Journal of Psychosomatic Research 78 (2015) 478–483

Contents lists available at ScienceDirect

Journal of Psychosomatic Research

Indirect associations of combat exposure with post-deployment physical

symptoms in U.S. soldiers: Roles of post-traumatic stress disorder,
depression and insomnia☆
Phillip J. Quartana ⁎, Joshua E. Wilk, Thomas J. Balkin, Charles W. Hoge
Center for Military Psychiatry and Neuroscience, Walter Reed Army Institute of Research, United States

a r t i c l e i n f o a b s t r a c t

Article history: Objective: To characterize the indirect associations of combat exposure with post-deployment physical
Received 30 June 2014 symptoms through shared associations with post-traumatic stress disorder (PTSD), depression and insomnia
Received in revised form 17 November 2014 symptoms.
Accepted 20 November 2014 Methods: Surveys were administered to a sample of U.S. soldiers (N = 587) three months after a 15-month de-
ployment to Iraq. A multiple indirect effects model was used to characterize direct and indirect associations be-
Keywords:
tween combat exposure and physical symptoms.
Combat exposure
Physical symptoms
Results: Despite a zero-order correlation between combat exposure and physical symptoms, the multiple indirect
PTSD effects analysis did not provide evidence of a direct association between these variables. Evidence for a significant
Insomnia indirect association of combat exposure and physical symptoms was observed through PTSD, depression, and in-
Depression somnia symptoms. In fact, 92% of the total effect of combat exposure on physical symptoms scores was indirect.
Pain These findings were evident even after adjusting for the physical injury and relevant demographics.
Military Conclusion: This is the first empirical study to suggest that PTSD, depression and insomnia collectively and inde-
pendently contribute to the association between combat exposure and post-deployment physical symptoms.
Limitations, future research directions, and potential policy implications are discussed.
Published by Elsevier Inc.

Introduction concussion/mild traumatic brain injury (mTBI), to which many post-

Post-deployment physical symptoms are commonly reported by
war veterans [1–4]. The prevalence of physical symptoms following
combat deployment was intensively studied in the context of reports
of ‘Gulf War Syndrome.’ Although no empirical confirmation of a specif-
ic syndrome has been identified, studies have consistently revealed as-
sociations between Gulf War deployment and physical symptoms
spanning multiple health domains. Studies of the most recent conflicts
in Iraq and Afghanistan also provide evidence of high rates of mental
and physical health problems, with rates of generalized physical health
problems considerably higher in service members who have deployed
compared with those who have not [1,4].
Determining mechanisms by which combat deployment contributes
to physical symptoms (independent of physical injury) has proven
challenging. The most recent wars have focused attention primarily on
☆ Disclaimer: The opinions or assertions contained herein are the private views of the
authors, and are not to be construed as official, or as reflecting true views, of the
Department of the Army or the Department of Defense.
⁎ Corresponding author at: Center for Military Psychiatry and Neuroscience, Walter
Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, MD 20910,
United States. Tel.: +1 301 319 9777; fax: +1 301 319 9484.
E-mail address: phillip.j.quartana2.civ@mail.mil (P.J. Quartana).
deployment physical symptoms have been attributed. Most studies
have failed to support this association, and instead suggest that PTSD
and depression are more robust correlates of persistent physical
symptoms in service members and veterans with a history of mTBI
[5–7]. Although there are notable differences in the combat experiences,
environmental exposures, operational tempo, and mission characteris-
tics across wars throughout history, it is perhaps telling that each, up
to and including the most recent wars in Iraq and Afghanistan, has
been associated with similar generalized post-deployment health con-
cerns [2–4,8]. These data collectively suggest that a common set of
deployment-related factors contribute to post-combat generalized
physical health problems. However, debate has existed after every war
as to the relative contribution of physical, psychological, or environ-
mental causes of these post-war health concerns. We hypothesize that
mental health conditions are essential considerations for fully explicat-
ing the association between combat exposure and physical symptoms.
Collectively, PTSD, depression and insomnia are associated with endo-
crine [9–11], inflammatory [12–16] and autonomic nervous system [9,
17] dysregulation, each of which could contribute to general physical
health symptoms. The purpose of the present study is to elucidate the
relationships among these variables with the aim of determining their
role in the development of post-deployment physical symptoms — an
important step toward development of optimal treatment strategies.

http://dx.doi.org/10.1016/j.jpsychores.2014.11.017
0022-3999/Published by Elsevier Inc.
 P.J. Quartana et al. / Journal of Psychosomatic Research 78 (2015) 478–483
Previous studies suggest that the degree of combat exposure is pre-
dictive of post-deployment physical symptomology [2,4,18–21]. Addi-
tional research is needed to more fully characterize the nature of this
association. Studies have rarely systematically examined indirect asso-
ciations of combat exposure and physical symptoms through highly
prevalent mental health consequences of combat deployment, and
studies that have done so have typically included one mental health in-
dicator, such as PTSD, in the analysis [21]. It is estimated that 6–13% of
soldiers who have returned from Iraq or Afghanistan meet criteria for
post-traumatic stress disorder (PTSD), and a high rate of major depres-
sion disorder has also been documented [22]. Both of these conditions
have been strongly correlated with post-deployment physical symp-
toms (e.g., [3]). Therefore, we hypothesize that the association between
combat exposure and physical symptoms is at least in part attributable
to variance shared in common with PTSD and/or depression symptoms.
Indeed, it has been hypothesized elsewhere that PTSD can account for
the effect of prior combat experience on physical health outcomes [21,
23], and numerous studies have shown that PTSD accounts for greater
variance in persistent post-concussive symptoms than mTBI itself
(e.g., [7]). Additional research is needed to determine the unique and
combined contributions of these factors to the association between
combat exposure and post-deployment physical symptoms.
