Hypertension in Relation to Posttraumatic Stress Disorder and Depression in the US National Comorbidity Survey

Hypertension in Relation to Posttraumatic
Stress Disorder and Depression in the
US National Comorbidity Survey
Jeffrey L. Kibler, PhD; Kavita Joshi, MS; Mindy Ma, PhD
The clinical literature increasingly indicates that cardiovascular risk factors
and cardiovascular disease (CVD) are more common among individuals
with posttraumatic stress disorder (PTSD). Depression also poses a risk for
CVD and is often comorbid with PTSD. Research to date has not established
whether PTSD is associated with additional CVD risk beyond the risks
associated with comorbid depression. The authors examined relationships
of lifetime PTSD and depression with high blood pressure in data from the
US National Comorbidity Survey. They divided participants into 4 mutually
exclusive diagnostic groups: (1) PTSD history and no depression history,
(2) PTSD and depression history, (3) depression history and no PTSD history,
and (4) no history of mental disorder. Hypertension prevalence was higher
for the PTSD, no depression and PTSD plus depression groups compared
with the depression only and no mental disorder groups. PTSD appears to be
related to hypertension independent of depression. This may partially explain
elevated rates of CVD in PTSD patients.
Index Terms: cardiovascular, depression hypertension, posttraumatic
stress, PTSD
Several researchers to date have identified a relationship
a nonveteran, same-age control group.2 There is also evi-
between posttraumatic stress disorder (PTSD) and either car-
dence of significantly higher rates of CVD in World War
diovascular disease (CVD) or symptoms of CVD.1–4 This evi-
II and Korean War veterans with PTSD.4 However, these
dence primarily involves retrospective examination of PTSD
studies are limited by designs that relied on retrospective
status among combat veterans who exhibit clinical symptoms
data as well as the fact that the researchers did not account
of CVD. Civilians with PTSD—and women in particular—
for history of depression or depressive symptoms in the
have been understudied with regard to cardiovascular risk.
analyses. High rates of comorbid major depression are con-
Depression is also related to CVD risk, and the literature on the
sistently reported among persons diagnosed with PTSD.5–7
PTSD?CVD risk relationship has not systematically examined
Data from 1,007 participants in the Epidemiologic Study
the extent to which the CVD risk is accounted for by the tre-
of Young Adults indicate that individuals with PTSD are
mendous overlap of depression with PTSD.
approximately 3 times more likely to experience major
In one study, heart disease was nearly 2.5 times as com-
depression than are individuals not exposed to a traumatic
mon in World War II veterans with PTSD compared with
event.8 Epidemiological studies and meta-analytic data
suggest that depressive symptoms may play a causal role
Dr Kibler and Ms Joshi are with the Center for Psychological
in the development of CVD.9–13 Because depression often
Studies at Nova Southeastern University in Fort Lauderdale, FL.
co-occurs with PTSD and is associated with CVD risk, it is
Dr Ma is with the Division of Social and Behavioral Sciences at
Nova Southeastern University.
critical to determine whether PTSD is associated with CVD
Copyright © 2008 Heldref Publications
risk beyond the risk associated with depression.
125

