Content uploaded by Dawne Vogt
Author content
All content in this area was uploaded by Dawne Vogt on Nov 20, 2014
Content may be subject to copyright.
The Impact of Killing on Mental Health Symptoms in Gulf War Veterans
Shira Maguen
San Francisco VA Medical Center and University of
California, San Francisco
Dawne S. Vogt, Lynda A. King, Daniel W. King,
and Brett T. Litz
National Center for PTSD, Department of Veterans Affairs
Boston Healthcare System and Massachusetts Veterans
Epidemiological Research and Information Center, and Boston
University School of Medicine
Sara J. Knight and Charles R. Marmar
San Francisco VA Medical Center and University of California, San Francisco
This study examined the impact of killing on posttraumatic stress symptomatology (PTSS), depression,
and alcohol use among 317 U.S. Gulf War veterans. Participants were obtained via a national registry of
Gulf War veterans and were mailed a survey assessing deployment experiences and postdeployment
mental health. Overall, 11% of veterans reported killing during their deployment. Those who reported
killing were more likely to be younger and male than those who did not kill. After controlling for
perceived danger, exposure to death and dying, and witnessing killing of fellow soldiers, killing was a
significant predictor of PTSS, frequency and quantity of alcohol use, and problem alcohol use. Military
personnel returning from modern deployments are at risk of adverse mental health symptoms related to
killing in war. Postdeployment mental health assessment and treatment should address reactions to killing
in order to optimize readjustment.
Keywords: killing, veterans, gulf war, PTSD, alcohol
Military personnel involved in war kill as part of their mission,
although few studies have examined the rates of those involved in
killing or the mental health impact of taking another life. In a
population-based epidemiological study of Gulf War veterans,
14% of male soldiers reported killing another person (Carney et
al., 2003), and these numbers are even higher when surveying
combat veterans. Hoge and colleagues (2004) found that 77% to
87% of soldiers in infantry units returning from Operation Iraqi
Freedom (OIF) reported shooting or directing fire at the enemy,
48% to 65% reported being responsible for the death of an enemy
combatant, and 14% to 28% reported being responsible for the
death of a noncombatant.
The few studies that have examined killing in the war zone as a
risk factor for combat-related posttraumatic stress disorder (PTSD)
have found that taking another life is an important and robust
predictor (Fontana & Rosenheck, 1999; MacNair, 2002). Each of
these studies controlled for exposure to other combat variables to
ensure that the results were not attributable to combat exposure
alone, but due to the particular effects of killing. More specifically,
in their model of war zone stressors and PTSD, Fontana and
Rosenheck (1999) demonstrated that the impact of killing differ-
entially impacted PTSD symptoms compared with other combat-
related factors, such as exposure to death or perceived threat.
While killing had a strong direct effect on PTSD, exposure to death
and injury and perceived threat did not. Our goal was to determine
whether these results would replicate in Gulf War veterans.
Thus far, the few studies that exist mainly examined the rela-
tionship between killing and PTSD, although alcohol problems and
depression also are common and important to assess in veterans
returning from overseas deployments. After controlling for general
combat experiences, Maguen and colleagues (2010, 2009) found
an association between killing and problem drinking among Viet-
nam and Iraq War veterans. Although PTSD and depression are
often highly comorbid, Maguen and colleagues (2010, 2009) did
not find an association between killing and depression among
either Vietnam or Iraq War veterans. Consequently, our goal was
to determine whether these alcohol and depression results would
replicate in Gulf War veterans.
There are a number of variables that are important to consider
when evaluating the mental health impact of killing. First, it is
This article was published Online First October 4, 2010.
Shira Maguen, Sara J. Knight, and Charles R. Marmar, San Francisco
VA Medical Center, San Francisco, California, and University of Califor-
nia, San Francisco; Dawne S. Vogt, Lynda A. King, Daniel W. King, and
Brett T. Litz, National Center for PTSD, Department of Veterans Affairs
Boston Healthcare System and Massachusetts Veterans Epidemiological
Research and Information Center, Boston, Massachusetts; Boston Univer-
sity School of Medicine.
