The following hyperlink is a really helpful online book for PTSD.

Click Here

NPR Radio Segment

Treating Iraqi Children for PTSD

* What are some unique aspects of PTSD in children exposed to combat in Iraq?

* In what ways are the Iraqi children experiencing similar traumatic events as in the United States?

Meta-Analyses on Treatment of PTSD

Journal of Traumatic Stress, Vol. 11, No. 3, 1998
Effects of Psychotherapeutic Treatments for
PTSD: A Meta-Analysis of Controlled Clinical
Jeffrey J. Sherman1,2

This meta-analysis synthesized the results from controlled, clinical trials of psychotherapeutic
treatments for posttraumatic stress disorder (PTSD). Psychotherapeutic
modalities included behavioral, cognitive, and psychodynamic
treatments, in group and individual settings. Participants in the studies included
combat veterans from the Vietnam and Lebanon Wars, crime-related victims,
and severe bereavement sufferers. The impact of psychotherapy on PTSD and
psychiatric symptomatology was significant, d = .52, r = .25, when measured
immediately after treatments were administered. Similarly, there was no decay
in the effect of treatment at follow-up, d = .64, r = .31. Moreover, for target
symptoms of PTSD and general psychological symptoms (intrusion, avoidance,
hyperarousal, anxiety, and depression), effect sizes were significant, ranging
from r's of .2-.49. Results suggest substantial promise for improving psychological
health and decreasing related symptoms for those suffering from PTSD.
KEY WORDS: PTSD; treatment; outcome; control.
Posttraumatic stress disorder (PTSD) affects up to 12% of the U.S.
population (Breslau, Davis, Andreski, & Peterson, 1991; Davidson, Hughes,
Blazer, & George, 1991; Helzer, Robbins, & McEvoy, 1987; Kessler, Sonnega,
Bromet, Hughes, & Nelson, 1995; Resnick, Kilpatrick, Dansky, Saunders,
& Best, 1993), with prevalence estimates varying based on the
methods used for detection of PTSD and the populations sampled (Ameri-
1University of Washington, College of Medicine, Department of Psychiatry and Behavioral
Science, Box 356560, Seattle, Washington 98195-6560.
2Correspondence should be addressed to Jeffrey J. Sherman, c/o Dr. Charles R. Carlson,
Department of Psychology, University of Kentucky, 112 Kastle Hall, Lexington, Kentucky
0894-9867/98/0700-0413$15.00/1 © 1998 International Society for Traumatic Stress Studies
can Psychiatric Association [APA], 1994; Davidson et al., 1996; Friedman,
Schnurr, & McDonagh-Coyle, 1994; Kilpatrick, Saunders, Best, & Von,
1987; Norris, 1992; Resnick et al., 1993). While lower prevalence estimates
are found in studies employing conservative methodologies that focus on
community samples (Helzer et al., 1987), much higher rates are found in
studies focusing on at risk populations such as veterans or motor vehicle
accident or assault victims (Blanchard, Hickling, Taylor, & Loos, 1995;
Breslau et al., 1991; Kilpatrick et al., 1987).
According to the Diagnostic and Statistical Manual of Mental Disorders
(DSM-IV; APA, 1994), criteria for diagnosis require a recognizable
traumatic stimulus followed by episodes of recurrent and intrusive recollections
or feelings, avoidance of stimuli associated with the trauma, and
persistent symptoms of increased arousal. The traumatic stimulus was originally
used for combat-related events, but is now frequently associated with
physical assault, abuse, incarceration, automobile accidents and life threatening
illness (APA, 1994).
The psychological sequelae of traumatic events have been studied for
over 100 years (Putnam, 1881; Salmon, 1919), and a substantial amount of
interest and research has focused on the treatment of these psychological
sequelae. Some of the early treatment outcome research lacked the methodological
rigor and sophistication (Foa & Meadows, 1997, McFarlane,
1994) that is now deemed important for treatment outcome studies. Foa
and Meadows (1997) clearly delineate the criteria for such research, which
include use of reliable and valid measures, unbiased assignment to treatment
conditions, and use of clearly defined target symptoms. In a review
of theoretical and empirical issues facing investigators in the treatment of
PTSD, Fairbank and Nicholson (1987) stated that no controlled studies had
yet appeared in the literature. In 1992, Solomon and colleagues reviewed
pharmacotherapeutic and psychotherapeutic treatments for PTSD and
found six randomized, controlled clinical trials for psychotherapeutic treatments
in the literature. In the last decade, however, there has been a dramatic
increase of interest in the treatment for PTSD and a comparable
increase in controlled treatment studies. For example, a PsycLit search from
January 1974 through March 1996, performed for this review, revealed over
608 published studies containing the key words "PTSD" and "treatment."
Still, only a small percentage of these were randomized controlled clinical
Individual psychotherapy is the most commonly proposed treatment
for PTSD (Fairbank & Nicholson, 1987), and a wide variety of individual
treatments has been advocated. Behavioral interventions such as flooding
or systematic desensitization emphasize the importance of extinguishing the
anxiety associated with intrusive symptoms and reducing avoidant behav-
414 Sherman
iors. Flooding exposes the person either in vivo or imaginally to the aversive
stimuli and thus extinguishes the fear response. In systematic desensitization,
the pairing of relaxation and exposure results in a reciprocal inhibition
of the fear response. Many of these treatments are based on modifications
of Mowrer's (1960) two-factor learning theory as it applies to PTSD
(Boudewyns, 1996; Keane, Zimmering, & Caddell, 1985), where PTSD is
seen as a condition resulting from both classical and instrumental conditioning.
Another exposure-based strategy combines saccadic eye movements
with an exposure component. During eye movement desensitization and
reprocessing (EMDR), patients perform rhythmic, multisaccadic eye movements
while concentrating on their own description of a traumatic memory
(Jensen, 1994; Shapiro, 1989,1991; Silver, Brooks, & Obenchain, 1995; Wilson,
Becker, & Tinker, 1995). While the eye movements distinguish this
strategy, the technique shares many characteristics with desensitization or
flooding (Boudewyns, 1996). Importantly, recent data comparing EMDR
with and without eye movements (Boudewyns & Hyer, 1996; Pitman et al.,
1996) suggested that EMD/R without eye movements, essentially a desensitization
or flooding technique, was equivalent to EMDR with eye movements.
Cognitive behavioral interventions focus on the acquisition and maintenance
of coping skills and on the assumptions and misattributions that
underlie the patient's behavior, expectations, and appraisal of events (Foa
& Rothbaum, 1989; Foa, Rothbaum, Riggs, & Murdock, 1991; Foa,
Steketee, & Rothbaum, 1989). This view suggests that the underlying problem
is not solely anxiety and the maintenance of the fear response, but
that additional difficulties also exist in the overactive cognitive patterns or
schemata that interpret external and internal experiences as dangerous. The
dysfunctional thinking may lead to depression, anxiety, and PTSD (Foa,
Rothbaum, & Steketee, 1993). Therefore, cognitive theory as applied to
the treatment of PTSD (Veronen & Kilpatrick, 1983) focuses on two primary
processes: (1) changing a person's cognitive appraisal of the traumatic
event or changing the process by which an individual attaches meaning to
an event and (2) changing a person's attribution of the event.
Veronen and Kilpatrick (1983) define attribution as the human need
to understand our experiences and continually interpret events and search
for explanations about why things happen. Cognitive therapeutic techniques
help the person identify and correct distorted schemata and attributions
by training the patient to monitor negative thoughts, identify cognitive distortions,
and substitute reality-oriented interpretations (Meichenbaum,
1985). Stress inoculation training (SIT, Foa et al., 1991; Veronen & Kilpatrick,
1983), a widely used stress management technique, focuses on treat-
Meta-Analysis of PTSD 415
ing the anxiety associated with PTSD by teaching coping strategies such as
relaxation training, problem solving, anger control, and self-monitoring
(Meichenbaum, 1985). The goals of SIT focus on increasing patients'
awareness of conditioned stimuli and improving early detection of anxietyprovoking
cues to facilitate the use of coping skills early in the stress response
to reduce anxiety (Litz & Roemer, 1996).
In psychodynamic therapy, the traumatized individual must reconcile
the occurrence of the traumatic event and its meaning with his or her concept
of the self and the world (Horowitz, 1974; Horowitz & Kaltreider,
1980). The patient's response to stress will vacillate between two phases of
denial and intrusive symptoms. Avoidance, denial and emotional numbness
are the results of defensive overcontrol (Horowitz, 1974). When this overcontrol
fails or is overwhelmed, this phase gives way to intrusive thoughts,
flashbacks, or nightmares which are seen as the result of failed defense
mechanisms. The nature of the therapy is dependent on the phase during
which the patient presents. The primary goal of psychodynamic treatment
is integration of the traumatic experience by means of therapeutic reexperiencing
in a supportive environment (Horowitz, 1974). Insight into the
conscious and unconscious meaning of the symptoms can help the patient
master the trauma and restore functioning. The therapist must proceed with
the difficult challenge of developing a therapeutic alliance and handling
transference and countertransference issues (Lindy, 1989). While doing so,
the therapist must recognize that every person brings to any serious life
event his or her own history and existential interpretation of the traumatic
moment (Horowitz, 1977). In treatment, the therapist must consider the
person's self-image, strategies for coping, and defense mechanisms against
intrapsychic and interpersonal threats. The treatment recognizes the influence
of dispositional variables on the response to both the trauma and
therapy. More thorough explanations of these and other treatment strategies
such as group interventions, prevention programs, and anger management
training can be found in additional review and original papers (Allen
& Bloom, 1994; Boudewyns, 1996; Foa et al., 1989; Frueh, Turner, & Beidel,
1995; Shalev, Bonne, & Eth, 1996; Solomon, Gerrity, & Muff, 1992).
When treating a patient suffering from PTSD, the therapist has a variety
of treatment modalities from which to choose, but the absence of a
large and sophisticated outcome literature on PTSD is an important concern
(Fairbank & Nicholson, 1987; Foa & Rothbaum, 1989; Frueh, Turner,
Beidel, Mirabella, & Jones, 1996; McFarlane, 1994). Treatment of PTSD
is a challenge to the therapist because one must attend to and contain
severe distress and suffering. It is particularly difficult to do so if there is
uncertainty about the most appropriate and effective interventions. This
concern has led to an outcry for more work that meets the criteria for
416 Sherman
good outcome research (Foa et al., 1993; McFarlane, 1989,1994; Solomon,
Gerrity et al., 1992). In response, several teams of experimenters have embarked
on and published controlled clinical trials of psychotherapy outcome
for PTSD. Of these and the many noncontrolled studies, a variety of treatments
including behavioral, psychodynamic, and group therapies have been
found to be effective. However, several studies using these treatments report
null findings (Boudewyns & Hyer, 1990; Funari, Piekarski, & Sherwood,
1991; Hyer, Woods, Bruno, & Boudewyns, 1989), or at times,
negative findings where the treatment group worsens in psychological functioning
(Pitman, Altaian, Greenwald, & Longpre, 1991; Solomon, Shalev
et al., 1992).
Several qualitative reviews have been conducted on the PTSD treatment
literature (Boudewyns, 1996; Frueh et al., 1995; McFarlane, 1989,