---

HYPERTENSION IN RELATION TO PTSD
In a recent study of military veterans, Kubzansky et al14
US National Comorbidity Survey (NCS)28,29 indicated that
found evidence that the level of posttraumatic stress symp-
childhood sexual abuse was associated with increased risk
toms, controlling for depressive symptoms, is prospectively
of cardiac disease and other health problems.30,31 Although
predictive of CVD. However, conclusions regarding the
researchers in these epidemiological studies did not specifi-
effects of PTSD were limited because the authors did not
cally examine PTSD diagnosis in their samples, their results
assess PTSD diagnosis; instead, they assessed PTSD symp-
provide a broader developmental context for other reports
toms using a self-report survey, which indicated that levels
that indicate that adults with a PTSD diagnosis generally
of posttraumatic stress symptoms were low to moderate (ie,
have poorer health and their health problems occur earlier in
generally not indicative of PTSD). Thus, researchers must
life than in the general population.32–34
systematically assess history of both PTSD and depression
The present study is an examination of the relationship
diagnoses when examining the relative associations of these
of hypertension with lifetime history of PTSD and depres-
disorders to CVD risk.
sion among respondents to the NCS. We examined whether
One risk factor that may underlie the elevated risk for
PTSD is significantly associated with hypertension and
CVD among individuals with PTSD is high blood pressure
whether this association is independent of depression. The
(BP). Combat veterans with PTSD have evidenced exagger-
NCS sample was designed to be representative of the US
ated BP that was measured using both laboratory15–19 and
population; therefore, men and women were adequately
24-hour ambulatory methodologies.20 These findings sug-
represented, as were civilian trauma victims. The NCS used
gest a condition of sympathetic nervous system overdrive
a sampling approach that limited respondent ages to 15–54
in persons with PTSD,21 which may put the system at risk
years. This methodological strategy resulted in the assess-
for cardiovascular injury. McEwen et al22–25 characterized
ment of hypertension diagnosed at a relatively early age.
the phenomenon of chronic sympathetic overactivation in
This artificial restriction of age may contribute to improve
PTSD patients using the concept of allostatic load. In this
understanding of how PTSD and depression are associated
model, allostasis is an individual’s normal, adaptive process
with early onset of cardiovascular risk.
of responding physiologically to acute stress (ie, to main-
tain homeostasis and cope behaviorally) with increases in
METHODS
cortisol levels, neurotransmitter release, and cardiovascular
Participants
reactivity. One byproduct of normal response to acute stress
is enhanced immune function.23 However, chronic overacti-
We analyzed data from the NCS,35 a large-scale epidemi-
vation of the stress-response system (as evidenced in PTSD
ological study conducted in the United States from 1990 to
patients) and corresponding immune suppression, chronic
1992. We obtained approval from the appropriate university
inflammation, and behavioral changes may wear down the
institutional review board for the present analyses.
body and pose risk for illness.22–25 This wear and tear is
The NCS involved administration of a structured psychi-
termed allostatic load.22–25
atric interview to determine the prevalence of psychiatric
One limitation of the studies indicating exaggerated BP
disorders based on criteria outlined in the Diagnostic and
in PTSD patients is that the investigators did not control
Statistical Manual of Mental Disorders, 3rd Edition, Revised
for the potential role of comorbid depression. In addition,
(DSM-III-R).36 This survey was conducted with a nation-
the finding of elevated BP has not yet been established in
ally representative sample of 8,098 participants aged 15–54
civilian PTSD patients. In general, the literature on cardio-
years. NCS researchers administered a second survey that
vascular risk in PTSD patients has excluded civilians, and
assessed psychosocial and health variables in relation to
few studies have included women. Yet many civilians suffer
psychiatric disorders in a subsample of 5,877 participants.
from PTSD resulting from the occurrence of threatening inci-
For the purposes of the present study, we identified 4,008
dents, including sexual abuse or rape, physical assault, and
participants who fit into 1 of 4 mutually exclusive diagnostic
automobile accidents. The available evidence suggests that
groups: individuals with (1) a history of PTSD diagnosis
for civilians who are traumatized at an early age and do not
(lifetime) and no history of major depression (n = 219), (2)
recover, the cumulative impacts of traumatic events may span
a lifetime history of both PTSD and major depression (n
the lifetime. Two epidemiological studies in the United States
= 210), (3) a history of major depression (lifetime) and no
and the United Kingdom have demonstrated linear relation-
PTSD (n = 785), and (4) no history of mental illness (n =
ships between the number of adverse childhood experiences
2794). We excluded respondents with a history of mental ill-
(eg, abuse or other potentially traumatic events) and cardio-
ness but no history of PTSD or major depression (n = 1,869).
vascular risk factors evidenced as adults.26,27 Data from the
Participants had a mean age of 34 years (SD = 10.8 y). The
126
Behavioral Medicine