This study was funded by a VA Health Services Research and Devel-
opment (HSR&D) Career Development Award (Shira Maguen) and by the
Department of Defense (U.S. Army Medical Research and Materiel Com-
mand) in collaboration with the Department of Veterans Affairs (DoD-87,
“Measurement and Validation of Psychosocial Risk and Resilience Factors
Associated with Physical and Mental Health and Health-Related Quality of
Life in Gulf War Veterans” (Daniel W. King). We thank Jeane Bosch and
Thomas Metzler for their assistance with this article.
Correspondence concerning this article should be addressed to Shira
Maguen, San Francisco VA Medical Center, PTSD Program (116-P), 4150
Clement Street, San Francisco, CA 94121. E-mail: shira.maguen@va.gov
Psychological Trauma: Theory, Research, Practice, and Policy In the public domain
2011, Vol. 3, No. 1, 21–26 DOI: 10.1037/a0019897
21
important to ensure that the impact of killing is not attributable to
related variables, such as exposure to death or witnessing killing.
For example, one study found that among Gulf War veterans, 52%
of men and 45% of women saw dead bodies or individuals who
were seriously maimed or injured, with 22% of men and 14% of
women witnessing someone dying (Carney et al., 2003). Another
study found that among soldiers serving in Operation Iraqi Free-
dom (OIF), 77% reported seeing dead bodies and 56% reported
witnessing killing (Maguen et al., 2010). Consequently, in this
study, we controlled for exposure to death and dying and witness-
ing killing in our examination of the association between killing
and a variety of postdeployment mental health measures.
Perceived threat is another variable that has been found to be
important in predicting PTSD among veterans and is often in-
cluded in comprehensive models of postdeployment functioning
(e.g., King, King, Foy, Keane & Fairbank, 1999; King, King,
Gudanowski & Vreven, 1995; Kulka et al., 1990). Perceived
threat, which is a subjective response to a hostile environment and
an important marker of peritraumatic response, is consistently
associated with the onset of adverse mental health symptoms
across samples (e.g., Holbrook, Hoyt, Stein, & Sieber, 2001;
Marmar, Weiss, Metzler, Ronfelt, & Foreman, 1996; McCaslin et
al., 2006). As a result, we also controlled for perceptions of danger
in our exploration of the impact of killing on postdeployment
mental health.
The purpose of this study was to examine the relationship
between killing and postdeployment mental health in Gulf War
veterans, after controlling for important variables such as per-
ceived danger, exposure to death and dying, and witnessing killing.
To our knowledge, this is the first study to examine the conse-
quences of taking another life among Gulf War veterans. Although
prior research has mainly focused on PTSD, our goal was to
replicate and extend these findings by examining the impact of
killing on additional postdeployment mental health measures, such
as depression and alcohol use. Identifying the impact of killing has
important implications for the evaluation and treatment of our
newly returning service members.
Method
Procedure and Participants
Participants were obtained via the Defense Manpower Data
Center and a national registry of Gulf War veterans. Mangione’s
(1998) multistep method was employed to optimize participant
response rate. First, potential participants were mailed a letter
explaining the study’s purpose. More specifically, participants
were told that the study would pertain to multiple aspects of their
deployment experiences. Confidentiality was assured, and the vol-
untary nature of participation was emphasized. Second, the origi-
nal letter was followed by a survey package containing a collection
of stressor and health outcome measures. Third, a reminder card
was sent, followed by a remailing of the package to nonrespon-
dents, and then a final reminder card. Of the 495 veterans who
were identified, 320 returned completed questionnaires in 2002.
Our study sample included 317 veterans who responded to the
question about killing in the war zone. Although posttraumatic
stress symptomatology (PTSS) was assessed among the entire
sample, depression and alcohol questionnaires were only admin-
istered to half the sample as part of a purposeful design to reduce
time burden on participants while still retaining a broad array of
measures. All procedures and measures were approved by the VA
Boston Healthcare System Institutional Review Board and con-
formed to standards for the protection of human subjects. Partic-
ipants were 64.8% Caucasian, 15.2% Black, 13.7% Hispanic, 1.6%
American Indian/Alaskan Native, and 1% Asian/Pacific Islander;
3.8% identified as Other. For other demographic characteristics of
participants, see Table 1.