Sleep disturbance and insomnia are prevalent in combat theater and
in garrison [24–26], and have been recognized and highlighted as one of
three critical areas of focus for improved soldier health and fitness,
as outlined in the Army Surgeon General's ‘Soldier Performance
Triad’. To date, sleep has received little consideration as a risk factor
for post-deployment physical symptoms. Prior studies examining
post-deployment correlates of physical health have characterized and
operationally defined sleep disturbance as but one component of the
physical symptom milieu [1,3,7,27]. However, there exists compelling
evidence from studies of civilian populations that sleep disturbance
might play an important role in the development, maintenance and exac-
erbation of physical health problems. Sleep disturbance can be triggered
and/or exacerbated by stress, and has been linked to physical symptom
complaints in a number of field and laboratory studies [28]. Sleep distur-
bance has also been shown to enhance sensitivity to noxious stimuli in
otherwise healthy individuals [29,30], and objective polysomnography
indicators of sleep disturbance are common among those with chronic
pain [31,32] and other physical health problems [33–35].
In the present study, we hypothesized that PTSD, depression and in-
somnia symptoms would emerge as important contributory factors to
the link between combat exposure and post-deployment physical
symptoms. Because insomnia is a phenomenological correlate of PTSD
and depression, and may even share some common biological sub-
strates with PTSD and depression [36–38], we carefully modeled the rel-
ative, or unique, contribution of each of these factors on the combat–
physical symptom link. To do so, we utilized a multiple indirect effects
model [39] that allowed us to systematically evaluate the direct associ-
ation of combat exposure with physical symptoms, as well as whether
this association was indirect and attributable to variance shared in com-
mon with PTSD, depression, and insomnia symptoms (i.e., indirect
effects). We examined these associations using an extant dataset from
a large cohort of U.S. soldiers following a 15-month deployment to Iraq.
Methods
Participants and procedures
Participants were 587 U.S. soldiers who completed a 15-month
deployment to Iraq early in support of Operation Iraqi Freedom (OIF)
in 2003–2004. Data for this study were collected as part of the Land
Combat Study (LCS), a large multi-year investigation of the health con-
sequences of the Iraq and Afghanistan wars on soldiers and their fami-
lies. Response rates for 2003–2007 were approximately 62% [22].
Comparison of data between responders and non-responders was not
479
possible because only responders were given an opportunity to provide
informed consent for using their data for research purposes. However,
data from LCS have been highly comparable with the larger deployed
population based on post-deployment health assessments [22]. Data
collections were scheduled in the units in consultation with unit
leaders. All data collections occurred 3 months following return from
deployment. This post-deployment assessment window corresponds
with post-deployment health assessments [40]. Survey data collection
procedures in the present study were consistent with those reported
in other LCS publications [2,5,22]. For this study, we included question-
naires that assessed demographics, combat exposure, and PTSD, de-
pression, insomnia, and physical symptoms. Informed consent was
obtained from all study participants after receiving an in-person brief-
ing and prior to completing the survey. Study procedures were
approved by the Institutional Review Board at the Walter Reed Army
Institute of Research.
Survey instruments
Combat exposure was assessed using a 35-item scale used in prior
studies [41–43]. Three items were not included in the analyses. One of
these items assessed a positive combat experience (i.e., saved
someone's life). A second was an open-ended “other” option and thus
could not be standardized. A third item pertained to whether a soldier
experienced a combat-related injury. This item was evaluated as a co-
variate (see below). Soldiers responded to each combat exposure item
on a 1 (never) to 5 (ten or more times) categorical scale. To compute
an overall combat exposure indicator, responses from the remaining
32 items were dichotomized as ‘yes’ or ‘no,’ indicating whether a soldier
had experienced an event at least once during deployment, and then
summed to create a single index of combat exposure severity. This com-
putational approach taps the breadth of combat exposure [1,41,44,45].
The combat exposure scale showed excellent internal consistency,
Cronbach's α = .93.
Physical symptoms were assessed using 12 items from the 15-item
Patient Health Questionnaire (PHQ-15; [46]). Two items not included
in the present analysis were difficulty falling or staying asleep and feel-
ing tired because the item content overlap with our measure of insom-
nias (see below). A third item pertaining to menstrual cramps was also
excluded as it applied only to females. For each symptom, soldiers indi-
cated the extent to which they were bothered over the past month.
Reponses were provided on a 0 (not bothered) to 2 (bothered a lot)
scale. Item responses were summed to create an overall physical symp-
toms score. Cronbach's α = .84 for the PHQ-15.
PTSD symptoms were assessed using the 17-item National Center for
PTSD Checklist — (PCL; [47]. The PCL has been validated in civilian and
military populations and possesses acceptable levels of concurrent va-
lidity with structured interview assessments of PTSD [48]. Soldiers pro-
vided responses on a 1 (not at all) to 5 (extremely) scale indicating the
degree to which each symptom bothered them over the past month.
The nightmare and insomnia items were not included in the computa-
tion of the PCL total score, which was a sum of responses to the remain-
ing items, in an effort to minimize inflated covariance attributable to
shared item content between PCL and our insomnia measure. For
PTSD caseness, a cut-off score of 50 or greater was used [22,48,49].
Note that we included the sleep item in the caseness definition so as
not to underestimate probable PTSD. Internal consistency of the PCL
was excellent, Cronbach's α = .96.