1994; Shalev et al., 1996; Solomon, Gerrity, et al., 1992), but none have
done so using a meta-analytic approach. Meta-analysis represents a systematic
approach toward integrating findings from a research domain through
statistical analysis of individual studies (Smith & Glass, 1977; Strube &
Hartmann, 1983). The purpose of this article was to review the empirical
evidence for the efficacy of psychotherapeutic treatments for PTSD using
a meta-analytic approach.
Qualitative reviews including case reports, studies without control
groups, and nonrandomized trials concluded that several different types of
treatment protocols reduce both PTSD symptoms and general psychiatric
symptomatology and improve quality of life (McFarlane, 1989, 1994). Recent
reviews on various behavioral, exposure-based strategies (Boudewyns,
1996; Frueh et al., 1995) concluded that these treatments were effective in
reducing typical features of PTSD such as intrusive and avoidant symptoms
and physiological reactivity to stimuli associated with the trauma. After reviewing
five randomized, controlled clinical trials on behavioral treatment
for PTSD, Solomon, Gerrity, et al. (1992) concurred that such strategies
effectively reduced intrusive symptoms indicative of PTSD. Blake (1993)
reviewed eight studies and compared various psychotherapeutic treatments.
This review also concluded that therapies were consistently effective in reducing
the "positive" symptoms of PTSD (i.e., psychophysiological arousal,
intrusive thoughts, nightmares, and anger), but that there was no clear reduction
in "negative" symptoms of PTSD (i.e., numbing, alienation, and
restricted affect).
In contrast, Shalev et al. (1996) reviewed various treatment modalities
and designs including case reports, and controlled and uncontrolled clinical
trials. They concluded that the magnitude of the result was often limited
and that remission was rarely achieved. They also included a framework
for identifying target symptoms and specific treatment modalities that might
Meta-Analysis of PTSD 417
be effectively combined to treat these target symptoms. These authors determined
from their review that a substantial alleviation of suffering did
occur with treatment, but that in controlled studies, the effect size was limited
and that although there was typically alleviation of depression, detachment
and anxiety, there was less of a reduction in the typical symptoms of
PTSD (intrusion, avoidance).
Such questions might be better addressed using a quantitative review
of both overall effects and specific effects on target symptoms such as depression,
anxiety, intrusion, and avoidance. The present study was designed
to answer these questions using meta-analytic techniques to review controlled
treatment outcome studies in which participants received a PTSD diagnosis
and which used objective measures administered before and after treatment
to monitor outcome. This approach made it possible to provide an estimate
of the overall effect size of psychotherapeutic treatments and to determine
the efficacy of these treatments for ameliorating specific symptoms.
Identification of Studies
The major sources for identifying the empirical studies used for the
meta-analysis were computerized databases including Psyclit, ERIC,
Medline, Cinahl Nursing Database, Dissertation Abstracts, and the PILOTS
Traumatic Stress Database from Dartmouth College. These were
searched using the following keywords and combinations of keywords:
"posttraumatic stress disorder," "PTSD," "therapy," "treatment," "outcome,"
and "control." Reference lists from existing reviews and empirical
studies were also consulted. One criticism of meta-analytic techniques is a
bias toward analyzing only significant, published findings. Tb include such
"fugitive literature" (Rosenthal, 1995, p. 184), attempts were made to include
unpublished studies in the meta-analysis by contacting 25 authors in
the field and requesting any unpublished work. Replies via e-mail, U.S.
mail, or telephone from 13 of these authors resulted in receipt of seven
additional studies, two of which were included in the analysis (Boudewyns
& Hyer, 1996; Foa, Freund et al., 1994).
Inclusion Criteria
Only clinical trials performed predominantly on participants who met
threshold DSM-III, III-R, or IV criteria for PTSD were included in the
418 Sherman
meta-analysis. Analyses were also restricted to studies that used a comparison
group, provided inferential statistics with which to calculate relevant
effects sizes, and which used objective measures of outcome taken before
and after treatment. Eleven studies included participants meeting the criteria
based on combat-related events, and six studies included participants
meeting the criteria based on noncombat-related trauma (crime-related or
bereavement). The meta-analysis did not include studies on general care
or counseling programs for victimized or traumatized people unless the participants
met criteria for PTSD.
Coding System
All studies were coded based on the following criteria: (a) author, year
of study, (b) total N, (c) type of participants (combat veteran vs. noncombat),
(d) type of intervention vs. type of control, (e) design and measures
used to calculate effect sizes at posttreatment and follow-up, and (f) length
of follow-up.
Two strategies were used for comparing effect sizes across the studies.
First, effect sizes were averaged across all dependent measures for each
study. These "supervariables" (Rosenthal, 1984, p. 36) were weighted
equally in determination of an overall effect for all of the studies. One
advantage of averaging a single effect size is that it avoids spurious inflation
of effect sizes from studies in which a number of dependent measures are
used. A second advantage is that it provides a measure of overall change
across several PTSD symptom categories, social functioning, and psychological
adjustment. One disadvantage, however, is that given the multidimensional
nature of PTSD, aggregating a single effect size based on these
symptom clusters and social functioning results in the loss of a theoretically
meaningful outcome. Thus, a supplementary strategy was adopted in order
to provide a more specific measure of treatment outcome. Similar to methods
used by Wolraich, Wilson, and White (1995), dependent measures were
categorized for target symptoms relevant to the disorder (i.e., intrusion,
avoidance, arousal, depression, and anxiety). When the data were available,
effect sizes were then calculated from the available studies based on these
target symptoms.
Meta-Analysis of PTSD 419
Type of Dependent Variables
Effect sizes were calculated for a composite effect from each study.
This effect combined all of the dependent measures evaluated in each
study. In addition, five categories of dependent measures were developed
and effect sizes were estimated for each symptom category. The intrusion
category included measures from the PTSD Structured Interview (SIPTSD;
Davidson, et al., 1989), the intrusion scale of the Impact of Event
Scale (IBS; Horowitz et al., 1979), clinical interviewer ratings of intrusive
experiences, the sleep disturbances subscale of the Modified Vietnam Experiences
Questionnaire (MVEQ; Cooper & Clum, 1989), and other selfreports
of nightmares or flashbacks. The hyperarousal category included
self-reports, interviewer ratings and/or psychophysiological test results of
arousal, anxiety, EMG activity, heart rate, and skin conductance in response
to stressful images. This category also included scores on the arousal
subscale of the PTSD Symptom Scale (PSS; Foa et al., 1993). The avoidance
category included interviewer ratings of avoidance and scores on the
avoidance subscales of the PSS and IES. The depression category included
scores from the Hamilton Rating Scale for Depression (Hamilton, 1960),
Beck Depression Inventory (BDI; Beck et al., 1961), and the depression
subscale of the Symptom Checklist 90-Revised (SCL-90-R; Derogatis,
1977). The anxiety category included scores from the state version of the
State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, & Luschene,
1970) and the anxiety subscale of the SCL-90-R.
While none of the studies in the meta-analysis made any claims that
long-term personality change would occur with treatment, two studies
(Brom, Kleber, & Defares, 1989; Keane, Fairbank, Caddell, & Zimering,
1989) included global measures of personality functioning. Brom, et al.
(1989) stated that these data were reported only because of significant, but
not predicted change on several subscales. Since the intent of few PTSD
treatments is global personality change, and since this study does not seek
to evaluate such change, neither global personality nor trait measures are
included in the analyses. Subscales on the MMPI or other personality measures
are not included as dependent measures unless specifically associated
with PTSD (i.e., PTSD subscale of the MMPI).
Calculation of Effect Sizes
When the data were available, effect sizes were calculated for treatment
effects immediately after treatment (post) and from 3 months to 2
years after treatment (follow-up). If means and standard deviations were
420 Sherman
available for treatment and no-treatment control conditions, effect sizes
were calculated by subtracting the treatment group mean from the no treatment
or control group mean and dividing the result by the pooled standard
deviation (Glass, McGaw, & Smith, 1981; Johnson, 1989). When insufficient
information was supplied for calculating the effect size in such a way,
i.e., differences reported in terms of nonparametric tests, t statistics or F
statistics formulas applied by Johnson (1989) were used to obtain estimates
of the effect size (ES). When results were reported as nonsignificant and
insufficient data were reported to calculate an ES, the ES was conservatively
estimated as an effect size of 0. Only studies providing comparisons
between treatment and control groups were used to estimate the post treatment
effect size. At follow-up, effect sizes were estimated for studies providing
either comparisons between treatment and control conditions or
between pre-treatment and follow-up periods for a treatment group.
On some measures, a negative treatment-control difference reflected
that treatment improved patient status by reducing symptoms (i.e., lower
intrusion or avoidance scores for treatment group). On other measures, a
positive treatment-control condition reflected that treatment improved psychosocial
functioning (e.g., increased self-efficacy). Therefore, the direction
of effect was standardized across effect sizes so that positive effects indicated
a desirable effect of treatment, and negative effects indicated an undesirable
effect of treatment.
The effect size calculated was g, defined as the difference between the
means of the intervention and comparison group, divided by the pooled
standard deviation (Cohen, 1977; Johnson, 1989). The g's were converted
to d's by correcting them for bias (i.e., g's may overestimate the population
effect size for small samples) (Johnson, 1989). For convenience, effect sizes
are expressed as both d's and r's. Finally, a homogeneity statistic, Q, was
used to determine whether the effect sizes were consistent across all studies
(Hedges & Olkin, 1985; Johnson, 1989).
Table 1 presents the study characteristics and effect sizes for each
study. The combined sample represents 690 participants from 17 studies.
Of these, 11 studies used a total of 281 Vietnam veterans and 81 Israeli
combat veterans from the Lebanon war, four studies used 180 female
rape/assault victims, and two studies (Brom et al., 1989; Vaughan et al.,
1994) used 148 subjects (111 females) who were victims of violent crimes,
motor vehicle accidents (MVA), child abuse, or suffered PTSD from the
traumatic loss of a loved one.
Meta-Analysis of PTSD 421
422 Sherman
Meta-Analysis of PTSD 423
The predominant treatment modality in the studies was behavioral or
cognitive behavioral. In addition to a cognitive behavioral treatment, one
