Journal of Consulting and Clinical Psychology
2014, Vol. 82, No. 4, 551–561
In the public domain
Cognitive Processing Therapy for Veterans With Posttraumatic Stress
Disorder: A Comparison Between Outpatient and Residential Treatment
Kristen H. Walter
Ruth L. Varkovitzky
Veterans Medical Research Foundation, San Diego, California,
and Veterans Affairs San Diego Healthcare System, San Diego,
Veterans Affairs Puget Sound Healthcare System, American
Lake Division, Tacoma, Washington
Gina P. Owens
University of Tennessee–Knoxville
Cincinnati Veterans Affairs Medical Center, Cincinnati, Ohio
Kathleen M. Chard
Cincinnati Veterans Affairs Medical Center, Cincinnati, Ohio, and University of Cincinnati
Objective: Across the Veterans Affairs (VA) Healthcare System, outpatient and residential posttraumatic
stress disorder (PTSD) treatment programs are available to veterans of all ages and both genders;
however, no research to date has compared these treatment options. This study compared veterans who
received outpatient (n ⫽ 514) to those who received residential treatment (n ⫽ 478) within a VA
specialty clinic on demographic and pretreatment symptom variables. Further, the study examined preto posttreatment symptom trajectories across the treatment programs. Method: All 992 veterans met
diagnostic criteria for PTSD and attended at least 1 session of cognitive processing therapy (CPT) in
either the outpatient or residential program. Bivariate analyses were utilized to investigate differences
between samples on demographic variables and severity of pretreatment symptoms. Multilevel modeling
(MLM) was used to investigate the change in symptomatology between the 2 samples from pre- to
posttreatment. Results: Analyses indicated that the samples differed on all demographic and pretreatment
symptom variables, with residential patients reporting higher scores on all assessment measures. MLM
results demonstrated that symptom scores improved for all veterans across time, with outpatients
consistently reporting fewer symptoms at both time points. The time by program interaction was
significant for PTSD-related symptom trajectories, but not for the depression-related symptom trajectory.
Conclusion: This is the 1st study to compare pretreatment characteristics and treatment outcome between
veterans receiving outpatient and residential PTSD treatment. Findings may help clinicians select
appropriate care for their patients by identifying relevant pretreatment characteristics and generally
informing expectations of treatment outcome.
Keywords: posttraumatic stress disorder, cognitive processing therapy, outpatient treatment, residential
Many veterans seeking mental health care at the Department of
Veterans Affairs (VA) have experienced various traumatic events,
such as combat, childhood trauma, intimate partner violence, and
sexual assault (Clancy et al., 2006). Such traumatic experiences may
lead to development of posttraumatic stress disorder (PTSD), depression, other anxiety disorders, and related symptomatology (Hoge et
al., 2004). Estimates suggest that 19%–30% of Vietnam veterans have
met lifetime criteria for PTSD (Dohrenwend et al., 2006), and as
many as 22% of Operation Iraqi Freedom/Operation Enduring Freedom veterans seeking VA care met PTSD criteria (Seal et al., 2009).
To meet these critical treatment needs, the VA provides a
continuum of options for PTSD care, including outpatient and
This article was published Online First June 9, 2014.
Kristen H. Walter, Health Services Research & Development, Veterans Medical Research Foundation, San Diego, California, and Research Service, Veterans Affairs San Diego Healthcare System, San
Diego, California; Ruth L. Varkovitzky, PTSD Outpatient Clinic, Veterans Affairs Puget Sound Health Care System, American Lake Division, Tacoma, Washington; Gina P. Owens, Department of Psychology,
University of Tennessee–Knoxville; Jennifer Lewis, Trauma Recovery
Center, Cincinnati Veterans Affairs Medical Center, Cincinnati, Ohio;
Kathleen M. Chard, Trauma Recovery Center, Cincinnati Veterans
Affairs Medical Center, Cincinnati, Ohio, and Department of Psychiatry, University of Cincinnati.
We would like to thank the staff of the Trauma Recovery Center at the
Cincinnati Veterans Affairs Medical Center, particularly Jeremiah Schumm, as
well as the data preparation efforts of Lindsey Davidson and Misty Wolfe. Content
of this article does not necessarily reflect the views of the United States government or the Department of Veterans Affairs.
Correspondence concerning this article should be addressed to Kristen H.