Measures
Veterans reported age, gender, race/ethnicity, educational status,
and relationship status. They also were asked to respond to four
questions, each indexing a particular component of their war zone
experience: (a) perceived danger, (b) exposure to death and dying,
(c) witnessing killing of a fellow soldier, and (d) killing. Perceived
danger was assessed using the item, “I felt that I was in danger of
being killed or wounded” and was rated on a five-point Likert
scale, ranging from strongly disagree to strongly agree. Those
endorsing somewhat agree or strongly agree were rated as expe-
riencing perceived danger when percentages of individuals expe-
riencing each item were calculated, and this item was used as a
continuous variable in the regression equations. Exposure to death
and dying was assessed using the item, “I was exposed to the sight,
sound, or smell of dying men and women” and was rated on a
dichotomous scale. Witnessing killing of a fellow soldier was
assessed using the item, “I personally witnessed someone from my
unit or an ally unit being seriously wounded or killed” and was
rated on a dichotomous scale. Killing was assessed using the item,
“I killed or think I killed someone in combat” and also was rated
on a dichotomous scale.
PTSS was assessed using the PTSD Checklist, Military Version,
a 17-item measure (PCL; Blanchard, Jones-Alexander, Buckley, &
Forneris, 1996; Weathers et al., 1993). Each item was rated on a
five-point Likert scale, with responses ranging from not at all to
extremely, and participants were asked to rate PTSS over the last
3 months. The PCL is widely used as a screen for PTSD, has been
shown to have very good internal consistency, and correlates
strongly with other measures of PTSD symptoms (Weathers et al.,
1993). The PCL also demonstrates high diagnostic efficiency (i.e.,
.90; Blanchard et al., 1996). For the purposes of this study, we used
a recommended cutoff score of 50 when reporting a positive screen
for PTSD (Weathers et al., 1993) and a continuous symptom score
in our regression analysis. The internal consistency of the PCL for
our sample was .96.
Depression symptoms were assessed using an adapted version
of the Beck Depression Inventory-Primary Care (BDI-PC; Beck,
Steer, Ball, Ciervo, & Kabat, 1997). The original seven items from
the BDI-PC were used with a variation in the response format.
More specifically, participants were asked to rate symptoms over
the last three months (e.g., “In the last three months, I have felt like
a failure”), and unlike the original, each item was rated on a
five-point scale, with responses ranging from strongly disagree to
strongly agree. The BDI has excellent psychometric properties
(Beck, Steer, & Garbin, 1988). We used a cutoff score of four
when determining a positive screen for depression (Beck, Guth,
Steer, & Ball, 1997), with somewhat or strongly agree being used
as the threshold for a positive symptom score; we also include a
22 MAGUEN ET AL.
stricter criterion for depression, using only those who responded
strongly agree to at least four items. In our regression analysis, we
used a continuous depression symptom score. The internal consis-
tency of the BDI-PC for our sample was .91.
Alcohol use was assessed using two separate measures: (a) an
index of frequency and quantity of alcohol use and (b) the CAGE
(Ewing, 1984), a measure of problem alcohol use.
Frequency and quantity of alcohol use was assessed using the
following two questions: (a) “In the past three months how often
have you had a drink containing alcohol?” with responses rated on
a five-point scale ranging from never to four or more times a week
and (b) “In the past three months, how many drinks containing
alcohol have you had on a typical day when you were drinking?”
with responses rated on a five-point scale ranging from none to 7
or more. For the purposes of this study, a product of these two
questions was used in the multiple regression equation indexing
alcohol use, given that there is precedent for examining the product
term of alcohol frequency and quantity among veterans (e.g.,
Savarese, Suvak, King, & King, 2001).