---

KIBLER ET AL
sample consisted of both men (45%) and women (55%). The
We used chi-square analysis to test for significant differ-
racial breakdown of the sample was 73.6% Caucasian, 13.0%
ences in hypertension among the 4 diagnostic groups. In the
African American, 9.8% Hispanic, and 3.6% other race.
event that the omnibus test of between-group differences was
significant, we conducted additional chi-square analyses to
Sampling and Weighting
identify significant differences between each pair of diagnos-
To recruit a nationally representative sample, NCS inves-
tic groups. We calculated odds ratios (ORs) for hypertension
tigators used stratified, multistage area probability sampling
associated with each of the 3 mental illness groups, relative to
for participant selection. The overall response rate was 82.6%.
the no mental illness group, using logistic regression analy-
The researchers offered a small percentage of nonrespondents
ses. We conducted an additional set of logistic regression
a financial incentive to participate, and rates of psychiatric
analyses to calculate ORs for the 2 PTSD groups relative to
disorders in this population were higher. We weighted the
the depression, no PTSD group. We controlled for sex, age,
data to account for this difference in prevalence. To reflect
and history of regular cigarette use (daily for more than a
differences in selection from within and between households,
month [yes or no]) in the logistic regression by entering these
we also used a second weight. We applied a third weight to
variables in the regression equation. Questions about ciga-
modify the data to the national population distribution.
rette use were administered to only a subsample of the NCS
part II survey participants (4,414 of the 5,877 respondents).
Measures
Therefore, data on smoking history were available for only
3,090 of the 4,008 participants from this analysis. Because
Psychiatric Interviews
the inclusion of smoking as a control variable significantly
NCS researchers interviewed participants in their homes
reduced the sample size, we conducted the logistic regression
after obtaining informed consent. To assess the diagnoses
analyses both with and without smoking included.
of psychiatric disorders, they used a modified version of
the Composite International Diagnostic Interview (CIDI).37
RESULTS
The CIDI is a structured diagnostic interview based on the
Of the study respondents, 7.3% reported a history of
Diagnostic Interview Schedule.38 It is designed for use by
PTSD (4.6% of men, 10.0% of women) and 17.0% reported
trained interviewers who are not clinicians to assess the
a history of depression (11.8% of men, 22.0% of women).
lifetime prevalence of psychiatric disorders. The disorders
Because our study sample was relatively young, the rate
assessed for the NCS were alcohol and drug abuse or depen-
of hypertension was modest (7.8% overall among our 4
dence, major depression, bipolar disorder, dysthymia, panic
diagnostic groups). Among the subsample of participants
disorder, agoraphobia, social phobia, simple phobia, general
for whom we assessed smoking history, 43.9% reported a
anxiety disorder, and PTSD.
history of regular cigarette use.
Field staff from the Survey Research Center at the
Chi-square analysis showed an overall difference in rates
University of Michigan administered the NCS. Interview-
of hypertension among the 4 diagnostic categories, ?2(3, N
ers were 158 individuals who completed a 7-day training
= 4,008) = 34.65, p < .001. Follow-up analyses to contrast
course, had an average of 5 years of experience interviewing
the diagnostic groups revealed that the group with a history
at the Survey Research Center, and were supervised by the
of PTSD and no history of depression had the highest rate
Survey Research Center throughout data collection.
of hypertension (14.5%) and that this rate was significantly
higher than the rate for the no mental illness group (6.5%),
Health Conditions
?2(1, N = 4,008) = 20.21, p < .001, and the group with his-
Participants were given a list of serious health problems
tory of depression and no PTSD (9.7%), ?2(1, N = 4,008) =
and indicated which, if any, they had experienced in the
4.24, p < .05. We observed similar findings when compar-
past 12 months. Fourteen physical health conditions were
ing hypertension rates in the PTSD plus history of depres-
listed, including heart disease or other serious heart trouble
sion group (13.9%) with the rate in the no mental illness
and high BP or hypertension, which were our outcomes of
group, ?2(1, N = 4,008) = 16.36, p < .001, and the group
interest. On the basis of participants’ responses, we coded
with history of depression and no PTSD, ?2(1, N = 4,008)
the health conditions as 0 (no) or 1 (yes).
= 3.07, p = .08. The PTSD, no history of depression group
did not differ significantly from the PTSD plus history of
Data Analyses
depression group. The hypertension rate in the depression
We used descriptive analyses to characterize rates of
no PTSD group was greater than the rate in the no mental
PTSD, depression, and hypertension in the study sample.
illness group, ?2(1, N = 4,008) = 9.44, p < .01.
Vol 34, Winter 2009
127