study (Brom et al, 1989) used psychodynamic and hypnotherapy. Less traditional
treatments included the Koach program (Solomon, Shalev et al.,
1992), anger treatment (Chemtob, Novaco, Hamada, & Gross, 1997),
EMDR (Boudewyns & Hyer, 1996; Jensen, 1994; Vaughan et al., 1994),
adventure-based activities and psychodrama (Ragsdale, Cox, Finn, & Eisler,
1996), and the Coatesville PTSD treatment program (Hammarberg & Silver,
The Koach program was a month-long residential treatment program
that included self-help groups, milieu therapy, family therapy, supportive
counseling, cognitive behavioral education groups, relaxation and in vivo
desensitization. Desensitization included living in tents, wearing military
uniforms, military combat and weapons training, mountain climbing, and
other military exercises throughout the treatment period. In addition to
routine care, participants in the Chemtob et al. (1997) study received manual-
guided individual psychotherapy sessions for anger management. Sessions
focused on anger self-monitoring exercises, relaxation, cognitive
restructuring of anger by modifying appraisals or altering attentional focus,
training in behavioral coping, assertiveness, and communication skills, and
role-playing devised to arouse anger. Ragsdale et al. (1996) treated patients
with a 26-day inpatient program. Patients engaged in activities designed to
increase trust by requiring mutual cooperation such as traversing a 12-ft
wall and falling backwards off of a platform to be caught by other group
members. Group sessions involved PTSD education, grief work, processing
experiences during the adventure-based activities, and psychodrama, a
reenactment through role play of the original traumatic event, Hammarberg
& Silver (1994) treated veterans in a 90-day inpatient, PTSD treatment
program, using multiple treatment modalities. Treatments were usually in
a group format and consisted of patient education, relaxation training,
dreamwork, and peer groups that were run without therapist assistance.
The effect sizes displayed in Table 1 represent an average effect size
for each study across all the dependent measures and for each type of treatment
modality, within each study. Thus, this composite effect represents
the extent to which treatment contributes to eliminating PTSD, reducing
PTSD symptoms (intrusion, avoidance, hyperarousal), and reducing psychiatric
symptomatology (depression, anxiety). These effect sizes are displayed
on the right side of Table 1 at posttreatment (immediately after treatment
for all studies) and at the first follow-up for each study. The effect sizes
immediately after treatment ranged from to .03 to 8.4. The overall effect
was significant, d = .54, r = .26, with a non-zero, 95% confidence interval
suggesting that the true effect lies between .39 and .68. One study (Penis-
424 Sherman
ton, 1986) contributed to a significant test of heterogeneity, Qw(23) =
37.24, p < .05. Following the methods of Hedges and Olkin (1985), that
study was removed from further analysis. With this outlier removed, the
overall effect was significant, d = .52, r = .25, with a 95% confidence interval
between .37 and .67. The effect sizes for the remaining studies were
homogeneous, Qw(22) = 11.94,p = .96. This suggests that the mean from
the effect sizes can be presumed to estimate a single underlying population
effect size.
Twelve studies provided data for follow-up occurring between 3 months
and 2 years after the end of treatment (Table 1). The effect sizes at the
first follow-up period ranged from -.25 to +1.69. The overall effect was
significant, d = .53, r = .25, with a non-zero, 95% confidence interval suggesting
that the true effect lies between .37 and .69. One study (Solomon,
Shalev et al., 1992) contributed to a significant test of homogeneity, Qw(\8)
= 36.61, p < .05. With this outlier removed, the overall effect was significant,
d = .64, r = .31, with a 95% confidence interval between .47 and
.81. The effect sizes for the remaining studies were homogeneous, Qw(17)
= 22.59, p = .21.
The effect sizes presented in Table 2 represent the effect of psychotherapy
treatment on specific target symptoms. The effect sizes immediately
after treatment ranged from .53 for anxiety to .79 for avoidance symptoms.
All effect sizes are significant and nonzero and are presented after correction
for heterogeneity. The only effect size with significant variability
(-.15-+8.4) was hyperarousal. This variability may reflect the wide variety
of assessment techniques used to assess hyperarousal such as EMG, HR,
SCL, SUDS, and self-reports.
The overall effect, d = .52; r = .25, achieved from examination of the
17 studies (690 participants) demonstrates strong empirical support for the
Meta-Analysis of PTSD 425
Table 2. Effect Sizes Across Treatments by Target Variables
No. of
No. of
efficacy of psychotherapeutic interventions for the treatment of combat,
crime, and traumatic bereavement-related PTSD. This indicates that psychotherapeutic
treatments have positive effects on characteristic PTSD
symptoms and symptoms of depression and anxiety. In order to put magnitudes
of effect size into perspective, Cohen (1977) provided guidelines
for interpretations of effect sizes and defined conventional magnitudes corresponding
to small (r = .1), medium (r = .3), and large (r = .5) effect
sizes. From this perspective, an r of .25 corresponds to a moderate correlation
between treatment and outcome. Similarly, this effect size reflects
an improvement in the treatment group of over one-half of a standard deviation
greater than that for the comparison group. Further, these effect
sizes were relatively consistent across the studies reviewed. This is likely
due to the similarity in research design between the controlled treatment
outcome studies used in the analysis.
While the effect sizes observed in this study are consistent with other
studies of psychotherapy outcome (Clum, Clum, & Surls, 1993; Hunter &
Schmidt, 1990; Smith & Glass, 1977), it is also important to evaluate the
practical usefulness and clinical utility of such an effect. Rosenthal (1984)
recommends using the binomial effect size display (BESD) for an intuitively
appealing interpretation of an average effect. The BESD represents an effect
size in terms of overall success rate for the treatment group. The BESD
can be interpreted as the estimated difference in success probability between
the treatment and control groups. Thus, a correlation coefficient of
.25 corresponds to increasing a success rate from 38% to 62% by means
of an intervention (Rosenthal, 1984; Wolfe & Cornell, 1986). This means
that the expected improvement rate for the psychotherapy condition was
62%, in contrast to an expected improvement rate of 38% for the nonintervention
group. A large proportion of the subjects used in the examined
studies were inpatients diagnosed with chronic PTSD. Considering the debilitating
nature of PTSD, the severity of symptoms endorsed by subjects
prior to participating in the studies, and the chronic course of PTSD, this
change in functioning is clinically significant and represents important practical
usefulness for psychotherapeutic treatment.
This meta-analysis also addressed some specific questions about the
treatment of PTSD. Shalev et al. (1996) stated that the magnitude of the
results in PTSD treatments are often limited, that remission is rarely
achieved, and that there is typically alleviation of depression, detachment
and anxiety, but less of a reduction in the typical symptoms of PTSD (intrusion
and avoidance). In contrast, these results suggest that the magnitude
of improvement due to psychotherapeutic treatments is moderate and
that these treatments are effective in reducing PTSD symptoms, depression,
and anxiety. Moreover, an analysis of effect sizes at follow-up suggests that
426 Sherman
these treatment effects are maintained even after discontinuation of treatment.
Further, six of the studies reviewed in the meta-analysis assessed patients
for diagnosis of PTSD after treatment and provided the necessary
data to compute percentage remission. Of the 103 patients with a full PTSD
diagnosis before treatment, 43% (44) improved to such a degree that they
did not meet diagnostic criteria after treatment. Considering the chronic
nature and severity of the disorder and the relative brevity of many of these
treatments, we would interpret this as a considerable percentage of subjects
who achieved improvement and remission. Finally, the meta-analysis of
treatment effects for specific symptoms was also revealing. Results of the
current review suggest that psychotherapeutic treatment was indeed effective
in reducing symptoms of depression and anxiety. Treatment was also
effective in reducing target symptoms of PTSD such as intrusion, arousal,
and avoidance. While treatment effects for all symptom categories were
significant, the greatest treatment effects were found for the intrusion and
avoidance categories of target symptoms.
While the results from this review lead to somewhat different conclusions
from those of Shalev et al. (1996), they are more consistent with conclusions
from other reviews that find that treatment effectively reduces the
hallmark symptoms of PTSD (Blake, 1993; Frueh et al., 1995; Solomon,
Gerrity et al., 1992). The present review focused only on controlled clinical
trials that adhered to a somewhat rigorous inclusion criteria. As such, this
review may have been more similar to the review by Solomon, Gerrity et
al. (1992) that focused only on randomized and controlled clinical trials.
All six studies reviewed in their report involved some form of exposure. In
contrast, the review by Shalev and colleagues provides an excellent and
broad overview of treatments for PTSD, but it includes studies of more
variable quality such as case reports, controlled and open trials. Moreover,
their review represents a broader array of treatment modalities (behavioral,
cognitive, psychodynamic, group, hypnosis, rehabilitation) in an outpatient
and inpatient environment than either the present or other reviews
(Boudewyns, 1996; Frueh et al., 1995; Solomon, Gerrity et al., 1992).
This review included published and unpublished studies on various
treatments of PTSD. However, the modality with the greatest representation
is the use of exposure techniques in a behavioral or cognitive
behavioral treatment setting. In fact, all but two studies used some form
of exposure technique (flooding, desensitization, implosion, EMDR, psychodrama)
with or without relaxation training. Thus, consistent with other
data (Boudewyns, 1996, Foa & Rothbaum, 1989; Frueh et al., 1995; Litz
& Roemer, 1996; Solomon, Gerrity et al., 1992), findings lend general support
for exposure-based therapies as effective treatment for both combat
and noncombat-related PTSD.
Mete-Analysis of PTSD 427
One study (Brom et al., 1989) included treatment groups receiving
either trauma desensitization, hypnotherapy, or psychodynamic therapy.
Foa et al. (1991, 1994) included treatment groups receiving stress inoculation
training, prolonged exposure, supportive counseling or a combination
of prolonged exposure and stress inoculation training. Two studies
(Boudewyns & Hyer, 1996; Vaughan et al., 1994) compared EMDR to
other forms of exposure or muscle relaxation therapy. If more controlled
studies had been available that compared a variety of treatments, a more
detailed analysis could have compared these treatments. Nevertheless, the
data from the Foa et al. (1991, 1994) studies suggest that effect sizes for
supportive counseling are lower than effect sizes for exposure therapies.
The data from the Brom et al. (1989) study suggest that hypnotherapy and
psychodynamic therapy are comparable to exposure therapy, and the data
from the Boudewyns and Hyer (1996) study suggest that effects from
EMDR are comparable to exposure therapy. In contrast, the data from
Vaughan et al. (1994) suggest that EMDR was superior to either an image
habituation treatment or a muscle relaxation treatment. This is consistent