The CAGE is a four-item measure in which participants were
asked four dichotomously rated questions related to their alcohol
use prior to deployment and in the present. Questions included
indicators of problem alcohol use, such as feelings of guilt in
reaction to drinking, criticism from others regarding drinking, and
wanting to cut down on drinking. For the purposes of this study,
responding positively to two or more questions in the present was
considered a positive screen for problem alcohol use (e.g., Hearne,
Connolly, & Sheehan, 2002), although we also report results for a
cut off score of one, given that there is also precedent of using a
lower threshold with military samples (e.g., LeardMann, Smith,
Smith, Wells, & Ryan, 2009). In our regression analyses, we used
a continuous measure of problem alcohol use prior to deployment
and in the present. The alpha reliability of this measure was .76
prior to deployment and .81 in the present.
Data Analysis
All of the analyses in this study were performed using the
statistical software package SPSS version 17.0 for Windows. First,
we calculated percentages of individuals who reported exposure to
indices of combat and killing in war. Next, we computed the
percentage of individuals in our sample who met screening criteria
for PTSD, depression, and problem alcohol use. Finally, we con-
ducted a series of regression analyses to identify predictors of each
of the postdeployment mental health measures.
We conducted four hierarchical regressions to determine
whether reported killing was significantly associated with PTSS,
depression, frequency and quantity of alcohol use, and current
problem alcohol use. In these analyses, we included related expo-
sure variables to ensure that the results were not attributable to
merely participating in combat (i.e., we controlled for perceived
danger, exposure to death and dying, and witnessing killing of a
fellow soldier).
In the first step of each regression, we entered demographic
variables (i.e., age, gender, race/ethnicity, educational status, and
relationship status). In the two regression equations related to
alcohol use, we also were able to control for prior problem drink-
ing in the first step. In the second step, we controlled for related
exposure variables; and in the third step we entered reported
killing.
Results
In this study, 46% of veterans reported perceiving danger during
their deployment, 42% reported exposure to death and dying, 19%
reported witnessing killing of a fellow soldier, and 11% reported
killing in combat. We compared those who endorsed killing to
those who did not and found that younger, F(1, 313) ⫽5.41, p⬍
.05, and male veterans,
2
(1) ⫽4.18, p⬍.05, were more likely
to report killing during the Gulf War than their counterparts.
Table 1
Sociodemographic Characteristics of Gulf War Veterans
n(%)
Killed (n⫽34) Did not kill (n⫽283) Total (n⫽317)
Age, M(SD)
ⴱ
41 (8.8) 45 (8.9) 44 (9.0)
Gender
ⴱ
Male 30 (88) 202 (72) 234 (74)
Female 4 (12) 79 (28) 83 (26)
Ethnicity
White 22 (65) 178 (65) 202 (65)
Minority 12 (35) 96 (35) 108 (35)
Education
Some high school 0 (0) 1 (.4) 1 (.3)
High school graduate 5 (15) 17 (6) 22 (7)
Vocation or technical training 2 (6) 27 (10) 30 (10)
Some college 15 (44) 119 (42) 135 (43)
Four-year college graduate 5 (15) 45 (16) 50 (16)
Some graduate or professional school 3 (9) 18 (6) 21 (7)
Graduate or professional degree 4 (12) 54 (19) 58 (18)
Marital status
Single 9 (27) 74 (26) 84 (26)
Married 25 (74) 207 (74) 233 (74)
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
23
KILLING IN WAR
Veterans also endorsed a wide range of mental health symp-
toms: 20% met threshold screening criteria for PTSS, 45% for
depression (11% using a stricter cutoff score; see measures sec-
tion), and 6% for current problem alcohol use (11% using a less
strict cutoff score; see measures section). Veterans reported drink-
ing an average of 2 to 3 times a month (SD ⫽1.35), with the mean
amount of alcohol consumed being 1 to 2 drinks per occasion
(SD ⫽1.08).
In the final hierarchical regression model predicting PTSS
(Table 2), older age, ethnic minority status, lower education, per-
ceived danger, and exposure to death and dying were each signif-
icant predictors. Reported killing remained significant, even after
controlling for perceived danger, exposure to death, and witness-
ing killing.
In the final hierarchical regression model predicting depression
symptoms (Table 2), being single, perceived danger, and exposure
to death and dying were each significant predictors. However,
neither witnessing killing nor reported killing were significant
predictors of depression symptoms.