---

HYPERTENSION IN RELATION TO PTSD
Table 1 shows the occurrence of hypertension by
COMMENT
diagnostic group and the ORs for each mental disorder
group, after controlling for sex and age. We repeated
Our results suggest that a history of PTSD is associated
the regression analysis for participants whose smoking
with greater rates of hypertension compared with depression
history we had assessed; Table 2 presents the results
in the absence of PTSD or no mental illness. These findings
after controlling for sex, age, and smoking history.
are consistent with studies that have identified higher levels
Diagnostic group was significantly related to hyperten-
of BP among combat veterans with PTSD relative to con-
sion prevalence, controlling for a significant positive
trols15–20 and suggest that such findings may generalize to
relationship between age and hypertension rate, Wald(1,
civilian samples. The finding that the PTSD, no depression
4007) = 110.79, p < .001. Sex was not significantly
group had a comparable (and slightly higher) rate of hyper-
related to hypertension: the prevalence rates for men
tension relative to the PTSD plus depression group suggests
(7.8%) and women (8.0%) were comparable. Although
that the relationship of PTSD to high BP is independent of
the rate of hypertension was higher for respondents with
comorbid depressive symptoms. In addition, the finding
a smoking history (9.4%) than for participants with no
that PTSD diagnosis was associated with hypertension after
history of regular cigarette use (7.4%) using a univari-
controlling for age and smoking history provides further
ate analysis, ?2(1, N = 3,090) = 3.93, p < .05, smoking
evidence for the richness of this relationship.
history was not a significant predictor in the logistic
Onset of hypertension in PTSD patients may represent
regression analysis.
an early premorbid risk factor for CVD. Elevations in
TABLE 1. Rates of Hypertension and Odds Ratios, by Diagnostic History, Controlling for Sex and Age (N = 4,008)

Hypertension
Diagnostic group
Pos
Neg
Prevalence (%)
ORa
CI
ORb
CI
PTSD without depression
32
188
14.5
2.56***
1.69–3.88
1.62*
1.03–2.55
PTSD with depression
29
180
13.9
2.59***
1.68–3.99
1.61*
1.01–2.56
Depression without PTSD
76
709
9.7
1.56**
1.17–2.08
—
—
No mental illness
181
2613
6.5
—
—
—
—
Note. CI = 95% confidence interval for odds ratio; Neg = negative case for hypertension; OR = odds ratio; Pos = positive case for hypertension.
aReference group for odds ratio calculation is no mental illness group.
bReference group for odds ratio calculation is depression without history of posttraumatic stress disorder (PTSD).
*p < .05. **p < .01. ***p < .001.
TABLE 2. Rates of Hypertension and Odds Ratios, by Diagnostic History, Controlling for Sex, Age, and
Smoking History (N = 3,090)

Hypertension
Diagnostic group
Pos
Neg
Prevalence (%)
ORa
CI
ORb
CI
PTSD without depression
25
127
16.4
2.94***
1.82–4.74
2.16**
1.28–3.63
PTSD with depression
24
139
14.7
2.81***
1.72–4.60
1.99*
1.17–3.38
Depression without PTSD
55
549
9.1
1.30
0.93–1.82
—
—
No mental illness
153
2018
7.0
—
—
—
—
Note. CI = 95% confidence interval for odds ratio; Neg = negative case for hypertension; OR = odds ration; Pos = positive case for hypertension.
aReference group for odds ratio calculation is no mental illness group.
bReference group for odds ratio calculation is depression without history of posttraumatic stress disorder (PTSD).
*p < .05. **p < .01. ***p < .001.
128
Behavioral Medicine