Walter, Veterans Medical Research Foundation, 3350 La Jolla Village Dr. (111N1), Building 13, San Diego, CA, 92161. E-mail: Kristen.Walter@va.gov
WALTER, VARKOVITZKY, OWENS, LEWIS, AND CHARD
residential treatment programs. The Veterans Health Administration (VHA) has produced guidelines for matching patient needs to
the intensity of care required (U.S. Department of Veterans Affairs, 2010). According to these guidelines, outpatient treatment is
intended to serve individuals with new onset or severe PTSD,
while residential treatment is meant to provide a semistructured
environment of rehabilitation where veterans can engage in PTSD
treatment and sobriety efforts and address medical and psychological needs prior to full reintegration into the community (U.S.
Department of Veterans Affairs, 2010). In part based on the
intended population that residential programs aim to serve, a
common assumption exists among providers that individuals in
residential treatment present with greater symptom severity and
functional impairment and, in turn, may not achieve as many
treatment gains as outpatients. Despite this assumption, no empirical literature to date has compared patients in residential and
outpatient treatment for PTSD.
Evidence-Based PTSD Treatment in the VA
Two evidence-based treatments for PTSD are currently being
disseminated through the VA system (Karlin et al., 2010), including cognitive processing therapy (CPT; Resick, Monson, & Chard,
2010) and prolonged exposure (PE; Foa, Hembree, & Rothbaum,
2007). Both CPT and PE are considered best practice models (Foa,
Keane, Friedman, & Cohen, 2008) and have been shown to effectively reduce PTSD and depression symptoms among veterans
(Monson et al., 2006; Schnurr et al., 2007; Tuerk et al., 2011).
Further treatment outcome research has explored the use of CPT
with veterans in residential and outpatient programs and has demonstrated that CPT effectively reduces symptoms of PTSD and
depression for veterans receiving treatment in both types of settings (Alvarez et al., 2011; Monson et al., 2006). CPT (rather than
PE) was the treatment of focus in the present study, as it was the
primary treatment modality used in the residential program at the
VA clinic in which this study was conducted. In an effort to make
meaningful comparisons between residential and outpatient treatment, we included only veterans receiving CPT in the current
CPT can be delivered in several formats, including individual
(Resick et al., 2010), group, and in a format which combines
individual and group sessions received concurrently (Chard, Resick, Monson, & Kattar, 2009), with the various formats effectively reducing PTSD symptoms in veteran samples (Monson et
al., 2006; Walter, Bolte, Owens, & Chard, 2012). Literature supporting the use of CPT in a group format was another reason CPT
was selected as the trauma-focused treatment modality. To date, no
treatment outcome studies have compared outpatient and residential samples in terms of either pretreatment individual differences
or symptom change trajectories. Such comparisons may shed light
on the implementation of VA levels of care and allow clinicians to
identify and use specific patient characteristics to improve treatment planning.
Based on review of the literature previously mentioned, the goal
of the current study was to empirically examine whether veterans
receiving CPT for PTSD in outpatient treatment (individual CPT
format) versus residential treatment (combined individual and
group CPT format) differed in terms of pretreatment demographic
variables, symptom severity, and treatment outcome. We hypoth-
esized that as residential treatment is designed for veterans with
more impairing symptoms and complex psychosocial stressors,
residential patients would endorse greater PTSD and depression
symptom severity at pretreatment compared with outpatients. We
also hypothesized that veterans in both programs would benefit
from treatment as evidenced by reductions in PTSD and depression symptoms. Due to the lack of preexisting research, analyses
comparing treatment outcome trajectory across groups were exploratory.
Participants included 992 male and female veterans admitted to
either the outpatient (n ⫽ 514) or residential (n ⫽ 478) PTSD
programs at a midwestern Veterans Affairs medical center between 2007 and 2011. All participants met full diagnostic criteria
for PTSD as set forth in the Diagnostic and Statistical Manual of
Mental Disorders (4th ed., text rev.; DSM–IV–TR; American Psychiatric Association, 2000) and as assessed by the ClinicianAdministered PTSD Scale (CAPS; Blake et al., 1995). All participants attended at least one, but no more than 15, sessions of
individual CPT in either the outpatient or residential treatment
program. A referral was the only eligibility criterion and was used
to determine whether treatment was received in the outpatient or
residential treatment program. To be eligible for the outpatient and
residential PTSD programs, veterans could either be self-referred
or referred by a care provider within the VA Healthcare System.