In the final hierarchical regression models predicting current
frequency and quantity of alcohol use and current problem alcohol
use (Table 3), after controlling for prior problem drinking, reported
killing was the only significant predictor of each alcohol-related
measure.
We used the false discovery rate procedure specified by Benja-
mini and Hochberg (1995) to adjust for multiple comparisons in
our four outcomes, and we found that the three significant killing
findings each remained significant.
Discussion
Despite the fact that the Gulf War was a relatively short and
targeted mission, a significant number of veterans reported killing
during their deployment, with even larger numbers reporting ex-
posure to death and dying and witnessing killing of a fellow
soldier. Killing in combat was a significant predictor of PTSS and
multiple indicators of alcohol use, even after controlling for highly
salient variables such as perceived danger, exposure to death and
dying, and witnessing killing, suggesting that taking a life in
combat is a critical ingredient in the development of postdeploy-
ment mental health concerns.
Our finding that killing is associated with PTSS is consistent with
the few existing studies that evaluated the mental health impact of
taking a life (e.g., Fontana & Rosenheck, 1999; MacNair, 2002),
although this is the first study to document this association in Gulf
War veterans. This also is the first study demonstrating an asso-
ciation between killing and several indices of alcohol use in Gulf
War veterans. Of particular importance, killing was the only war
zone variable that significantly predicted alcohol use, even when
including important war zone experiences and controlling for
preexisting alcohol problems. Given recent findings of elevated
alcohol rates in veterans returning from modern deployments (e.g.,
Calhoun, Elter, Jones, Kudler, & Straits-Tröster, 2008; Maguen et
al., 2010), the impact of killing seems critical and may play an
important role in the evaluation and treatment of veterans with
problem alcohol use.
The relationship between killing and alcohol use may reflect a
method for veterans who are impacted by killing in war to regulate
difficult emotions. Indeed, prior researchers have postulated that
alcohol abuse may represent an effort to self-medicate avoidance
or hyperarousal symptoms associated with PTSD (Bremner et al.,
1996; Kulka et al., 1990). Another possibility is that the relation-
ship between killing in war and alcohol use is mediated by vari-
ables such as impulsivity or invincibility. A recent study postulated
that exposure to violent combat, including killing, may alter an
individual’s perceived threshold of invincibility, thereby increas-
ing the likelihood of engaging in high risk behavior, including
increased alcohol use (Killgore et al., 2008). It is important to note
that these authors did not find a relationship between killing and
impulsivity, suggesting that invincibility may be the important
link.
We did not find an association between killing and depression,
which is consistent with past studies (Maguen et al., 2010, 2009).
Although depression is highly associated with PTSD and the two
are often comorbid, when depression is examined as a whole, it is
not associated with killing. It is important to note that while PTSD
is always event-based, depression is not. A diagnosis of PTSD
cannot be given without a potentially traumatic event. Relatedly,
depression may be better predicted by other variables, such as
family history, rather than by killing. While killing and exposure to
killing were not strongly associated with depression, exposure to
Table 2
Hierarchical Regression Models of Posttraumatic Stress Symptomatology (PTSS) and Depression
Predictors
PTSS Depression
(Step 1) (Step 2) (Final model) R
2
(⌬R
2
)(Step 1) (Step 2) (Final model) R
2
(⌬R
2
)
1. Age .01 .10 .11
ⴱ
.14
ⴱⴱ
⫺.17 ⫺.07 ⫺.07 .10
ⴱⴱ
Gender .06 .08 .10 .05 .08 .08
Ethnicity .18
ⴱⴱ
.16
ⴱⴱ
.15
ⴱⴱ
⫺.02 ⫺.02 ⫺.02
Education ⫺.26
ⴱⴱ
⫺.17
ⴱⴱ
⫺.17
ⴱⴱ
⫺.15 ⫺.10 ⫺.10
Marital status ⫺.13
ⴱ
⫺.09 ⫺.09 ⫺.15 ⫺.16
ⴱ
⫺.16
ⴱ
2. Perceived danger .32
ⴱⴱ
.31
ⴱⴱ
.37
ⴱⴱ
(.23)
ⴱⴱ
.18
ⴱ
.18
ⴱ
.23
ⴱⴱ
(.13)
ⴱⴱ
Exposure to death .25
ⴱⴱ
.23
ⴱⴱ
.29
ⴱⴱ
.29
ⴱⴱ
Witnessing killing .12
ⴱ
.10 .05 .05
3. Killing .11
ⴱ
.38
ⴱⴱ
(.01)
ⴱ
.00 .23
ⴱⴱ
(.00)
ⴱ
Note. For final models, F(9, 297) ⫽20.05, p⬍.01 for PTSS; F(9, 148 ⫽4.86, p⬍.01 for Depression, Ethnicity: 0 ⫽Caucasian,1⫽ethnic minority;
Gender: 0 ⫽male,1⫽female; Spouse: 0 ⫽single,1⫽married/in a relationship. Numbers vary because of missing data.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