with other research groups who have failed to find reliable differences between
therapeutic modalities for the treatment of PTSD (Frank et al., 1988;
Frank & Stewart, 1983, 1984).
A review of the treatment modalities most commonly used in the
treatment of PTSD suggests that all treatment approaches share a number
of common elements (McFarlane, 1994). All treatments share a primary
aim to help the patient develop a realistic appraisal of the threat experienced
during the trauma. All treatments seek to help the patient overcome
the cognitive and behavioral avoidance of internal cues and external reminders
of the trauma. All treatments also seek to assist individuals to work
through the meaning of their traumatic experience and to gain a sense of
mastery over their intrusive recollections. In one form or another, all treatments
focus on Horowitz' (1974) conceptualization suggesting that the goal
of psychotherapy is to help the patient work through the trauma experience
through gradual reexposure of the actual traumatic event and reinterpretation
of its meaning. Thus, while exposure-based techniques clearly have
the greatest representation when reviewing the literature on controlled trials,
many treatment modalities may have similar active therapeutic components.
This review and others (Shalev et al., 1996) call into question the efficacy
of using in vivo exposure-based techniques without sufficient patientbased
control of the exposure situation and therapeutic support. One
controlled, clinical trial reviewed in this article (Solomon, Shalev et al.,
1992) involved a residential treatment of 4 weeks exposure to military cues
including living in tents, wearing uniforms, mountain climbing, weapons,
428 Sherman
artillery, and hand-to-hand combat training for Lebanon War veterans.
Treated patients experienced considerable deterioration for up to 9 months
posttreatment. Similarly, Scurfield, Wong, and Zeerocah (1992) exposed
war veterans to helicopter rides and found increases in intrusive and painful
memories. While this study provided intense exposure, it may not have provided
sufficiently broad exposure to the full-range symptom-provoking cues.
In contrast, Richards, Lovell, and Marks (1994) found improvement with
a behavioral treatment program that included live exposure but also included
imaginal exposure. These results suggest the exercise of considerable
caution in the use of in vivo techniques for the treatment of PTSD.
One limitation of the study is that although the finding of significant
impact of treatments for PTSD is encouraging, this finding cannot be used
as full support for any single theoretical, therapeutic rationale. While most
of the studies used an exposure and/or relaxation technique, results for
alternative treatments were similar to those for the more traditional cognitive
behavioral techniques. Another concern is that eight studies in the
analysis used wait-list, no-treatment control groups, rather than a psychological
placebo control group. Since McConaghy (1990) demonstrated that
placebo treatments show superior outcomes to wait-list controls, it is possible
that the effect from studies using a wait-list control may be inflated.
Further, it is possible that the improvement outcomes in the treatment
groups could be due to nonspecific effects of therapy. On the other hand,
nine studies used control groups receiving standard, routine-care treatments
such as supportive counseling and individual and group therapy. Results
suggest that even when compared to control groups receiving standard care,
additional treatment is beneficial. As such, using active, psychological placebo
groups in future treatment studies would seem appropriate.
An important concern, especially with the limited number of studies
used in this analysis, is whether the combined effect sizes presented here
represent all of the research on this topic, including unpublished studies.
In an attempt to address this limitation, 25 authors in the field were contacted
in the hopes of gaining any unpublished work. These requests resulted
in seven responses and two studies that qualified for inclusion in
these analyses. In addition, in order to account for this "file-drawer" problem,
methods from Rosenthal (1979) were used to estimate the number of
new, filed, or unretrieved studies that would be required to bring the results
to nonsignificance. An estimated 57 nonsignificant studies would be required
to reduce the overall effect size to nonsignificance.
Certain treatment modalities were underrepresented or not represented
in this meta-analysis such as psychological debriefing after trauma
(Raphael, Meldrum, & McFarlane, 1996), anger management (Chemtob,
Hamada, Roitblat, & Muraoka, 1994; Gerlock, 1994), and treatment of
Meta-Analysis of PTSD 429
traumatized children (McNeil & Todd, 1986; Saigh, 1986, 1987a, 1987b).
This limitation reflects a dearth of controlled, clinical trials in these important
areas. For example, Raphael and colleagues (1996) discuss the merits
of debriefing programs in the prevention of PTSD and call for
randomized controlled trials in this area. Similarly, the present literature
review for this study and prior reviews (Boudewyns, 1996) found no controlled
treatment studies on children with PTSD, although children clearly
suffer from the disorder and may be particularly vulnerable (Boudewyns,
1996; Lyons, 1987). Further, results from the one study on anger management
(Chemtob et al., 1997), although based on a small sample, are quite
promising and suggest the need for continued study in this area. The current
and other reviews elucidate the need for more controlled trials in these
Miller, Kamenchenko, & Krasniasnski (1992) emphasize the importance
of making a clear differentiation with respect to the processing and
impact of traumas from man-made vs. natural traumas, physical vs. psychological
traumas, and traumas associated with domestic violence. Another
limitation to this meta-analysis is that we combined studies on PTSD
sufferers due to a variety of etiological traumas (war, crime, motor vehicle
accident, bereavement). While we recognize the potential limitations in aggregating
results across different types of traumatic experiences, insufficient
sample size precluded an analysis of type of etiology as a moderator variable.
It is important to note, however, that regardless of etiology, psychotherapeutic
treatments had similar efficacies.
This study reviewed and quantitatively synthesized results of psychotherapeutic
treatment modalities for the treatment of PTSD. The effect
sizes observed in this review were significant and encouraging. Psychotherapeutic
treatment reduces PTSD, and general psychiatric symptomatology,
and these effects are maintained even after termination of treatment. The
data support the continued use and investigation of psychotherapeutic modalities
for the treatment of PTSD.
I am very grateful to Drs. Judith Lyons, Thomas Miller, Charles
Carlson, James McCubbin, John Wilson and Jeffrey Okeson for their assistance
and reviews of earlier versions of this article. I thank Dr. Bryan
430 Sherman
Johnstone for his technical assistance and helpful discussions about metaanalytic
issues. I also thank three anonymous reviewers and the editor for
their very helpful comments and recommendations.
Allen, S. N, & Bloom, S. L. (1994). Group and family treatment of posttraumatic stress
disorder. Psychiatric Clinics of North America, 17, 425-437.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, D.C.: Author.
Beck, A. T, Ward, C. H., Mendelsohn, M., Mock, J., & Erbaugh, J. (1961). An inventory for
measuring depression. Archives of General Psychiatry, 4, 561-571.
Blake, D. D. (1993). Treatment outcome research on PTSD. Clinical Psychology Clinician
Newsletter, 3, 14-17.
Blanchard, E. B., Hickling, E. J., Taylor, A. E., & Loos, W. (1995). Psychiatric morbidity
associated with motor vehicle accidents. Journal of Nervous and Mental Diseases, 183,
Boudewyns, P. A. (1996). Posttraumatic stress disorder: Conceptualization and treatment. In
M. Hersen, R. M. Eisler & P. M. Miller (Eds.), Progress in behavior modification: Vol. 30
(pp. 165-189). Pacific Grove: Brooks/Cole Publishing Co.
*Boudewyns, P. A., & Hyer, L. (1990). Physiological response to combat memories and
preliminary treatment outcome in Vietnam veteran PTSD patients treated with direct
therapeutic exposure. Behavior Therapy, 21, 63-87.
*Boudewyns, P. A., & Hyer, L. (1996). Eye movement desensitization and reprocessing
(EMDR) as treatment for posttraumatic stress disorder (PTSD). Clinical Psychology and
Psychotherapy, 3, 185-195.
*Boudewyns, P. A., Hyer, L., Woods, M. G., Harrison, W. R., McCranie, E. (1990). PTSD
among Vietnam veterans: An early look at treatment outcome using direct therapeutic
exposure. Journal of Traumatic Stress, 3, 359-368.
Breslau, N., Davis, G. C., Andreski, P., & Peterson, E. (1991). Traumatic events and
posttraumatic stress disorder in an urban population of young adults. Archives of General
Psychiatry, 48, 216-222.
*Brom, D., Kleber, R., & Defares, P. B. (1989). Brief psychotherapy for posttraumatic stress
disorders. Journal of Consulting and Clinical Psychology, 57, 607-612.
Chemtob, C. M., Hamada, R. S., Roitblat, H. L., & Muraoka, M. Y. (1994). Anger, impulsivity,
and anger control in combat-related posttraumatic stress disorder. Journal of Consulting
and Clinical Psychology, 62, 827-832.
*Chemtob, C. M., Novaco, R. W., Hamada, R. S., & Gross, D. M. (1997). Cognitive-behavioral
treatment for severe anger in posttraumatic stress disorder. Journal of Consulting and
Clinical Psychology, 65, 184-189.
Clum, G. A., Clum, G. A., & Surls, R. (1993). A meta-analysis of treatments for panic disorder.
Journal of Consulting and Clinical Psychology, 61, 317-326.
Cohen, J. (1977). Statistical power analysis for the behavioral sciences (Revised ed.). New York:
Academic Press.
*Cooper, N. A., & Clum, G. A. (1989). Imaginal flooding as a supplementary treatment for
PTSD in combat veterans. A controlled study. Behavior Therapy, 20, 381-391.
Davidson, J., Foa, E. B., Blank, A. S., Brett, E. A., Fairbank, J., Green, B. L., Herman, J.
L., Keane, T., Kilpatrick, D. G., March, J. S., McNally, R. J., Pitman, R. K., Resnick, H.
S., & Rothbaum, B. O. (1996). Posttraumatic stress disorder. In T. A. Widiger, A. J.
3References marked with an asterisk indicate studies included in the meta-analysis.
Meta-Analysis of PTSD 431
Francis, H. A. Pincus, R. Ross, M. B. First, W. W. Davis (Eds.) DSM-IV Sourcebook (pp.
577-605). Washington DC: American Psychiatric Association.
Davidson, J. R. T., Hughes, D., Blazer, D. G., & George, L. K. (1991). Posttraumatic stress
disorder in the community: An epidemiological study. Psychological Medicine, 21, 713-721.
Davidson, J., Smith, R., & Kudler, H. (1989). Validity and reliability of the DSM-III criteria
for posttraumatic stress disorder: Experience with a structured interview. Journal of
Nervous and Mental Disease, 177, 336-341.
Derogatis, L. R. (1977). SCL 90: Administration, scoring, and procedure manual-I from the R
(revised) version. Baltimore: Johns Hopkins University School of Medicine.
Fairbank, J. A., & Nicholson, R. A. (1987). Theoretical and empirical issues in the treatment
of posttraumatic stress disorder in Vietnam veterans. Journal of Clinical Psychology, 43,
*Foa, E. B., Freund, B. E, Hembree, E., Franklin, M. E., Perry, K. J., Riggs, D. S., & Molnar,
C. (1994). Efficacy of short-term behavioral treatments of PTSD in sexual and non-sexual
assault victims. Presented at the annual meeting of the Association for the Advancement
of Behavior Therapy, San Diego, CA.
*Foa, E. B., Hearst-Ikeda, D., & Perry, K. J. (1995). Evaluation of a brief cognitive behavioral
program for the prevention of chronic PTSD in recent assault victims. Journal of
Consulting and Clinical Psychology, 63, 948-955.
Foa, E. B., & Meadows, E. A. (1997). Psychosocial treatments for posttraumatic stress
disorder: A critical review. Annual Review of Psychology, 48, 449-480.