Eligibility for the residential program did not require veterans to
have comorbid diagnoses or to have received (or not responded to)
prior treatment; however, veterans did commonly present with
comorbid conditions and prior treatment experience.
Participants were excluded from the study for the following
reasons: active substance dependence, current unmanaged psychosis, a significant interfering medical condition (e.g., unmedicated
seizure disorder), or the presence of suicidal or homicidal intentions (individuals with ideation but not intention were included).
Eleven participants were excluded as they did not attend at least
one CPT session after intake, and 14 were excluded because they
attended 16 or more sessions. Seventy-five individuals presented
to the clinic for at least two unique treatment periods (e.g., two
outpatient treatment periods, two residential treatment periods, or
one treatment period of each outpatient and residential). These
cases were handled by including data from the first treatment
period (whichever type of treatment was received first) and eliminating data from any subsequent treatment periods. This method
was selected as a conservative approach to reduce the effects of
previous treatment exposure.
The mean age for the outpatient sample was 43.48 years (SD ⫽
14.59), and the mean years of education was 13.21 (SD ⫽ 1.84).
For the residential sample, the mean age was 47.87 years (SD ⫽
10.96), and the mean years of education was 13.51 (SD ⫽ 2.43).
Additional demographic information for the outpatient and residential treatment samples can be found in Table 1.
The pre- and posttreatment assessment battery administered for
both outpatient and residential programs included structured clin-
OUTPATIENT AND RESIDENTIAL PTSD TREATMENT
Demographics Across Outpatient and Residential Groups
Adult sexual assault
Childhood sexual abuse
Sudden unexpected death
(n ⫽ 514)
(n ⫽ 478)
Note. OEF/OIF/OND ⫽ Operation Enduring Freedom, Operation Iraqi
Freedom/Operation New Dawn.
Other service eras include Korean War, between Korea and Vietnam,
Somalia, Kosovo, World War II, Bosnia, and Lebanon. b The five most
frequently reported types of index trauma.
ical interviews and self-report measures completed in approximately 2–3 hr. The measures used in the current study were
derived from these larger pre- and posttreatment batteries. The
summary scores used for the present analyses were obtained from
chart review, and as a result, item-level data were unavailable.
Clinician-Administered PTSD Scale (CAPS; Blake et al.,
1995). The CAPS is a structured clinical interview for the assessment of PTSD based on the DSM–IV–TR. A PTSD symptom,
as measured on the CAPS, counted toward meeting diagnostic
criteria if the frequency (on a 0 – 4 scale) was rated as at least 1
(symptoms occur at least monthly) and intensity (on a 0 – 4 scale)
was rated as at least 2 (moderate distress). This scoring procedure
was used to determine PTSD diagnostic status based on response
to CAPS items (Weathers, Ruscio, & Keane, 1999). A CAPS total
severity score was also computed by summing ratings of the
frequency and intensity of each PTSD symptom (higher scores
indicate greater distress or impairment). Prior research with the
CAPS has suggested that the interview is reliable and valid (Blake
et al., 1995; Weathers et al., 1999), with excellent psychometric
properties among veteran samples (Weathers, Keane, & Davidson,
2001). PTSD diagnostic status, as determined by the CAPS, was
used for inclusion criteria, while the CAPS total severity scores at
pre- and posttreatment were used for the current study analyses. At
posttreatment, the period of assessment used for the CAPS was the
previous week, which was intended for repeated assessment over
a reasonably brief time interval (Weathers et al., 2001). The
1-week time frame at posttreatment was selected to ensure that the
assessment period did not significantly overlap with the treatment
period in either program.
Structured Clinical Interview for DSM–IV Axis I Disorders
(SCID–I; First, Spitzer, Gibbon, & Williams, 1996). The
SCID–I is a semistructured interview designed to assess the presence of Axis I symptoms and disorders. The SCID–I was used in
the present study to detect current and lifetime mood disorders,
substance use disorders, and anxiety disorders other than PTSD,
and to screen for psychotic symptoms.