Depression symptoms were assessed using the 9-item depression
subscale from the Patient Health Questionnaire (PHQ-9; [50,51]),
which has excellent validation compared with structured diagnostic in-
terviews for major depression [50]. Soldiers provided responses indicat-
ing how much each symptom bothered them over the past month on a 0
(not at all) to 3 (nearly every day) scale. The PHQ-9 item concerning
insomnia/hypersomnia was omitted to minimize covariance between
the PHQ-9 and our measure of insomnia. The remaining 8 items were
480 P.J. Quartana et al. / Journal of Psychosomatic Research 78 (2015) 478–483
summed to create an overall indicator of depression symptoms. For de-
pression caseness, participants were considered probably depression if
they endorsed 5 items, including “feeling depressed or hopeless” or
“having little interest in doing things” for more than half of the days in
the past month [22]. Note that we included the sleep item in the
caseness definition. The PHQ-9 had adequate internal consistency,
Cronbach's α = .90.
Insomnia symptoms were assessed using the insomnia severity index
(ISI; [52]). The ISI is a 5-item questionnaire that evaluates: [1] the sever-
ity of sleep-onset, sleep maintenance and early morning awakening
problems, [2] satisfaction with current sleep pattern, [3] sleep-related
interference with daily function, [4] noticeability of impairment attrib-
uted to sleep problems, and [5] level of distress caused by sleep prob-
lems. Participants rate each of these factors on a 5-point (0–4) scale,
with possible scores ranging from 0 (No Clinically Significant Insomnia)
to 28 (Severe Clinical Insomnia). The measure has adequate psychomet-
ric characteristics [52–54]. Participants were asked to provide responses
to each item corresponding to, “over the past two weeks.” For insomnia
caseness, a standardized cut-off score of 15 or greater was used [52]. The
ISI possessed adequate internal consistency, Cronbach's α = .84.
Data analysis
Descriptive statistics were computed to examine sample demo-
graphic characteristics and to ascertain sample mean and standard devi-
ations on each study measure. We also present caseness for probable
PTSD, depression, and insomnia. Zero-order correlations were comput-
ed to examine basic associations between combat exposure and PHQ-
15, PCL, PHQ-9 and ISI scores. Direct and indirect associations of combat
stress with physical symptoms were examined using a bias-corrected
and bootstrapped multiple indirect effects model [39]. Fig. 1 provides
a graphical display of the model that we examined (including coeffi-
cients for a possible set of covariates). Bootstrapping is a nonparametric
re-sampling procedure that is robust to violations of the assumption of
normality for the indirect effect, or a ∗ b path coefficient. It involves
Combat c Physical
Exposure Symptoms
PTSD Symptoms
b1
a1
Depression
a2 b2
Symptoms
c’
Combat Physical
Exposure Symptoms
a3
Insomnia b3
Symptoms
Covariates
(e.g., age, injury)
Fig. 1. Graphical depiction of the multiple indirect effects model. Notes: c = total effect of
combat stress on physical symptoms; c′ = direct effect of combat stress on physical symp-
toms; a1 = direct effect of combat stress on PTSD symptoms; b1 = direct effect of PTSD
symptoms on physical symptoms; a2 = direct effect of combat stress on PTSD symptoms;
b2 = direct effect of depression symptoms on physical symptoms; a3 = direct effect of
combat stress on PTSD symptoms; b3 = direct effect of insomnia on physical symptoms.
Unique indirect effects path coefficients are computed via the aj ∗ bj path coefficient
cross-product. The total indirect effect path coefficient is computed by summing each
unique aj ∗ bj cross-product coefficient.
random and repeated sub-sampling of the data (usually a minimum of
k = 1000 bootstraps is recommended) to derive an estimate of the
a ∗ b (indirect effect) path coefficient and a 95% confidence interval
around the indirect effect point estimate. The a ∗ b path coefficient
should not be submitted to statistical significance testing based on nor-
mal theory because this estimate is positively skewed. Submitting the
a ∗ b estimate to normal theory significance tests leads to an underesti-
mation of the indirect effects.
The utilization of this approach to multiple indirect effects modeling
afforded some noteworthy advantages over separate tests of simple in-
direct effects. First, it allowed us to examine a total indirect effect, which
can tell us whether the (additive) combination of factors accounted for
an association between combat exposure and PHQ-15 scores. Second, it
allowed for an examination of a ∗ b (indirect effects) estimates for each
indirect effect factor under consideration conditional on the inclusion of
the other proposed indirect effects factors; that is, it provided factor-
specific (i.e., unique) a ∗ b (indirect effects) estimates and 95% confi-
dence intervals for each. We considered this a critical element of our
analysis because of the conceptual and statistical overlap between
PTSD, depression, and insomnia symptoms. Conducting numerous sim-
ple indirect effects models would not allow us to make conditional
statements concerning the unique and relative effects of one indirect ef-
fect relative to another. Third, it allowed us to conduct pairwise compar-
isons between the a ∗ b (indirect association) path coefficients, which
permitted statistical tests of the relative contribution of each indirect ef-
fect variable to the association between combat exposure and PHQ-15
scores. Lastly, deriving point estimates in a single model versus multiple
models mitigated Type I error rate inflation. Bias-corrected bootstraps
were used to derive path coefficients and corresponding 95% confidence
intervals for total indirect and individual indirect effects (k = 1000
bootstraps were used). Effect size estimates are provided as the ratio
of the total indirect effect to the total effect (path c, as shown in
Fig. 1). This represents the proportion of the total effect of that is medi-
ated by PTSD, depression, and insomnia symptoms considered together
[55]. We also present individual indirect effect size estimates for PTSD,
depression, and insomnia symptoms. Finally, we present a ratio of indi-
vidual indirect effect estimates to the total indirect effect, as well as
pairwise comparisons of the individual indirect effects to examine a rel-
ative contribution of each variable to the overall indirect effect of com-
bat exposure on PHQ-15 scores.