Foa, E. B., Riggs, D. S., Dancu, C. V, & Rothbaum, B. O. (1993). Reliability and validity of
a brief instrument for assessing posttraumatic stress disorder. Journal of Traumatic Stress,
6, 459-473.
Foa, E. B., & Rothbaum, B. O. (1989). Behavioral psychotherapy for posttraumatic stress
disorder. Special issue: Behavioural psychotherapy in the 1990's. International Review of
Psychiatry, 1, 219-226.
*Foa, E. B., Rothbaum, B. O., Riggs, D. S., & Murdock, T. B. (1991). Treatment of
posttraumatic stress disorder in rape victims: A comparison between cognitive behavioral
procedures and counseling. Journal of Consulting and Clinical Psychology, 59, 715-723.
Foa, E. B., Rothbaum, B. O., & Steketee, G. S. (1993). Treatment of rape victims. Special
section: Rape. Journal of Interpersonal Violence, 8, 256-276.
Foa, E. B., Steketee, G. S., & Rothbaum, B. O. (1989). Behavioral cognitive conceptualizations
of posttraumatic stress disorder. Behavior Therapy, 20, 155-176.
Frank, E., Anderson, B., Stewart, B. D., Dancu, S C., Hughes, C., & West, D. (1988). Efficacy
of cognitive behavior therapy and systematic desensitization in the treatment of rape
trauma. Behavior Therapy, 19, 403-420.
Frank, E., & Stewart, B. D. (1983). Treating depression in victims of rape. The Clinical
Psychologist, 36, 95-98.
Frank, E., & Stewart, B. D. (1984). Depressive symptoms in rape victims: A revisit. Journal
of Affective Disorders, 1, 269-277.
Friedman, M. J., Schnurr, P. P., & McDonagh-Coyle, A. (1994). Posttraumatic stress disorder
in the military veteran. Psychiatric Clinics of North America, 17, 265-277.
Frueh, B. C., Turner, S. M., & Beidel, D. C. (1995). Exposure therapy for combat-related
PTSD: A critical review. Clinical Psychology Review, 15, 799-817.
Frueh, B. C., Turner, S. M., Beidel, D. C., Mirabella, R. E, & Jones, W. J. (1996). Trauma
management therapy: A preliminary evaluation of a multicomponent behavioral treatment
for chronic combat-related PTSD. Behaviour Research and Therapy, 34, 533-543.
Funari, D. J., Piekarski, A. M., & Sherwood, R. J. (1991). Treatment outcomes of Vietnam
veterans with posttraumatic stress disorder. Psychological Reports, 68, 571-578.
Gerlock, A. A. (1994). Veterans' responses to anger management intervention. Issues in Mental
Health Nursing, 15, 393-408.
Glass, G. V, McGaw, B., & Smith, M. L. (1981). Meta-analysis in social research. Beverly
Hills, CA: Sage.
Hamilton, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery, and
Psychiatry, 23, 169-174.
432 Sherman
*Hammarberg, M., & Silver, S. M. (1994). Outcome of treatment for posttraumatic stress
disorder in a primary care unit serving Vietnam veterans. Journal of Traumatic Stress, 7,
Hedges, L. V, & OIkin, I. (1985). Statistical methods for meta-analysis. Orlando, FL: Academic
Helzer, J. E., Robbins, L. N., & McEvoy, L. (1987). PTSD in the general population: Findings
of the epidemiological catchment area survey. New England Journal of Medicine, 317,
Horowitz, M. (1974). Stress response syndromes, character style, and dynamic psychotherapy.
Archives of General Psychiatry, 31, 768-781.
Horowitz, M. J. (1977). Stress response syndromes and brief psychotherapy. Philadelphia,
Pennsylvania Hospital.
Horowitz, M. J., & Kaltreider, N. B. (1980). Brief psychotherapy of stress response syndromes.
In T. B. Karasu & L. Bellak (Eds.), Specialized techniques in individualized psychotherapy.
New York: Brunner/Mazel.
Horowitz, M. J., Wilner, N., & Alvarez, W. (1979). Impact of event scale: A measure of
subjective stress. Psychosomatic Medicine, 41, 209-218.
Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in
research findings. Newbury Park, CA: Sage.
Hyer, L., Woods, M. G., Bruno, R., & Boudewyns P. A. (1989). Treatment outcomes of
Vietnam veterans with PTSD and the consistency of the MCMI. Journal of Clinical
Psychology, 56, 85-90.
*Jensen, J. A. (1994). An investigation of eye movement desensitization and reprocessing
(EMDR) as a treatment for posttraumatic stress disorder (PTSD) symptoms of Vietnam
veterans. Behavior Therapy, 25, 311-325.
Johnson, B. T. (1989). DSTAT: Software for the meta-analytic review of research literatures. New
Jersey: Lawrence Erlbaum Associates.
*Keane, T. M., Fairbank, J. A., Caddell, J. M., & Zimering, R. T. (1989). Implosive (flooding)
therapy reduces symptoms of PTSD in Vietnam combat veterans. Behavior Therapy, 20,
Keane, T. M., Zimmering, R. X, & Caddell, J. M. (1985). A behavioral formulation of
posttraumatic stress disorder. The Behavior Therapist, 8, 9-12.
Kessler, R. C., Sonnega, A., Bromet, E. Hughes, M., & Nelson, C. B. (1995). Posttraumatic
stress disorder in the national comorbidity survey. Archives of General Psychiatry, 52,
Kilpatrick, D. G., Saunders, B. E., Best, C. L., & Von, J. M. (1987). Criminal victimization:
Lifetime prevalence, reporting to police, and psychological impact. Crime and Delinquency,
33, 479-489.
Lindy, J. D. (1989). Transference and posttraumatic stress disorder. Journal of the American
Academy of Psychoanalysis, 17, 397-413.
Litz, B. T., & Roemer, L. (1996). Posttraumatic stress disorder: An overview. Clinical
Psychology and Psychotherapy, 3, 153-168.
Lyons, J. A. (1987). Posttraumatic stress disorder in children and adolescents: A review of
the literature. Developmental and Behavioral Pediatrics, 8, 349-356.
McConaghy, N.(1990). Can reliance be placed on a single meta-analysis? Australian and New
Zealand Journal of Psychiatry, 24, 405-418.
McFarlane, A. C. (1989). The aetiology of posttraumatic morbidity: Predisposing, precipitating
and perpetuating factors. British Journal of Psychiatry, 154, 221-228.
McFarlane, A. C. (1994). Individual psychotherapy for posttraumatic stress disorder. Psychiatric
Clinics of North America, 17, 393-408.
McNeil, J. W., & Todd, F. J. (1986). The operant treatment of excessive verbal ruminations
and negative emotional arousal in a case of child molestation. Child and Family Behavior
Therapy, 8, 61-69.
Meichenbaum, D. (1985). Stress inoculation training. New York: Plenum Press.
Meta-Analysis of PTSD 433
Miller, T. W., Kamenchenko, P., & Krasniasnski, A, (1992). Assessment of life stress events:
The etiology and measurement of traumatic stress disorder. InternationalJoumal of Social
Psychiatry, 38, 215-227,
Mowrer, A. H. (1960). Learning theory and behavior. New York: Wiley.
Norris, F. H. (1992). Epidemiology of trauma: Frequency and impact of different potentially
traumatic events on different demographic groups. Journal of Consulting and Clinical
Psychology, 60, 409-418.
*Peniston, E. O. (1986). EMG biofeedback assisted desensitization treatment for Vietnam
combat veterans posttraumatic stress disorder. Clinical Biofeedback and Health, 9, 35-41.
Pitman, R. K., Altman, B., Greenwald, E., & Longpre, R. E. (1991). Psychiatric complications
during flooding therapy for posttraumatic stress disorder. Journal of Clinical Psychiatry,
52, 17-20.
Pitman, R. K., Orr, S. P., Altman, B., Longpre, R. E., Poire, R. E., & Macklin, M. L. (1996).
Emotional processing during eye movement desensitization and reprocessing therapy of
Vietnam veterans with chronic posttraumatic stress disorder. Comprehensive Psychiatry,
37, 419-429.
Putnam, J. J, (1881). Recent investigations into patients of so-called concussion of the spine.
Boston Medical and Surgical Journal, 109, 217.
*Ragsdale, K. G., Cox, R. D., Finn, P., & Eisler, R. M. (1996). Effectiveness of short-term
specialized inpatient treatment for war-related posttraumatic stress disorder: A role for
adventure-based counseling and psychodrama. Journal of Traumatic Stress, 9, 269-283.
Raphael, B., Meldrum, L., & McFarlane, A. C. (1996). Does debriefing after psychological
trauma work? Time for randomized controlled trials. Accident and Emergency Nursing, 4,
*Resick, P. A., & Schnicke, M. K. (1992). Cognitive processing therapy for sexual assault
victims. Journal of Consulting and Clinical Psychology, 60, 748-756.
Resnick, H. S., Kilpatrick, D. G., Dansky, B. S., Saunders, B. E., & Best, C. L. (1993).
Prevalence of civilian trauma and posttraumatic stress disorder in a representative
national sample of women. Journal of Consulting and Clinical Psychology, 61, 984-991.
Richards, D. A., Lovell, K., & Marks, I. M. (1994). Posttraumatic stress disorder: Evaluation
of a behavioral treatment program. Journal of Traumatic Stress, 7, 669-680.
Rosenthal, R. (1979). The "file drawer problem" and tolerance for null results. Psychological
Bulletin, 86, 638-641.
Rosenthal, R. (1984). Meta-analytic procedures for social research, Beverly Hills, CA: Sage.
Rosenthal, R. (1995). Writing meta analytic reviews. Psychological Bulletin, 118, 183-192.
Saigh, P. A. (1986). In vitro flooding in the treatment of a 6-year old boy's posttraumatic
stress disorder. Behaviour Research and Therapy, 24, 685-688.
Saigh, P. A. (1887a). In vitro flooding of childhood posttraumatic stress disorders: A systematic
replication. Professional School Psychology, 2, 135-146.
Saigh, P. A. (1987b). In vitro flooding of an adolescent's posttraumatic stress disorder. Journal
of Clinical Child Psychology, 16, 147-150.
Salmon, T. W. (1919). The war neuroses and their lessons. New York State Medical Journal,
59, 993-994.
Scurfield, R. M., Wong, L. E., & Zeerocah, E. B. (1992). An evaluation of the impact of
"helicopter ride therapy" for in-patient Vietnam veterans with war-related PTSD, Military
Medicine, 157, 67-73.
Shalev, A. Y., Bonne, O., and Eth, S. (1996). Treatment of posttraumatic stress disorder: A
review. Psychosomatic Medicine, 58, 165-182.
Shapiro, F. (1989). Efficacy of the eye movement desensitization procedure in the treatment
of traumatic memories. Journal of Traumatic Stress, 2, 199-223.
Shapiro, F. (1991). Eye movement desensitization and reprocessing procedure: From EMD to
EMDR—A new model for anxiety and trauma. The Behavior Therapist, 14, 133-145.
Silver, S. M., Brooks, A., & Obenchain, J. (1995), Treatment of Vietnam war veterans with
PTSD: A comparison of eye movement desensitization and reprocessing, biofeedback,
and relaxation training. Journal of Traumatic Stress, 8, 337-342.
434 Sherman
Smith, M. L., & Glass, O. V. (1977). Meta-analysis of psychotherapy outcome studies. American
Psychologist, 32, 752-760.
Solomon, S. D., Certify, E. T, & Muff, A. M. (1992). Efficacy of treatments for posttraumatic
stress disorder: An empirical review. Journal of the American Medical Association, 268,
*Solomon, Z., Shalev, A., Spiro, S. E., Dolev, A., Bleich, A., Waysman, M., & Cooper, S.
(1992). Negative psychometric outcomes: Self-report measures and a follow-up telephone
survey. Journal of Traumatic Stress, 2, 225-246.
Spielberger, C. D., Oorsuch, R. L., & Lushene, R. E. (1970). Manual for the State-Trait
Anxiety Inventory (self-evaluation questionnaire). Palo Alto, CA: Consulting Psychologists
Strube, M. J., & Hartmann, D. P. (1983). Meta-analysis: Techniques, applications and functions.
Journal of Consulting and Clinical Psychology, 51, 14-27.
*Vaughan, K., Armstrong, M. S., Gold, R., O'Connor, N., Jenneke, W., & Tarrier, N. (1994).
A trial of eye movement desensitization compared to image habituation training and
applied muscle relaxation in posttraumatic stress disorder. Journal of Behavior Therapy
and Experimental Psychiatry, 25, 283-291.
Veronen, L. J., & Kilpatrick, D. G. (1983). Stress management for rape victims. In D.
Meichenbaum & M. E. Jaremko (Eds.), Stress Reduction and Prevention. New York:
Plenum Press.
Wilson, S. A., Becker, L. A., & Tinker, R. H. (1995). Eye movement desensitization and
reprocessing (EMDR) treatment for psychologically traumatized individuals. Journal of
Consulting and Clinical Psychology, 63, 928-937.
Wolfe, E M., & Cornell, R. G. (1986). Interpreting behavioral, biomedical, and psychological
relationships in chronic disease from 2x2 tables using correlation. Journal of Chronic
Disease, 39, 605-608.
Wolraich, M. L., Wilson, D. B., & White, W. (1995). The effect of sugar on behavior or
cognition in children. Journal of the American Medical Association, 274, 1617-1621.
Meta-Analysis of PTSD 435