In the current study, pretreatment SCID–I diagnoses were used for
descriptive purposes. Among veterans in the residential program, the
most frequent current psychiatric conditions comorbid with PTSD, as
diagnosed by the SCID–I, were major depressive disorder (66.7%),
panic disorder (13.7%), bipolar I disorder (8.5%), social phobia
(7.3%), and specific phobia (6.3%). Among veterans in the outpatient
program, the most frequent current comorbid conditions were major
depressive disorder (56.5%), panic disorder (8.2%), social phobia
(4.5%), alcohol abuse (3.7%), and specific phobia (3.4%).
PTSD Checklist–Stressor-Specific Version (PCL–S; Weathers, Huska, & Keane, 1991). The PCL–S is a widely used 17item self-report measure of PTSD symptomatology that corresponds
with diagnostic criteria in the DSM–IV–TR. The PCL–S was used to
assess PTSD symptom severity in relation to participants’ reported
index trauma (i.e., worst identified trauma or the trauma that they
reported as affecting them the most). PCL–S scale items are rated on
a 5-point Likert-type scale ranging from 1 (not at all) to 5 (extremely),
with total scores ranging from 17 to 85 (higher scores indicate greater
distress). The PCL–S has demonstrated reliability and validity across
various populations (Blanchard, Jones-Alexander, Buckley, & Forneris, 1996), as well as sensitivity to change over the course of
psychological treatment (McDonald & Calhoun, 2010). PCL–S total
scores from pre- and posttreatment were used for the current study
analyses. Consistent with the CAPS assessment, the time period used
for the PCL–S at posttreatment was the previous week.
Beck Depression Inventory-II (BDI–II; Beck, Steer, &
Brown, 1996). The BDI–II is a well-established 21-item selfreport measure used to assess the presence and severity of depressive symptoms. Items are rated on a 4-point Likert-type scale
ranging from 0 to 3. Ratings are summed to yield a total severity
index, with higher scores indicating more severe depressive symptoms. Research has demonstrated that the BDI–II has strong psychometric properties, including internal reliability, test–retest reliability, and convergent validity (Beck et al., 1996). For the current
study, BDI–II total scores at pre- and posttreatment were used for
the analyses and assessed symptoms over the past 2 weeks.
All data were collected as part of routine clinical care at a VA
PTSD specialty clinic in the Midwest. A waiver of written consent
based on archival data was obtained from a university institutional
WALTER, VARKOVITZKY, OWENS, LEWIS, AND CHARD
review board and the VA Research and Development Office.
During admission into both outpatient and residential treatment
programs, veterans were administered a series of assessment instruments to determine their pretreatment diagnostic status and
functioning. All assessments were conducted by clinicians with
extensive training and supervision in the provision of psychological tests. Diagnostic interviews and treatment sessions were not
recorded in the clinic, precluding assessment of interrater reliability.
Following the pretreatment assessment, veterans received either
the 12-session individual standard CPT protocol (Resick et al.,
2010) on an outpatient basis or the 12-session combined group and
individual CPT protocol (Chard et al., 2009) within the residential
PTSD treatment program. Prior work has demonstrated similar
effect sizes for treatment outcome using various CPT protocols,
including individual only and combined group and individual
formats (see Chard, 2005; Resick, Nishith, Weaver, Astin, &
Feuer, 2002). Veterans who received residential treatment were
grouped in cohorts of approximately 10 people.
Within both outpatient and residential programs, CPT treatment
consisted of three stages: (a) an initial stage consisting of examination of the meaning of the trauma and the relationship between
thoughts and emotions and identification of “stuck points” (i.e.,
disruptive cognitions about trauma that exacerbate PTSD symptoms); (b) a second stage that included the writing of a trauma
account(s) to facilitate activation and normalizing natural emotions
related to trauma, and further identification of stuck points; and (c)
a final stage that focused on challenging stuck points and replacing
them with more adaptive cognitions. Throughout treatment, cognitive techniques (including Socratic dialogue) were employed to
facilitate further evaluation of potentially maladaptive beliefs.
During the final CPT session, veterans were again asked to discuss
the meaning of their trauma in order to highlight changes in their
beliefs that may have emerged through the course of treatment.
CPT was provided by licensed clinicians (i.e., nurse practitioners, social workers, psychiatrists, psychologists) or by trainees
(i.e., psychiatry residents or psychology interns) supervised by a
licensed clinician. Although providers varied in their ...
Purchase answer to see full