Results
Descriptive statistics
The majority of the soldiers assessed were men (82.4%). Most soldiers (63.8%) were
between 18 and 29 years of age, with 29.6% being 30–39 years and 6.6% 40 years or
older; 56.1% of the sample was Caucasian; 41.2% had received high school diploma/GED
or at least attended some high school, 44% attended some college or possessed an
Associate's degree, and 14% had a Bachelor's or graduate degree. The majority of the sam-
ple was E1–E4 (53%), with 36% E5–E9 and 10% commissioned officers. These data, and
other data derived from LCS surveys, are demographically representative of the infantry
population for the Army at-large [22,56]. Mean combat exposure was 10.86 combat events
(SD = 6.93). The mean PHQ-15 score was 5.34 (SD = 4.36); mean PCL score was 26.19
(SD = 12.53); mean PHQ-9 score was 4.73 (SD = 5.01); and mean ISI score was 7.47
(SD = 6.38). Approximately 10% (n = 66) of the sample reported being wounded or in-
jured during combat. Soldiers who were injured had higher PCL and PHQ-9 scores (p-
values b .05). Injury was unrelated to ISI and PHQ-Physical Symptom scores (p-values N
.10). Approximately 11% of the sample met self-reported criteria for PTSD; 9.7% for depres-
sion; and 15.5% for insomnia.
Zero-order correlations
Zero-order correlations among study variables are provided in Table 1. Combat expo-
sure scores were positively correlated with PCL, PHQ-9, PHQ-15 and ISI scores. The percent
shared variance between these measures ranged from 3.6% to 13.1%. These findings sug-
gest a zero-order association between combat exposure and PTSD and depression symp-
toms, insomnia and physical symptoms. Not surprisingly, strong positive correlations
were observed between PCL and PHQ-9 scores. ISI scores were also correlated with PCL,
PHQ-9 and PHQ-15 scores. That these factors are correlated provides empirical support
 P.J. Quartana et al. / Journal of Psychosomatic Research 78 (2015) 478–483 481

Table 1 indirect effects of combat exposure on PHQ-15 scores through each of the factors under
Zero-order correlations among main study variables (N = 587). consideration, as evidenced by bias-corrected bootstrapped 95% confidence intervals
that did not contain zero. The ratio of the total effect of combat exposure on PHQ-15 scores
1 2 3 4 5 was .29, .37, and .26 for PCL, PHQ-9 and ISI scores, respectively. PCL, PHQ-9, and ISI scores
1. Combat exposure – accounted for 27%, 46%, and 27% of the total indirect association of combat exposure on
2. PHQ-15 .19 – PHQ-15 scores. Pairwise comparisons of the independent a ∗ b paths revealed that PCL,
3. PCL .36 .53 – PHQ-9 and ISI scores did not differentially account for the indirect association of combat
4. PHQ-9 .21 .56 .75 – exposure on PHQ-15 scores, as evidenced by bias-corrected bootstrapped 95% confidence
5. ISI .22 .45 .59 .58 – intervals that contained zero (see Table 2). It is worth noting that all of the analyses report-
ed herein were also conducted using sleep disturbance estimates only, derived by sum-
Note. All correlations were statistically significant at p b .05 (two-tailed). ming responses to the sleep continuity items from the ISI (i.e., difficulty falling asleep,
difficulty staying asleep, and waking up too early). Results of these analyses were nearly
identical with those reported. We also computed models using a sum score of combat ex-
posure items as the independent variable. Again, the results were highly similar to those
for the use of a multiple indirect effects approach versus multiple tests of simple indirect
reported.
effects.

Multiple indirect effects analysis

Discussion
Initial models were conducted in which a set of possible covariates were included in
the model, including age, sex, rank, ethnicity, educational attainment and whether a sol- The purpose of this study was to examine direct and indirect associ-
dier reported being wounded or injured during combat. Results of these models failed to
ations between combat exposure and self-reported physical symptoms
reveal partial effects of ethnicity, educational attainment and physical injury on PHQ-15
scores (all p-values N .10). Hence, these factors were not considered further. However, par- among U.S. soldiers after a 15-month combat deployment. We were
tial effects (p b .05) emerged for sex and age on PHQ-15 scores. Specifically, male soldiers specifically interested in examining the common and unique contribu-
had lower PHQ-15 scores than female soldiers and age was positively associated with tions of PTSD, depression and insomnia symptoms to the association be-
PHQ-15 scores. However, findings from a sex- and age-adjusted model did not deviate tween combat exposure and physical symptoms. Despite evidence for a
from the unadjusted model. Based on findings from these initial exploratory covariate
analyses, we reported path coefficients from the unadjusted model only.
zero-order correlation between combat exposure and physical symp-
Results of the model are presented in Table 2. Of particular note, the direct effect of toms, our analysis did not provide evidence for a direct association be-
combat exposure on PHQ-15 scores was not statistically significant (p N .10). However, tween these variables when other variables were considered in the
combat exposure was associated with PCL, PHQ-9 and ISI scores (p-values ≤ .03). More- model. Consistent with our hypotheses, we found clear evidence for in-
over, statistically significant direct associations with PHQ-15 scores were observed for
direct associations of combat exposure on post-deployment physical
PCL, PHQ-9 and ISI scores. The total a ∗ b (indirect effect) path was significant, as evidenced
by the bias-corrected bootstrapped 95% confidence interval that did not contain zero. The symptoms through greater PTSD, depression and insomnia symptoms.
ratio of the total indirect association to the total effect of combat exposure on PHQ-15 Critically, each of the mental health problems accounted for unique var-
scores was .92, suggesting that 92% of the effect of combat exposure on PHQ-15 scores iability in the combat exposure–physical symptom association. These
was indirect and accounted for by the set of PCL, PHQ-9, and ISI scores. Inspection of the findings support the hypothesis that this set of common deployment-
individual conditional a ∗ b (indirect effect) estimates revealed statistically significant
related factors plays an important role, collectively and independently,
in the association between combat exposure and post-deployment
Table 2
physical symptoms.