NASPs Effective Interventions for PTSD

Coping After a Crisis

In the wake of a crisis almost everyone is
hurt and confused, but some are at risk of
developing long-term consequences. Quick,
effective intervention can help alleviate
these people’s pain.


JANUARY 2 0 0 4 9
Until recently severe trauma
reactions were most commonly
associated with veterans
or survivors of major catastrophes,
not the young people filling
our nation’s schools. Generally,
when a student or a staff member
coped with the psychological aftermath
of a tragedy, they did so without
the involvement—or responsibility—
of school personnel. But
educators have come to recognize
that schools play a critical role in
any crisis response and care system
serving children and youth. This is
true whether dealing with an act of
violence on campus, an accident
involving a serious injury or loss of
life, a natural disaster, or terrorism.
For the most part, schools have
done an admirable job of improving
crisis support in terms of planning,
prevention, and ensuring physical
safety during and immediately after
an event. An essential part of this
effort, however, is the ability to recognize
and respond to the mental
health effects of a crisis—including
such serious disorders as posttraumatic
stress disorder (PTSD).
Although principals are primarily
concerned with understanding and
meeting the needs of students, they
also must pay attention to the
potential effect of a crisis on staff
members, particularly those who are
serving as crisis caregivers for their
Range of Reactions
Anyone, regardless of psychological
strength, can be initially affected by
exposure to an event that causes
fear, helplessness, or horror.
Symptoms can be emotional, physical,
cognitive, and social and interpersonal
and vary significantly in
severity and duration. Strong reactions,
such as anxiety or anger, can
be normal and usually are not debilitating.
With time, most people are
able to recover from the psychological
effects of a traumatic experience.
According to the National Institutes
of Mental Health (2002), a “sensible
working principle” in the immediate
aftermath of a traumatic event “is to
expect normal recovery.”
A minority of children and
adults, however, will develop longterm
difficulties that can significantly
impair their daily functioning. Some
will be at risk of developing PTSD,
the most common and devastating
of the clinical anxiety disorders
that are triggered by psychological
trauma. The exact size of this
minority will depend on the type
and severity of the event because
some events—especially acts of
human violence—are more traumatic
than others. Although only a
trained mental health professional
can diagnose and treat PTSD, educators
can help minimize potentially
serious consequences of the
disorder by recognizing warning
signs, identifying high-risk students,
and providing appropriate
interventions and referrals.
What Is PTSD?
PTSD is a characteristic set of
symptoms resulting from “exposure”
to a “traumatic stressor” (American
Psychological Association, 2000).
The kinds of stressors most likely to
result in PTSD include death, serious
injury/harm, and other threats
to physical integrity. Exposure is
defined as directly experiencing or
witnessing a traumatic event or
learning about an event being
experienced by a family member,
close friend, or another loved one.
Coping After a Crisis
In the wake of a crisis almost everyone is
hurt and confused, but some are at risk of
developing long-term consequences. Quick,
effective intervention can help alleviate
these people’s pain.
Stephen E. Brock is a nationally certified school psychologist and school psychology
trainer at California State University–Sacramento. Kathy Cowan is director of
communications for the National Association of School Psychologists (NASP). This
article was written in cooperation with NASP.
C O U N S E L I N G 1 0 1
This explains why a number of
people developed PTSD after
September 11 although they were
not directly affected by the attacks:
They knew someone close who had
died or they “experienced” the catastrophe
on television.
The general symptom categories
of PTSD are persistent reexperiencing
of the traumatic stressor, persistent
avoidance of reminders of the
traumatic event, emotional numbing,
and persistent symptoms of
increased arousal. PTSD is different
from a more “typical” traumatic
event response (i.e., a normal
response to abnormal circumstances)
by the duration and intensity of the
symptoms. They must last for at
least one month and cause significant
disruption to the individual’s
daily life. For example, following a
school-related traumatic event a student
or a staff member may be
unable to return to the building.
Different Ages, Different
PTSD symptoms of adolescents
closely resemble those of adults.
Teenagers with PTSD may become
more aggressive, start fights, or
behave irrationally. They may have
trouble sleeping, participating in
class, and completing assignments
or lose interest in cocurricular activities
and friends. They are also at
increased risk of substance and
alcohol abuse, reckless behavior,
and suicide.
Younger middle school students
and children may display a different
pattern of symptoms. Specifically,
their anxieties can be more generalized
than those of older individuals
whose symptoms are much more
clearly linked to the traumatic event.
For example, they may display generalized
fear of strangers; separation
anxiety; and sleeping difficulties,
including frightening dreams that
do not necessarily reflect the stressor.
They may engage in repetitive
play or representations in art or
writing that reflect the trauma. They
may also be more disruptive in class,
express the belief that they will not
live to adulthood, believe that certain
omens foretell traumatic events,
and exhibit such physical symptoms
as headaches and stomachaches.
How Common Is PTSD?
The lifetime prevalence of PTSD
among adults is slightly less than
8% and among children and adolescents
slightly more than 10%.
Females are two times more likely
than males to develop the disorder.
The rate of PTSD after any specific
event varies greatly depending on its
type and intensity. Sudden, humancaused
disasters involving assault,
injury, physical threats, or fatalities
can be particularly traumatic, as are
unusually intense and long-lasting
experiences. In addition, PTSD
seems more likely to develop in
young people if the perpetrator of a
violent act is a trusted adult. It is
important to keep in mind, however,
that virtually any traumatic experience
has the potential to cause
PTSD among some victims, given
the right set of circumstances and
the particular vulnerabilities of the
individuals involved.
Risk Factors
Usually exposure to a traumatic
event is not sufficient in itself to
generate PTSD in most people.
Certain factors put specific individuals
at greater risk for developing
the disorder. A fatal school bus accident
may not cause PTSD in most
of the students involved but might
trigger it in the student whose father
recently died in a car crash. Being
aware of the more powerful predictors
of PTSD can help school personnel
and parents identify students
who are potentially more vulnerable.
Physical and emotional proximity
to the traumatic event. The closer
people are to a traumatic event,
the greater the likelihood that they
will develop PTSD. Proximity
includes both direct personal exposure
to the event (physical proximity)
and relationships with crisis victims
(emotional proximity), particularly
when the victim is killed. Not
surprisingly, crisis victims, especially
those who were physically injured,
are at greatest risk.
Subjective perceptions of the
traumatic event. The individual’s
subjective impression of the traumatic
event can be more important than
the event itself. Simply put, those
who develop PTSD perceive traumatic
events as extremely threatening.
Those who do not view an
event as threatening (no matter how
horrific others may judge it to be)
are unlikely to develop PTSD. This
is why the demeanor of teachers and
other staff members is important
during and after a crisis. The reactions
of trusted adults can help shape
students’ subjective perceptions of a
traumatic experience, particularly for
middle level students.
Family and social factors.
Students who no longer live with a
nuclear family member, are exposed
to family violence, have a family his-
School counselors
are staunch
student advocates
who work closely
with teachers and
administrators to
ensure students’
welfare and
protect their rights.
JANUARY 2 0 0 4 11
tory of mental illness, or live with
caregivers who have PTSD are more
likely to develop PTSD themselves.
In addition, the absence of close
peer friendships, access to positive
adult models outside of the family,
or connections to prosocial organizations
or institutions (such as
schools) increases vulnerability.
Mental health and trauma
history. A preexisting mental illness
influences the development of
PTSD. Students who had mental
health problems (such as depression
or bipolar disorders) before experiencing
the trauma are more likely to
develop PTSD than those with good
baseline mental health. Individuals
who previously experienced psychological
trauma also are more likely
to develop (or reexperience) PTSD.
What Are the Warning Signs of
Immediately after a traumatic event,
it may be difficult to identify those
who will develop PTSD because
symptoms can mirror many normal
reactions (e.g., anger, difficulty concentrating,
and nightmares) that will
dissipate on their own. Distinguishing
the difference requires training; any
concern about a student should be
referred to a mental health professional.
However, initial reactions to
the event are important indicators of
whether someone may develop
The initial response of individuals
who develop PTSD usually involves
“intense fear,” “helplessness,” or
“horror.” Not only does a severe
immediate response (e.g., panic) act
as a powerful warning sign, but it
also influences the individual’s ability
to cope with the experience independently
and adaptively. Educators
should note these immediate
responses in students and consider
them when determining who
requires crisis intervention assistance
and support. The presence of any
persistent PTSD symptom in the
Assessing the Risk for PTSD
A local gang, in response to the beating of a fellow gang member by a student at
your high school, has come on campus. A fight breaks out in the student parking lot
between the gang and the student’s friends. A 15-year-old gang member is hospitalized
with a stab wound, and one of your students is killed by a gunshot wound to
the head. A teacher was in the immediate area and tried to intervene; she was
hospitalized with a serious stab wounds but is expected to live.
How Traumatic Is the Crisis Event?
The situation described in this case study is obviously extreme and very rare, but it