24 MAGUEN ET AL.
death seemed to be a strong predictor, suggesting that some aspects
of exposure to combat increase risk for depression, while others do
not.
There are several limitations of this study that should be noted.
First, the current study was retrospective; consequently recall bias
should be taken into account when interpreting these results.
Relatedly, because these data were collected several years after the
war, measures may reflect more chronic symptomology. Second,
this investigation was conducted with American Gulf War veter-
ans; therefore these results may not generalize to veterans of other
wars. Additionally, although this was a national sample, results are
not necessarily representative of the population of Gulf war vet-
erans. Third, it is important to account for the fact that our
postdeployment mental health measures were self-report and used
for screening rather than as diagnostic instruments. Despite the fact
that clinically significant cutoff scores have been recommended
for each of these measures, these results should be replicated with
clinician-rated diagnostic tools (e.g., Clinician Administered
PTSD Scale for DSM–IV; Blake et al., 1995). Exposure variables,
including our index of killing, were assessed using a single indi-
cator, and future studies should replicate these results with more
comprehensive measures. This is especially important in order to
better understand which aspects of killing (e.g., circumstances of
killing, person killed, etc.) are associated with increased mental
health problems. However, it is important to note that there is
precedent for examining single exposure items (e.g., LeardMann,
Smith, Smith, Wells, & Ryan, 2009). Although we were able to
index predeployment alcohol use, a limitation is that these reports
were retrospective; additionally we were not able to control for
predeployment PTSD or depression. Future investigations that
utilize broader sets of health and deployment information are
needed to further understand potential mediators and moderators in
this model, such as prior mental health difficulties and prior
trauma.
Our finding that killing in war is a significant, independent
predictor of multiple mental health symptoms has important im-
plications for the health care of veterans. A comprehensive eval-
uation of veterans returning from combat should include an as-
sessment of killing experiences as well as reactions to killing and
precursors to killing, including witnessing the death of a fellow
soldier and perceived danger, factors that may place veterans at
even greater risk of developing mental health complications such
as PTSD. This information can be used to inform larger conceptual
models of mental health response, as we continue to expand our
understanding of how veterans are impacted by taking a life. For
example, the experience of killing also may be associated with
moral injury (Litz et al., 2009) and changes in spirituality/
religiosity (e.g., Fontana & Rosenheck, 2004). It is critical that
future research examine the broad impact of taking another life in
combat.
Overall, we found that a significant percentage of veterans
serving in the Gulf War reported killing, which places them at risk
for PTSD and elevated alcohol use. Including killing experiences
as part of postdeployment evaluation and treatment planning will
ensure that we are providing comprehensive health care to our
modern deployment veterans as they embark on the journey of
reintegration and readjustment to civilian life.
References
Beck, A. T., Guth, D., Steer, R. A., & Ball, R. (1997). Screening for major
depression disorders in medical inpatients with the Beck Depression
Inventory for Primary Care. Behaviour Research and Therapy, 35,
785–791.
Beck, A. T., Steer, R. A., Ball, R., Ciervo, C. A., & Kabat, M. (1997). Use
of the Beck Anxiety and Depression Inventories for primary care with
medical outpatients. Assessment, 4, 211–219.
Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties
of the Beck Depression Inventory: Twenty-five years later. Clinical
Psychology Review, 8, 77–100.
Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery
rate: A practical and powerful approach to multiple testing. Journal of
the Royal Statistical Society Series B, 57, 289 –300.
Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman,
F. D., Charney, D. S., et al. (1995). The development of a clinician-
administered PTSD scale. Journal of Traumatic Stress, 8, 75–90.
Blanchard, E. B., Jones-Alexander, J., Buckley, T. C., & Forneris, C. A.
(1996). Psychometric properties of the PTSD Checklist. Behaviour
Research and Therapy, 13, 669 – 673.
Bremner, J. D., Southwick, S. M., Darnell, A., & Charney, D. S. (1996).
Table 3
Hierarchical Regression for Models of Alcohol Use and Problem Alcohol Use
Predictors
Alcohol use Problem alcohol use
(Step 1) (Step 2) (Final model) R
2
(⌬R
2
)(Step 1) (Step 2) (Final model) R
2
(⌬R
2
)
1. Age ⫺.01 ⫺.03 ⫺.01 .07 .01 .02 .04 .09
ⴱ
Gender ⫺.16 ⫺.17 ⫺.13 ⫺.09 ⫺.08 ⫺.04
Ethnicity ⫺.01 ⫺.02 ⫺.05 .16
ⴱ
.16 .13
Education ⫺.04 ⫺.04 ⫺.05 .03 .03 .02
Marital status ⫺.15 ⫺.16 ⫺.15 ⫺.16 ⫺.17
ⴱ
⫺.16
Prior alcohol use .13 .15 .17
ⴱ
.18
ⴱ
.19
ⴱ
.22
ⴱⴱ
2. Perceived danger .01 ⫺.02 .09 (.01) .00 ⫺.03 .10 (.01)
Exposure to death ⫺.05 ⫺.08 .07 .04
Witnessing killing ⫺.10 ⫺.14 ⫺.06 ⫺.10
3. Killing .20
ⴱ
.12
ⴱ
(.03)
ⴱ
.21
ⴱ
.13
ⴱ
(.04)
ⴱ
Note. For final models, F(10, 145) ⫽1.92, p⬍.05, for alcohol use; F(10, 145) ⫽2.23, p⬍.05, for problem alcohol use. Ethnicity: 0 ⫽Caucasian,
1⫽ethnic minority; Gender: 0 ⫽male,1⫽female; Spouse: 0 ⫽single,1⫽married/in a relationship.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
25
KILLING IN WAR
Chronic PTSD in Vietnam combat veterans: Course of illness and
substance abuse. American Journal of Psychiatry, 153, 369 –375.
Calhoun, P. S., Elter, J. R., Jones, E. R., Kudler, H., & Straits-Tröster, K.
(2008). Hazardous alcohol use and receipt of risk-reduction counseling
among U.S. veterans of the wars in Iraq and Afghanistan. Journal of
Clinical Psychiatry, 69, 1686 –1693.
Carney, C. P., Sampson, T. R., Voelker, M., Woolson, R., Thorne, P., &
Doebbeling, B. N. (2003). Women in the Gulf War: Combat experience,
exposures, and subsequent health care use. Military Medicine, 168,
654 – 661.
Ewing, J. A. (1984). Detecting alcoholism, the CAGE questionnaire.
Journal of the American Medical Association, 252, 1905–1907.
Fontana, A., & Rosenheck, R. (2004). Trauma, change in strength of
religious faith, and mental health service use among veterans treated for
PTSD. Journal of Nervous and Mental Disease, 192, 579 –584.
Fontana, A., & Rosenheck, R. (1999). A model of war zone stressors and
posttraumatic stress disorder. Journal of Traumatic Stress, 12, 111–126.
Hearne, R., Connolly, A., & Sheehan, J. (2002). Alcohol abuse: Prevalence
and detection in a general hospital. Journal of the Royal Society of
Medicine, 95, 84 – 87.
Hoge, C. W., Castro, C. A., Messer, S. C., McGurk, D., Cotting, D. I., &
Koffman, R. L. (2004). Combat duty in Iraq and Afghanistan, mental
health problems, and barriers to care. New England Journal of Medicine,
351, 13–22.