Bias corrected bootstrapped multiple indirect effects of combat exposure on physical Previous research concerning post-deployment physical symptoms
symptoms through PTSD symptoms, depression symptoms and insomnia (N = 587). in the context of the Iraq and Afghanistan wars has provided evidence
that PTSD and depression are important factors related to physical
Criterion = PHQ-15a Point SE t-Test Bias corrected
estimateb statisticc 95% CI symptoms [2–4,21]. Indeed, Hoge et al. [3] underscored a strong associ-
ation between PTSD and physical symptom complaints and overall rat-
Lower Upper
ings of physical health among veterans of the Iraq war. Other studies
Total effect .12 .03 4.78⁎ – – have reported that persistent post-concussive symptoms are more
Direct effect: combat .01 .02 b1.00 – –
exposure ➔ PHQ-15
strongly related to PTSD and major depression disorder than mTBI [7,
Direct effect: combat .67 .07 10.20⁎ – – 57]. In the current study, we conducted a rigorous analytic examination
exposure ➔ PCL of PTSD and depression symptoms considered simultaneously in a sin-
Direct effect: combat .15 .03 5.44⁎ – – gle model and found evidence for unique indirect effects for each of
exposure ➔ PHQ-9
these factors.
Direct effect: combat .10 .02 5.71⁎ – –
exposure ➔ ISI Insomnia also exerted unique indirect effects on physical symptoms
Direct effect: PCL ➔ PHQ-15 .06 .02 3.00⁎ – – above and beyond shared variance with PTSD and depression. Put oth-
Direct effect: PHQ-9 ➔PHQ-15 .31 .04 6.98⁎ – – erwise, the indirect association of combat exposure with physical symp-
Direct effect: ISI ➔ PHQ-15 .21 .06 3.72⁎ – – toms that was attributable to insomnia was statistically independent of
Indirect effects (a ∗ b) any overlap with associations that were attributable to PTSD and de-
Total (combined) indirect effects .11 .02 – .07 .14 pression symptoms. This is a novel and important finding that extends
PCL .04 .02 – .01 .07 prior studies of post-deployment physical symptoms. It has immediate
PHQ-9 .05 .01 – .03 .07
ISI .02 .01 – .01 .04
implications for how researchers and clinicians assess and treat factors
related to post-deployment physical symptoms. Insomnia, or sleep
Pairwise comparisons for indirect effect estimates disturbance more specifically, should be considered an independent
PCL v PHQ-9 .01 .02 −.03 .04
risk factor for post-deployment physical symptoms instead of merely
PCL v ISI −.01 .02 −.05 .03
PHQ-9 v ISI −.01 .01 −.04 .01 as another component of the global physical symptom milieu [1,3]. In-
deed, studies of the civilian population, and some recent military
Notes. PHQ-15 = Patient Health Questionnaire, 15-item; PCL = PTSD Checklist;
PHQ-9 = Patient Health Questionnaire, 9-item; ISI = Insomnia Severity Index. population-based studies, have established insomnia, short sleep dura-
a
The total model accounted for 35.6% of variance in PHQ-15 scores, which was signif- tion and sleep disturbances as an important independent risk factor for
icant at p b .001 with df = 575. well-being across a wide range of health conditions [33–35]. Moreover,
b
Path coefficients are based on k = 1000 bootstraps. rates of insomnia are increasing at a rapid pace worldwide [35]. Inter-
c
t-Test statistics were computed only for direct effects estimates with α (two-tailed) =
.05. Normal theory significance test statistics were not computed for indirect effects (see
estingly, when cross-lagged analyses have been used, sleep problems
Data analysis for explanation). have emerged as a more robust predictor of later negative health out-
⁎ p b .05. comes than vice-versa [28,58–60].
482 P.J. Quartana et al. / Journal of Psychosomatic Research 78 (2015) 478–483
These findings can help guide future Army and Veterans Affairs pol-
icy concerning the assessment and management of generalized physical
symptom complaints. This type of change in standards of care has
the potential to mitigate some of the common and costly chronic phys-
ical health consequences associated with combat deployment. A num-
ber of randomized clinical trials support the efficacy of cognitive-
behavioral interventions for PTSD [61], depression [62] and insomnia
[63], and a number findings suggest that cognitive behavioral therapy
for insomnia (CBT-I) can lead to improved physical health in partici-
pants with knee osteoarthritis [64], as well as have salutary effects on
co-morbid mental health symptoms [65], providing evidence that im-
proving sleep quality can have measurable effects on physical and men-
tal health. For the Army, “battlemind” resilience training had some mild
protective effects on mental health problems in for soldiers with high
levels of combat exposure [41], suggesting that group-based preventive
approaches can be an effective strategy to mitigate health problems in
combat experienced populations. Future studies are needed to explore
the impact of resilience training on physical health outcomes, and no
work to date has examined the impact of large-scale trials on the
sleep of soldiers. Additionally, the relationship between PTSD, depres-
sion, sleep disturbances and physical symptoms suggests the need for
embedding mental health treatment within primary care settings. This
model of care would allow for coordination of care between primary
and specialty care and address all contributors to post-deployment
physical symptoms.