illustrates many of the principles of assessing the risk for PTSD. This event involves
violence that has resulted in death and injury and likely generated feelings of intense
fear, helplessness, and horror within the school community. This is the type of event
that is likely to be highly traumatic and generate symptoms of PTSD among survivors,
witnesses, and friends of the victims. This situation would probably require
comprehensive crisis intervention from the school, which may include crisis intervention
team members who are not typically a part of the school staff. The fact that a
teacher was injured may significantly affect school staff members and increase the
need for outside assistance.
Who Is At Risk?
Students who were involved in the fight (especially those who were injured or who
felt that they were threatened), those who were close friends of the student who was
killed, and other staff members and students who were especially close to the
injured teacher are at the greatest risk for developing PTSD symptoms. These individuals
were physically or emotionally proximal to the traumatic event. In particular,
any student or staff member who displays significant crisis reactions (e.g., who panicked
during or immediately after the fight) or are socially isolated (e.g., who viewed
the deceased student as their one and only friend or who do not have a supportive
family) are at particular risk.
To a lesser degree, the students and staff members who knew but were not
especially close to the fight victims are also at risk. Attention should also be directed
to students and staff members who have personal histories that include witnessing
or experiencing assaultive violence (especially those with a prior history of
PTSD) or experiencing any other mental disorder.
Students and staff members who are at low risk for developing PTSD would
include those who did not see the fight or its aftermath, did not have any relationship
with victims, did not ever feel that they were in danger, and do not display any
crisis reactions.
How Should the School Respond?
The school needs to classify students and staff members according to risk factors
and levels and carefully monitor their status. Psychological first aid should be
offered to anyone who requests it, but special attention should be directed to those
who were physically and emotionally close to the crisis. The school needs to be prepared
to refer any student or staff member whose symptoms do not remit within a
few weeks to a mental health professional who has expertise in dealing with traumatic
stress. In addition, the school should be prepared to make such a referral of
anyone who reports that they subjectively feel that they are having difficulty coping
with the traumatic event.
Although it is important for the school to identify those who are at high risk for a
traumatic stress reaction to provide them with immediate assistance, it is also important
to identify those who are at low risk and to give them the opportunity to cope
with the crisis. Doing so may help to generate feelings of empowerment and the
belief that they are able to manage stressful events. Providing crisis intervention
assistance to those who truly don’t need crisis team support may unintentionally
send the message that the event was more threatening then it actually was or that
the student is not a capable problem solver.
aftermath of a traumatic event is
also reason for concern and should
trigger careful monitoring of affected
students. Among those symptoms
that appear to be most worrisome
are those that reflect an
unusually high level of alertness
(e.g., being easily startled).
How Can Schools Help?
PTSD is a very serious mental
disturbance that cannot be taken
lightly. Effective treatment typically
requires the assistance of a clinical
psychologist or psychiatrist who has
specific training dealing with trauma
victims. The most important role
that principals can play is to ensure
that their staff members are able to
recognize risk factors and warning
signs of PTSD and know how to
make appropriate referrals. Although
usually not trained to treat PTSD,
most school-based mental health
professionals (e.g., school counselors
and school psychologists) can provide
the staff development training
needed to ensure that school personnel
can recognize PTSD symptoms
and provide appropriate screening of
students who might need referral to
expert providers in the community.
Other recommendations include the
Address the needs of any student
who exhibits signs of distress.
Being part of a caring support network
is one of the most important
ways to protect people against serious
trauma reactions. Educators
should respond to students who
appear in distress, even if they do
not have risk factors for PTSD.
Reaching out reinforces that adults
care and enables staff members to
build students’ natural resiliency as
well as identify those students who
need further monitoring or referral
to a mental health professional.
Establish a crisis intervention
team. These teams are designed to
help students and personnel cope
with psychologically traumatic
events and to identify those who
need professional mental health
assistance (e.g., those with PTSD).
Team members, including some
mental health personnel, may need
additional training in crisis response.
It also may be necessary to identify
trained personnel in the district or
coordinate training and staff
resources with other schools or
neighboring districts.
Develop a protocol for the school
crisis intervention response. Such
School personnel who provide crisis care to students may be at increased risk of developing PTSD
if they also are affected by the event or have personal risk factors. Offering the following advice
can help principals ensure that staff members receive adequate support:
• Don’t underestimate the effect of providing crisis care.
• Understand the nature of the crisis and the students who need care.
• Know your limitations and what you can feel reasonably comfortable handling in a crisis.
• Be aware of your history of personal loss or trauma.
• Be willing to decline an assignment or seek help from someone more experienced.
• Ask family members and friends to help with regular chores if your crisis responsibilities demand extra
• Maintain a healthy diet and water intake.
• Get plenty of sleep, preferably without the use of sleep aids or alcohol.
• Take periodic breaks while in the midst of a crisis response; go for a walk or call a friend.
• Connect with trusted friends or family members who can help counter negative feelings.
• Take time to process daily events with team members or colleagues.
• Find an acceptable outlet—for example, exercise, a favorite hobby, sports, music, art, or movies.
• Avoid excessive news coverage.
C O U N S E L I N G 1 0 1
JANUARY 2 0 0 4 13
a protocol should identify specific
individuals to fill specific crisisintervention
roles. Among these
roles is a mental health officer, who
is responsible for establishing referral
mechanisms and monitoring crisis
reactions among students and staff
members. The protocol should
include a school policy regarding
access to onsite psychological firstaid
assistance either by community
mental health professionals or
trained school personnel.
Learn which students may be
at increased risk for developing
PTSD. Provide parents with information
about the risk factors and
symptoms of PTSD. Encourage
them to tell their child’s teacher,
school psychologist, or counselor if
their child has experienced a previous
traumatic event or personal loss,
has a mental health problem, or is
exhibiting warning signs. Ideally,
parents should know to share this
information under any circumstances,
but they should be reminded
immediately following a crisis
event involving members of the
school community.
Encourage students to seek
help. Although individuals cannot
self-diagnose PTSD, students can
recognize when they or a friend
are experiencing problematic
symptoms. Schools need to reinforce
that students should tell a
trusted adult any time their feelings
or thoughts interfere with their normal
routine. Student self-reporting
can augment, but not replace, adult
observation of students affected by a
traumatic event.
Pay attention to the needs of
staff members. Depending on the
event, staff members may be at risk
of PTSD. They may be affected personally
or have individual risk factors.
They may be at additional risk
if they are acting as crisis caregivers
to students. Supporting the emotional
needs of students over
an extended period of time is draining,
particularly for teachers and
support personnel who must remain
“in control” day in and day out.
Principals should allow staff members
time to take care of their own
needs. The school psychologist can
help support individual staff members
and facilitate group discussions.
Ensure that a range of schooland
community-based interventions
are available for students
and staff members. For the minority
of individuals who need professional
mental health assistance, it is
important that the school identifies
who in the local mental health community
has expertise in working
with trauma victims. In particular, it
will be important to know who has
training in a form of psychotherapy
known as cognitive-behavioral treatment.
This form of therapy has documented
effectiveness in helping
individuals to recover from PTSD.
The school psychologist or social
worker is generally an ideal resource
for appropriate community referrals.
❏ American Psychiatric Association.
(2000). Diagnostic and statistical manual
of mental disorders (4th ed.). Washington,
DC: Author.
❏ Brock, S. E., Lazarus, P. J., &
Jimerson, S. R. (2002). Best practices in
school crisis prevention and intervention.
Bethesda, MD: National Association of
School Psychologists. Related document
can be found at
/NEAT /trauma.html
❏ National Institute of Mental Health.
(2002). Mental health and mass violence:
Evidence-based early psychological intervention
for victims/survivors of mass violence.
A workshop to reach consensus on
best practice [NIH Pub. No. 02-5138].
Washington, DC: U.S. Government
Printing Office. Retrieved July 25, 2003
Persistent reexperiencing of the traumatic stressor: Reoccurring intrusive and distressing thoughts,
images, or feelings associated with the event; reoccurring and upsetting dreams about the trauma.
Persistent avoidance of reminders of the event: Deliberate efforts to avoid thoughts, feelings,
discussions, activities, places, or people that are associated with the traumatic event; inability to
remember elements of the event.
Emotional numbing: Reduced interest in important and previously enjoyed activities; feeling all alone or
detached from others and unable to react emotionally; feeling as if there is no future.
Persistent symptoms of increased arousal: Difficulty falling or staying asleep; unusually alert and
easily startled; difficulty concentrating; increased irritability and anger.

