure that excluded other interven- Thestructure of the exposure group in this study improves
tions. Overall, PTSD symptoms decreased from pre- to post- on two key methodological issues of past studies10,11 by
therapy using this group exposure format, consistent with expanding the two in-session imaginal exposures to four and
established findings of exposure treatment in an individual isolating the exposure component from other interventions.
1,2,4,5 Repeatedimaginalexposuretoatraumamemoryisnecessary
format. The findings were particularly notable, as (1) the
clinical sample had multiple traumas highly comorbid with to attain desensitization effects and is less likely to occur with
other psychiatric diagnoses, and (2) the sample received other onlytwoin-sessionimaginalexposures.Infact, amorerecent
23
therapies, including an evidence-based cognitive therapy, study showedacorrelation between PTSD improvement and
before the exposure group therapy. Also notable was the low homeworkcomplianceinlisteningtorecordingsofthetrauma.
noncompleter/dropout rate of 12.5%, much lower than the Although this group protocol only provided 4 in-session
20
dropout rates in other studies (19–27%). The changes on the imaginal exposures—compared to 8 in the individual PE
total PTSD symptomseverityappeartobedrivenbydecreases protocol—improvement was shown, charted across time, and
in avoidance and numbing symptoms, which can be explained addressed the logistic problems by limiting inclusion to three
by current theory and empirical work that indicates emotional participants per group. The protocol more closely approaches
andbehavioralavoidanceareparticularlyresponsivetoexposure the individual PE protocol than previous group studies investi-
7,21
interventions. Thus, the significant reduction in avoidance/ gating exposure therapy and assured an adequate dose of the
numbingPTSDsymptomsfurthertestifiestotheutility of group therapy with imaginal exposure for every participant during
exposure therapy. Finally, increases followed by decreases in sessions when imaginal exposure was conducted. This study
MILITARYMEDICINE,Vol.177,December2012 1489
no reviews yet
Please Login to review.