Some study limitations deserve mention. These data were cross-
sectional and based on self-report. Retrospective reporting of combat
exposure is prone to potential reporting biases, though estimates of
reporting reliability appear to be acceptable [43]. Whether objective in-
dicators of insomnia would reveal effects similar to self-report indica-
tors is not known. The cross-sectional nature of these data prohibits
making any conclusions regarding causal associations between the con-
structs examined. That said, it seems plausible, based on the assessment
of the constructs examined and a priori theoretical grounds, that the de-
gree of combat exposure preceded the onset of PTSD, depression, and
insomnia symptoms, as well as physical symptoms. Future research
that is prospective or experimental in nature is required to determine
the temporal nature of associations between these behavioral and phys-
ical health factors. This study also did not consider the role of other po-
tentially important variables correlated with soldier physical health, as
well as PTSD, depression, and insomnia, including, for example, grief
[1] and physical activity and nutrition [66,67]. It will be important to in-
corporate these and additional factors into this and similar models to
delineate their precise contribution to the association between combat
deployment and physical symptoms.
Finally, this study was not able to assess possible shared neurophys-
iological mechanisms contributing to associations between combat
exposure and mental and physical health problems. One area to investi-
gate in future research is inflammation. Reviews have supported strong
and consistent links between depression and sleep disturbance with
systemic markers of inflammation (e.g., [13,15,16]), and there is also ev-
idence that PTSD is associated with heightened inflammation and alter-
ations in hypothalamic–pituitary–adrenal axis function [9,12]. Chronic
inflammation has been implicated in obesity, cognitive dysfunction, di-
abetes, chronic pain and cardiovascular disease, and more broadly relat-
ed to “sickness behaviors,” which can account for many of the physical
symptoms assessed in this and prior studies.
This study represents the first effort, to our knowledge, to character-
ize the combined and unique contributions to the association of combat
exposure and post-deployment physical symptoms with a set of highly
prevalent mental health concerns among service members previously
deployed to Iraq. Traditionally, it has proven difficult to identify the fac-
tors that adequately account for variability in non-specific physical
symptom expression among war Veterans. These findings highlight a
set of modifiable risk factors that appear to underlie the expression of
physical symptoms that may be harbingers of long-term health risk. It
is imperative that we continue to clarify the precise role of these and
other factors as they relate to the physical health and well-being of ser-
vice members worldwide.
Conflicts of interest and source of funding
The authors have no conflicts of interest to report. The U.S. Army
Medical Research and Materiel Command (USAMRMC) provides intra-
mural funding that supports enhancing the psychological resilience of
the warfighter.
References
[1] Toblin RL, Riviere LA, Thomas JL, Adler AB, Kok BC, Hoge CW. Grief and physical
health outcomes in U.S. soldiers returning from combat. J Affect Disord 2012;136:
469–75.
[2] Killgore WD, Stetz MC, Castro CA, Hoge CW. The effects of prior combat experi-
ence on the expression of somatic and affective symptoms in deploying soldiers.
J Psychosom Res 2006;60:379–85 [Comparative Study].
[3] Hoge CW, Terhakopian A, Castro CA, Messer SC, Engel CC. Association of posttrau-
matic stress disorder with somatic symptoms, health care visits, and absenteeism
among Iraq war veterans. Am J Psychiatry 2007;164:150–3 [Comparative Study
Research Support, U.S. Gov't, Non-P.H.S.].
[4] Vanderploeg RD, Belanger HG, Horner RD, Spehar AM, Powell-Cope G, Luther SL,
et al. Health outcomes associated with military deployment: mild traumatic brain
injury, blast, trauma, and combat associations in the Florida National Guard. Arch
Phys Med Rehabil 2012;93:1887–95.
[5] Hoge CW, McGurk D, Thomas JL, Cox AL, Engel CC, Castro CA. Mild traumatic brain in-
jury in U.S. soldiers returning from Iraq. N Engl J Med 2008;358:453–63 [Comparative
Study Research Support, U.S. Gov't, Non-P.H.S.].
[6] Fear NT, Jones E, Groom M, Greenberg N, Hull L, Hodgetts TJ, et al. Symptoms of post-
concussional syndrome are non-specifically related to mild traumatic brain injury in
UK Armed Forces personnel on return from deployment in Iraq: an analysis of self-
reported data. Psychol Med 2009;39:1379–87.
[7] Wilk JE, Herrell RK, Wynn GH, Riviere LA, Hoge CW. Mild traumatic brain injury
(concussion), posttraumatic stress disorder, and depression in U.S. soldiers involved
in combat deployments: association with postdeployment symptoms. Psychosom
Med 2012;74:249–57.
[8] Jones E, Hodgins-Vermaas R, McCartney H, Everitt B, Beech C, Poynter D, et al. Post-
combat syndromes from the Boer war to the Gulf war: a cluster analysis of their
nature and attribution. BMJ 2002;324:321–4.
[9] Gupta MA. Review of somatic symptoms in post-traumatic stress disorder. Int Rev
Psychiatry 2013;25:86–99.
[10] Xia L, Chen GH, Li ZH, Jiang S, Shen J. Alterations in hypothalamus-pituitary-adrenal/
thyroid axes and gonadotropin-releasing hormone in the patients with primary
insomnia: a clinical research. PLoS One 2013;8:e71065.
[11] Elder GJ, Wetherell MA, Barclay NL, Ellis JG. The cortisol awakening response —
applications and implications for sleep medicine. Sleep Med Rev 2014;18:215–24.
[12] Lindqvist D, Wolkowitz OM, Mellon S, Yehuda R, Flory JD, Henn-Haase C, et al.
Proinflammatory milieu in combat-related PTSD is independent of depression and
early life stress. Brain Behav Immun 2014;42:81–8.
[13] Slavich GM, Irwin MR. From stress to inflammation and major depressive disorder: a
social signal transduction theory of depression. Psychol Bull 2014;140:774–815.
[14] Valkanova V, Ebmeier KP, Allan CL. CRP, IL-6 and depression: a systematic review
and meta-analysis of longitudinal studies. J Affect Disord 2013;150:736–44.