Holbrook, T. L., Hoyt, D. B., Stein, M. B., & Sieber, W. J. (2001).
Perceived threat to life predicts posttraumatic stress disorder after major
trauma: Risk factors and functional outcome. The Journal of Trauma,
51, 287–292.
Killgore, W. D., Cotting, D. I., Thomas, J. L., Cox, A. L., McGurk, D., Vo,
A. H., et al. (2008). Post-combat invincibility: Violent combat experi-
ences are associated with increased risk-taking propensity following
deployment. Journal of Psychiatric Research, 42, 1112–1121.
King, D. W., King, L. A., Foy, D. W., Keane, T. M., & Fairbank, J. A.
(1999). Posttraumatic stress disorder in a national sample of female and
male Vietnam Veterans: Risk factors, war-zone stressors, and resilience-
recovery variables. Journal of Abnormal Psychology, 108, 164 –170.
King, D. W., King, L. A., Gudanowski, D. M., & Vreven, D. L. (1995).
Alternative representations of war zone stressors: Relationships to Post-
traumatic Stress Disorder in male and female Vietnam veterans. Journal
of Abnormal Psychology, 104, 184 –195.
Kulka, R. A., Schlenger, W. E., Fairbank, J. A., Hough, R. L., Jordan,
B. K., Marmar, C. R., et al. (1990). Trauma and the Vietnam War
generation: Report of the findings from the National Vietnam Veterans
Readjustment Study. New York: Brunner/Mazel.
LeardMann, C. A., Smith, T. C., Smith, B., Wells, T. S., & Ryan, M. A. K.
(2009). Baseline self reported functional health and vulnerability to
post-traumatic stress disorder after combat deployment: Prospective
U.S. military cohort study. British Medical Journal, 338, 1273.
Litz, B. T., Stein, N., Delaney, E., Lebowitz, L., Nash, W. P., Silva, C., &
Maguen, S. (2009). Moral injury and moral repair in war veterans: A
preliminary model and intervention strategy. Clinical Psychology Re-
view, 29, 695–706.
MacNair, R. M. (2002). Perpetration-inducted traumatic stress in combat
veterans. Peace and Conflict: Journal of Peace Psychology, 8, 63–72.
Maguen, S., Lucenko, B. A., Reger, M. A., Gahm, G. A., Litz, B. T., Seal,
K. H., et al. (2010). The impact of reported direct and indirect killing on
mental health symptoms in Iraq War veterans. Journal of Traumatic
Stress, 23, 86 –90.
Maguen, S., Metzler, T. J., Litz, B. T., Seal, K. H., Knight, S. J., &
Marmar, C. R. (2009). The impact of killing in war on mental health
symptoms and related functioning. Journal of Traumatic Stress, 22,
435– 443.
Mangione, T. (1998). Mail surveys. In L. Bickman & D. J. Rog (Eds.),
Handbook of applied social research methods. Thousand Oaks, CA:
Sage.
Marmar, C. R., Weiss, D. S., Metzler, T. J., Ronfeldt, H. M., & Foreman,
C. (1996). Stress responses of emergency services personnel to the Loma
Prieta earthquake Interstate 880 freeway collapse and control traumatic
incidents. Journal of Traumatic Stress, 9, 63– 85.
McCaslin, S. E., Rogers, C. E., Metzler, T. J., Best, S. R., Weiss, D. S.,
Fagan, J. A., et al. (2006). The impact of personal threat on police
officers’ responses to critical incident stressors. Journal of Nervous and
Mental Disease, 194, 591–597.
Savarese, V. W., Suvak, M. K., King, L. A., & King, D. W. (2001).
Relationships among alcohol use, hyperarousal, and marital abuse and
violence in Vietnam veterans. Journal of Traumatic Stress, 14, 717–732.
Weathers, F. W., Litz, B. T., Herman, J. A., Huska, J. A., & Keane, T. M.
(1993). The PTSD Checklist (PCL): Reliability, validity and diagnostic
utility in 9th Annual Conference of the ISTSS. San Antonio, TX.
Received August 18, 2009
Revision received March 8, 2010
Accepted March 18, 2010 䡲
26 MAGUEN ET AL.