Communique Article for School Psychologists

Crisis Management
in the Schools
A Cognitive Model of PTSD

Summarized by Karen Lake, Predoctoral
Intern, Psychological & Social
Services, APPIC Internship Site, Dallas
Independent School District, TX
In an invited essay for Behaviour
Research and Therapy, Anke Ehlers and
David Clark (2000) discuss posttraumatic
stress disorder as described in a cognitive
model. Their article, “A Cognitive
Model of Posttraumatic Stress Disorder,”
explores theoretical reasons that some
individuals overcome traumatic events
within months while others’ symptomology
persists for years.
This theoretical paper proposes that
persistent PTSD occurs when individuals
process traumatic events in a way that
reproduces a sense of threat in current situations.
The authors address reasons for
persistent symptomology including an
inability to see the trauma event as timelimited,
overgeneralization of the danger
from the event to a normal range of activities,
interpretation of one’s own reaction
as well as others’ reactions in the aftermath
of the event as negative, and maladaptive
behavioral strategies and cognitive
processing styles.
Ehlers and Clark also provide treatment
implications based on the proposed cognitive
model. Specifically, treatment should
address the trauma memory by elaborating
and integrating the experience in order to
reduce intrusive reexperiencing, and problematic
appraisals that maintain the current
threat should be explored and modified.
Finally, dysfunctional behavioral and
cognitive strategies preventing memory
elaboration, exacerbating symptoms,
and/or hindering reassessment of problematic
appraisals should be abandoned.
Although the model in its entirety has
not been evaluated empirically, several of
its central features have been tested with
promising results. Specifically, Ehlers and
Clark cite multiple supportive sources for
negative appraisals of the trauma, negative
interpretations of initial PTSD symptoms,
negative appraisals of others’ posttrauma
responses, and improvement of
symptoms with the development of a cohesive
and organized narrative of the event.
Many of the supportive studies, however,
were conducted by the authors themselves.
The cognitive model proposed by Ehlers
and Clark may have important implications
for school psychologists when faced with
crises in the schools. First, this model confirms
traditional knowledge of PTSD as
Editor’s Note: In this edition of the Research Reviews column, Crisis Management in the Schools
Interest Group members summarize recent crisis management publications. In this issue we will be
focusing on research related to posttraumatic stress disorder (PTSD) and below we bring to your
attention to three articles we believe have relevance to school psychologists. The first article summarized
provides a cognitive model of PTSD that may help to inform the school crisis intervention
response (especially as it relates to the identification of more serious psychological trauma victims).
The second paper reviews the literature regarding the treatment of PTSD among children and provides
guidance regarding best therapeutic practices. The final article summarized offers a metaanalytic
review of the predictors of PTSD.
developing due to a real or perceived serious
threat to one’s well-being. The authors
sought to explore preexisting characteristics
that perhaps make individuals more
susceptible to the development of PTSD.
The implications may be useful for crisis
response in the schools as a means of identifying
those individuals who may need secondary
and tertiary levels of treatment. The
authors also provided research-based treatment
protocols that may be used with
school personnel and students experiencing
distress following a school or community
crisis event. Furthermore, the central
features of PTSD presented in this model
may be modified through psychoeducational
groups in the schools. Specifically, school
personnel may be able to normalize the
emotional and psychological turmoil that
may result from school-based trauma.
Ehlers and Clark’s cognitive model of
posttraumatic stress disorder provides
theoretical foundations for the symptoms
of persistent PTSD as well as potentially
helpful treatment implications. This model
may prove useful for school personnel in
the identification of secondary and tertiary
treatment groups and in the treatment
planning for these groups. The model has
promising implications and warrants further
Ehlers, A., & Clark, D. (2000). A cognitive model of
posttraumatic stress disorder. Behaviour Research
& Therapy, 38, 319–345.
Posttraumatic Stress
Disorder in Youth
Summarized by Elizabeth J. Zhe,
PsyD, Culver Academies, Culver, IN
Professional Psychology: Research and
Practice published a review article
by Feeny, Foa, Treadwell, and March
(2004) critiquing research on PTSD treatments
for youth. Despite the substantial
amount of research on the treatment of
PTSD in adult populations, PTSD treatment
outcome studies in youth populations
are sparse. Feeny et al.’s review
strives to bring light to the clinical implications
of existing research and to detect
gaps where future research is needed.
The authors review the prevalence of
trauma exposure in school-age populations
and youth with emotional disturbances, as
well as common psychological reactions to
exposure. PTSD is pinpointed as the most
common type of psychological reaction to
trauma exposure, and yet few youth with
PTSD receive empirically supported treatments
for their diagnosis. Common selfreport
measures used to evaluate traumatized
youth are also briefly reviewed. Feeny
et al. identify numerous self-reports as adequate
initial screening instruments; however,
none provides a Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV;
American Psychiatric Association, 1994)
diagnosis of PTSD. Foa, Johnson, Feeny, &
Treadwell (2001) describe the Child PTSD
Symptom Scale as a solid measure that evaluates
both DSM-IV symptoms and functional
An empirical review of cognitive and
behavioral treatments (CBT) for youth
with PTSD is then provided. Due to the
lack of research on other types of treatments
for traumatized youth (i.e., one published
randomized controlled study of a
non-CBT psychosocial intervention and
no published pharmacological interventions
with controlled trials), only CBTs
are reviewed. Reviewed treatments were
individually delivered exposure treatment,
eye-movement desensitization and reprocessing
(EMDR), anxiety-management
training (AMT), group-delivered treatments,
and child–parent treatments for
sexual abuse-related trauma.
On the whole, the authors assert that
there is mounting empirical support for the
use of numerous CBT programs in the treatment
of youth with PTSD. Key findings are
(a) there is preliminary evidence found in
numerous single-case designs for the effectiveness
of exposure, group CBT with exposure,
and anxiety-management training; (b)
evidence is found in well-designed research
studies with randomized trials that brief
treatments for disaster- and violence-related
PTSD are efficacious; (c) sequential studies
of child and parent CBT programs
(including exposure and parent-training
for PTSD and symptoms of sexual abuse)
are suggestive of short- and long-term efficacy
when provided in either individual,
group, or parallel child and parent treatments;
and (d) effective CBT treatments
for sexually abused youth, with or without
a focus on treating PTSD, entail parent or
family involvement. While the review suggests
CBT is the most empirically supported
treatment of PTSD for youth, psychologists
should base their choice of CBT
methodology on consideration of whether
sexual trauma is involved, if there are cooccurring
mental health diagnoses, and the
youth’s developmental level.
Feeny et al. (2004) conclude with a discussion
of areas for future research and
the necessity for psychologists to obtain
additional supervision or education in CBT
as it relates to PTSD. By and large, additional
research is needed to conduct largescale
randomized control trials comparing
CBT and other common PTSD
Stephen Brock, PhD, NCSP, is on the school psychology faculty of California State University, Sacramento.
He is a member of NASP’s National Emergency Assistance Team (NEAT), a past-coordinator of
the Crisis Management Interest Group, and author of the recently developed NASP school crisis intervention
training curriculum (Crisis Intervention & Recovery: The Roles of School-Based Mental Health Professionals).
This column will appear several times during the year. If you would like to write a research summary
and/or know of a study that should be summarized please contact Dr. Brock at
treatments (e.g., pharmacological, nondirective
play therapy, psychodynamic therapies),
as well as the effectiveness of medication
alone versus medication plus
psychosocial treatment interventions. The
authors also cite the need for treatment
outcome research studies to include measures
beyond those of symptom reduction,
such as academic and family functioning.
There is also a need for large-scale dissemination
studies in natural settings and
investigation of how treatment outcome
is impacted by comorbidity, family functioning,
and early termination.
American Psychiatric Association. (1994). Diagnostic
and statistical manual of mental disorders (4th
ed.). Washington, DC: Author.
Feeny, N. C., Foa, E. B., Treadwell, K. R. H., & March,
J. (2004). Posttraumatic stress disorder in youth:
A critical review of the cognitive and behavioral
treatment outcome literature. Professional Psychology:
Research and Practice, 35, 466–476.
Foa, E. B., Johnson, K., Feeny, N. C., & Treadwell, K.
R. T. (2001). The Child PTSD Symptom Scale
(CPSS): Preliminary psychometrics of a measure
for children with PTSD. Journal of Clinical Child
Psychology, 30, 376–384.
Predictors of PTSD
Among Adults
Summarized by Calissia Thomas
Tasby, Predoctoral Intern, Psychological
& Social Services, APPIC Internship
Site, Dallas Independent School
District, Dallas, TX
In the past decade, studies have been
conducted to specifically examine the
role certain variables play in contributing
to PTSD in adults. Since more than half
of the United States population is exposed
to a traumatic event, but only approximately
7% actually develop PTSD, the importance
placed on determining why the
remaining 43% do not develop PTSD is warranted
(Kessler, Sonnega, Bromet, Hughes,
& Nelson, 1995). In examining these phenomena,
most reviews, with the exception
of Brewin, Andrews, and Valentine (2000),
failed to conduct a comprehensive review
of the PTSD literature or use quantitative
measures (effect size [ES]; Emery, Emery,
Shama, Quiana, & Jassani, 1991; Fontana &
Rosenheck, 1994; Green, 1994). This lack of
comprehensive and quantitative research
prompted the study by Ozer, Best, Lipsey,
and Weiss (2003) summarized below.
In a meta-analysis conducted through
a literature review examining symptoms
predictive of PTSD, Ozer et al. (2003) identified
2,647 articles that related to PTSD,
of which 476 met the inclusion criteria.
Studies that did not assess all of the clusters
of PTSD symptoms as defined in the
4th edition of the Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV;
American Psychiatric Association, 1994)
© 2008, National Association of School Psychologists September 2008, Volume 37, Number 1 | Communiqué | 35
were eliminated leaving 68 studies to be
included in the present study. Of the 68
studies, all met criteria under seven predictors
of PTSD in adults (ages 18 and
over) that included:
(a) a history of at least one other
trauma prior to the index traumatic
event, (b) psychological adjustment
prior to the traumatic event, (c)
family history of psychopathology,
(d) perceived life threat during the
traumatic event, (e) perceived
social support following the traumatic
event, (f) peritraumatic emotionality—
high levels of emotion
during or in the immediate aftermath
of the traumatic event, and (g)
peritraumatic dissociation—dissociative
experiences during or in the
immediate aftermath of the traumatic
event. (Ozer et al., 2003, p. 55)
Results of the analysis using the seven
criteria above yielded significant effect
sizes. Family history, prior trauma, and
prior adjustment produced the smallest
effect size (weighted r = .17). The largest
effect size was produced by peritraumatic
dissociation (weighted r = .35). From the
results, prior characteristics do not appear
to play a significant role in the development
of PTSD. Instead the psychological
response (or peritraumatic psychological
processes) following exposure to a traumatic
event is the best predictor of PTSD.
These findings are important for school
psychologists to understand when working
with children and their families, especially
during crisis. Understanding the factors
that contribute to PTSD in adults may
also help in providing psychological services
that counteract the development of
PTSD in children. Specifically, these results
suggest that it is important to mitigate the
psychological arousal level after a crisis.
The present study was not designed to
account for all mechanisms that contribute
to PTSD. For example, future investigators
may want to examine the statistical role
that distal predictors such as family history
of psychopathology and prior trauma
play in the development of PTSD. ■
American Psychiatric Association. (1994).
Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: Author.
Brewin, C. R., Andrews, B., & Valentine, J. D.
(2000). Meta-analysis of risk factors for posttraumatic
stress disorder in trauma-exposed
adults. Journal of Consulting and Clinical
Psychology, 68, 748–766.
Emery, V. O., Emery, P. E., Shama, D. K., Quiana,
N. A., & Jassani, A. K. (1991). Predisposing
variables in PTSD patients. Journal of Traumatic
Stress, 4, 325–343.
Fontana, A., & Rosenheck, R. (1994). Posttraumatic
stress disorder among Vietnam theater
veterans: A causal model of etiology in a community
sample. Journal of Nervous and Mental
Disease, 182, 677–684.
Green, B. L. (1994). Psychosocial research in
traumatic stress: An update. Journal of Traumatic
Stress, 7, 341–362.
Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M.,
& Nelson, C. B. (1995). Posttraumatic stress
disorder in the National Comorbidity Survey.
Archives of General Psychiatry, 52, 1048–1060.
Ozer, E. J., Best, S. R., Lipsey, T. L., & Weiss, D. S.
(2003). Predictors of posttraumatic stress
disorder and symptoms in adults: A metaanalysis.
Psychological Bulletin, 129, 52–73.
Culturally Responsive Practice
Resources and Initiatives
NASP is pleased to announce the release of the latest
career brochure, School Psychology: The Role of a
Native American School Psychologist. The brochure
outlines the important contributions indigenous
school psychologists make toward strengthening
Native communities and helping children succeed
in school and life. Additionally, information about
what services school psychologists provide and
how NASP supports culturally responsive practice
are noted. The brochure was developed by
the Native American Group of the Multicultural
Affairs Committee.
■ Read more about the recruitment initiatives
of the Multicultural Affairs Committee
on the Culturally Competent
Practice webpage at www.nasponline
■ Download the Native American
career brochure or order copies
from the NASP website
Directory of Bilingual School Psychologists. NASP is building an online directory
of bilingual schools psychologists. NASP members who are fluent in languages other
Mary Beth Klotz, NCSP, is NASP Director, IDEA Projects and Technical Assistance. Some of the