[15] Howren MB, Lamkin DM, Suls J. Associations of depression with C-reactive protein,
IL-1, and IL-6: a meta-analysis. Psychosom Med 2009;71:171–86.
[16] Motivala SJ. Sleep and inflammation: psychoneuroimmunology in the context of
cardiovascular disease. Ann Behav Med 2011;42:141–52.
[17] Shah AJ, Lampert R, Goldberg J, Veledar E, Bremner JD, Vaccarino V. Posttraumatic
stress disorder and impaired autonomic modulation in male twins. Biol Psychiatry
2013;73:1103–10.
[18] Nissen LR, Marott JL, Gyntelberg F, Guldager B. Danish soldiers in Iraq: perceived ex-
posures, psychological distress, and reporting of physical symptoms. Mil Med 2011;
176:1138–43.
[19] Benyamini Y, Solomon Z. Combat stress reactions, posttraumatic stress disorder, cu-
mulative life stress, and physical health among Israeli veterans twenty years after
exposure to combat. Soc Sci Med 2005;61:1267–77.
[20] Kline A, Falca-Dodson M, Sussner B, Ciccone DS, Chandler H, Callahan L, et al. Effects
of repeated deployment to Iraq and Afghanistan on the health of New Jersey Army
National Guard troops: implications for military readiness. Am J Public Health
2010;100:276–83.
[21] Maia A, McIntyre T, Pereira MG, Ribeiro E. War exposure and post-traumatic stress
as predictors of Portuguese colonial war veterans' physical health. Anxiety Stress
Coping 2011;24:309–25.
[22] Thomas JL, Wilk JE, Riviere LA, McGurk D, Castro CA, Hoge CW. Prevalence of mental
health problems and functional impairment among active component and National
Guard soldiers 3 and 12 months following combat in Iraq. Arch Gen Psychiatry 2010;
67:614–23 [Comparative Study Research Support, U.S. Gov't, Non-P.H.S.].
[23] Vasterling JJ, Schumm J, Proctor SP, Gentry E, King DW, King LA. Posttraumatic stress
disorder and health functioning in a non-treatment-seeking sample of Iraq war vet-
erans: a prospective analysis. J Rehabil Res Dev 2008;45:347–58.
 P.J. Quartana et al. / Journal of Psychosomatic Research 78 (2015) 478–483
[24] Luxton DD, Greenburg D, Ryan J, Niven A, Wheeler G, Mysliwiec V. Prevalence and
impact of short sleep duration in redeployed OIF soldiers. Sleep 2011;34:1189–95.
[25] Plumb TR, Peachey JT, Zelman DC. Sleep disturbance is common among
servicemembers and veterans of Operations Enduring Freedom and Iraqi Freedom.
Psychol Serv 2014;11:209–19.
[26] Seelig AD, Jacobson IG, Smith B, Hooper TI, Boyko EJ, Gackstetter GD, et al. Sleep pat-
terns before, during, and after deployment to Iraq and Afghanistan. Sleep 2010;33:
1615–22.
[27] Wilk JE, Thomas JL, McGurk DM, Riviere LA, Castro CA, Hoge CW. Mild traumatic
brain injury (concussion) during combat: lack of association of blast mechanism
with persistent postconcussive symptoms. J Head Trauma Rehabil 2010;25:9–14.
[28] Quartana PJ, Wickwire EM, Klick B, Grace E, Smith MT. Naturalistic changes in in-
somnia symptoms and pain in temporomandibular joint disorder: a cross-lagged
panel analysis. Pain 2010;149:325–31.
[29] Smith MT, Edwards RR, McCann UD, Haythornthwaite JA. The effects of sleep depri-
vation on pain inhibition and spontaneous pain in women. Sleep 2007;30:494–505.
[30] Goodin BR, Smith MT, Quinn NB, King CD, McGuire L. Poor sleep quality and exag-
gerated salivary cortisol reactivity to the cold pressor task predict greater acute
pain severity in a non-clinical sample. Biol Psychol 2012;91:36–41.
[31] Abad VC, Sarinas PS, Guilleminault C. Sleep and rheumatologic disorders. Sleep Med
Rev 2008;12:211–28.
[32] Finan PH, Goodin BR, Smith MT. The association of sleep and pain: an update and a
path forward. J Pain 2013;14:1539–52.
[33] Ferrie JE, Shipley MJ, Cappuccio FP, Brunner E, Miller MA, Kumari M, et al. A prospec-
tive study of change in sleep duration: associations with mortality in the Whitehall II
cohort. Sleep 2007;30:1659–66.
[34] Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated
with sleep duration and insomnia. Arch Gen Psychiatry 2002;59:131–6.
[35] Luyster FS, Strollo Jr PJ, Zee PC, Walsh JK. Sleep: a health imperative. Sleep 2012;35:
727–34.
[36] Harvey AG. Sleep and circadian functioning: critical mechanisms in the mood disor-
ders? Annu Rev Clin Psychol 2011;7:297–319.
[37] Harvey AG, Jones C, Schmidt DA. Sleep and posttraumatic stress disorder: a review.
Clin Psychol Rev 2003;23:377–407.
[38] Germain A. Sleep disturbances as the hallmark of PTSD: where are we now? Am J
Psychiatry 2013;170:372–82.
[39] Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and com-
paring indirect effects in multiple mediator models. Behav Res Methods 2008;40:
879–91.
[40] Bliese PD, Wright KM, Adler AB, Thomas JL, Hoge CW. Timing of postcombat mental
health assessments. Psychol Serv 2007;4:141–8.
[41] Adler AB, Bliese PD, McGurk D, Hoge CW, Castro CA. Battlemind debriefing
and battlemind training as early interventions with soldiers returning from Iraq:
Randomization by platoon. J Consult Clin Psychol 2009;77:928–40.
[42] Sipos ML, Bar-Haim Y, Abend R, Adler AB, Bliese PD. Postdeployment threat-related
attention bias interacts with combat exposure to account for PTSD and anxiety
symptoms in soldiers. Depress Anxiety 2014;31:124–9.
[43] Garvey Wilson AL, Hoge CW, McGurk D, Thomas JL, Castro CA. Stability of combat
exposure recall in Operation Iraqi Freedom veterans. Ann Epidemiol 2010;20:
939–47.
[44] Britt TW, Adler AB, Bliese PD, Moore D. Morale as a moderator of the combat
exposure–PTSD symptom relationship. J Trauma Stress 2013;26:94–101.
[45] Toblin RL, Quartana PJ, Riviere LA, Walper KC, Hoge CW. Chronic pain and opioid use
in US soldiers after combat deployment. JAMA Intern Med 2014;174:1400–1.
[46] Kroenke K, Spitzer RL, Williams JB. The PHQ-15: validity of a new measure for eval-
uating the severity of somatic symptoms. Psychosom Med 2002;64:258–66.
[47] Blanchard EB, Jones-Alexander J, Buckley TC, Forneris CA. Psychometric properties of
the PTSD Checklist (PCL). Behav Res Ther 1996;34:669–73.
483
[48] Bliese PD, Wright KM, Adler AB, Cabrera O, Castro CA, Hoge CW. Validating the pri-
mary care posttraumatic stress disorder screen and the posttraumatic stress disor-
der checklist with soldiers returning from combat. J Consult Clin Psychol 2008;76:
272–81.
[49] Terhakopian A, Sinaii N, Engel CC, Schnurr PP, Hoge CW. Estimating population
prevalence of posttraumatic stress disorder: an example using the PTSD checklist.
J Trauma Stress 2008;21:290–300.
[50] Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity
measure. J Gen Intern Med 2001;16:606–13.
[51] Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version
of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Dis-
orders. Patient Health Questionnaire. JAMA 1999;282:1737–44.
[52] Bastien CH, Vallieres A, Morin CM. Validation of the Insomnia Severity Index as an
outcome measure for insomnia research. Sleep Med 2001;2:297–307.
[53] Morin CM, Belleville G, Belanger L, Ivers H. The Insomnia Severity Index: psychomet-
ric indicators to detect insomnia cases and evaluate treatment response. Sleep 2011;
34:601–8.
[54] Gagnon C, Belanger L, Ivers H, Morin CM. Validation of the Insomnia Severity Index
in primary care. J Am Board Fam Med 2013;26:701–10.
[55] Mackinnon DP, Warsi G, Dwyer JH. A simulation study of mediated effect measures.
Multivar Behav Res 1995;30:41.
[56] Hoge CW, Auchterlonie JL, Milliken CS. Mental health problems, use of mental health
services, and attrition from military service after returning from deployment to Iraq
or Afghanistan. JAMA 2006;295:1023–32 [Research Support, U.S. Gov't, Non-P.H.S.].
[57] Yurgil KA, Barkauskas DA, Vasterling JJ, Nievergelt CM, Larson GE, Schork NJ, et al.
Marine Resiliency Study T. Association between traumatic brain injury and risk of
posttraumatic stress disorder in active-duty Marines. JAMA Psychiatry 2014;71:
149–57.
[58] Tang NK, Goodchild CE, Sanborn AN, Howard J, Salkovskis PM. Deciphering the tem-
poral link between pain and sleep in a heterogeneous chronic pain patient sample: a
multilevel daily process study. Sleep 2012;35:675–687A.
[59] Edwards RR, Almeida DM, Klick B, Haythornthwaite JA, Smith MT. Duration of sleep
contributes to next-day pain report in the general population. Pain 2008;137:202–7.
[60] Wright KM, Britt TW, Bliese PD, Adler AB, Picchioni D, Moore D. Insomnia as predic-
tor versus outcome of PTSD and depression among Iraq combat veterans. J Clin
Psychol 2011;67:1240–58.
[61] Bisson JI, Roberts NP, Andrew M, Cooper R, Lewis C. Psychological therapies for
chronic post-traumatic stress disorder (PTSD) in adults. Cochrane Database Syst
Rev 2013;12:Cd003388.
[62] Cuijpers P, Berking M, Andersson G, Quigley L, Kleiboer A, Dobson KS. A meta-
analysis of cognitive-behavioural therapy for adult depression, alone and in compar-
ison with other treatments. Can J Psychiatry 2013;58:376–85.
[63] Morin CM, Vallieres A, Guay B, Ivers H, Savard J, Merette C, et al. Cognitive behavioral
therapy, singly and combined with medication, for persistent insomnia: a random-
ized controlled trial. JAMA 2009;301:2005–15.
[64] McCurry SM, Shortreed SM, Von Korff M, Balderson BH, Baker LD, Rybarczyk BD,
et al. Who benefits from CBT for insomnia in primary care? Important patient selec-
tion and trial design lessons from longitudinal results of the Lifestyles trial. Sleep
2014;37:299–308.
[65] Taylor DJ, Pruiksma KE. Cognitive and behavioural therapy for insomnia (CBT-I) in
psychiatric populations: a systematic review. Int Rev Psychiatry 2014;26:205–13.
[66] Purvis DL, Lentino CV, Jackson TK, Murphy KJ, Deuster PA. Nutrition as a component
of the performance triad: how healthy eating behaviors contribute to soldier perfor-
mance and military readiness. US Army Med Dep J 2013:66–78.
[67] Smith TJ, Dotson LE, Young AJ, White A, Hadden L, Bathalon GP, et al. Eating patterns
and leisure-time exercise among active duty military personnel: comparison to the
Healthy People objectives. J Acad Nutr Diet 2013;113:907–19.