---

KIBLER ET AL
traditional cardiovascular risk factors, such as hypertension,
est use. However, we did not have enough information on
may partially explain the previously identified relationships
length of periods of use and nonuse or the typical amount of
between PTSD and CVD.1,2,4,14 Our findings suggest that
cigarettes used over the respondents’ history to calculate an
with regard to BP, the CVD risk associated with PTSD
index of pack per year history. Therefore, our use of regu-
is only partially explained by the presence of comorbid
lar smoking for a month or more is the best indication of
depressive symptoms. Despite the evidence that PTSD
smoking history available for the purposes of our analyses.
may be related to the development of CVD, there has been
Most of the respondents who reported a history of regular
little focus on modifying cardiovascular risk factors that
smoking (nearly 70%) reported regular smoking in the year
may underlie the preclinical progression of PTSD toward
preceding assessment. Given the age range of our sample,
CVD. Investigators in some assessment studies have exam-
this suggests that a positive smoking history was reflective
ined cigarette smoking, and the data indicate that Vietnam
of long-term use for many respondents.
combat veterans have higher rates of smoking than do non-
Another limitation relates to the qualitative nature of the
PTSD controls.39–41 Preliminary findings also suggest that
mental health diagnoses. Some individuals with a history
unhealthy lipid profiles may be more common in people
of PTSD and no history of major depression may have had
with PTSD.17,42 Along with previous research on BP in
depressive symptoms that did not meet diagnostic criteria
PTSD, our findings suggest that high BP may be another
for major depression. Likewise, some individuals in the
treatable condition that predisposes individuals with PTSD
depression, no PTSD group may have experienced post-
to develop CVD. Taken together, the findings pertaining to
traumatic stress symptoms that were below the threshold
cardiovascular risk are notable because smoking, in com-
for diagnosis. Thus, the PTSD, no depression and depres-
bination with hypertension and unhealthy lipids, increases
sion, no PTSD groups are not likely to be as pure as these
cardiovascular risk by a factor greater than the sum of their
categorical classifications suggest.
independent risks.43 In addition to our hypothesis that modi-
Despite its limitations, our study provided some cor-
fiable behavioral factors mediate the increased CVD risk in
roborating evidence for the relationship of PTSD to cardio-
PTSD, the relationship between PTSD and CVD may also
vascular risk. We were able to expand on previous research
be attributed to the direct effects of chronic exaggerated
by including comparable numbers of men and women, as
sympathetic responses to stress and the corresponding activ-
well as civilian trauma cases, and by demonstrating that the
ity of the hypothalamic pituitary adrenal axis.21,44
relationship of PTSD to hypertension is not significantly
The primary limitation to the present study is its nonex-
accounted for by a comorbid depression diagnosis. The
perimental design. Although the design was not prospective,
artificially restricted age sampling inherent in the meth-
and elevated BP could feasibly have developed at any time
odology of the NCS may also be viewed as a strength,
in relation to PTSD and depression, several factors suggest
in that it resulted in a study of early onset hypertension.
that PTSD preceded hypertension in our study sample. The
Clinicians and researchers should consider the implications
mean and median ages for each of the 4 diagnostic groups
of these and other findings relating PTSD to health out-
were between 33 and 34 years. Because hypertension is an
comes for early intervention focusing on health promotion.
age-related chronic illness that tends to affect middle-aged
Cognitive?behavioral stress management interventions and
and older adults, the hypertension evidenced in the seg-
relaxation techniques may assist in managing stress-related
ment of our study sample aged 54 and younger was likely
arousal and BP for young adults with PTSD. In addition,
relatively recent in the majority of cases. In contrast, PTSD
efforts to incorporate smoking cessation and dietary inter-
and depression occur across most age ranges, often affect-
ventions may assist in the prevention of the significant CVD
ing children, adolescents, and young adults. In addition,
risk associated with clustering of high BP with smoking and
the average age of PTSD onset in the present study was 17
elevated lipid levels.
years (SD = 9.0 y), which was considerably lower than the
average age of study participants and considerably lower
ACKNOWLEDGMENT
than the typical age of onset for hypertension.
The NCS was funded by the National Institute of Mental
A methodological limitation concerning the assessment
Health (Grants R01 MH/DA46376 and R01 MH49098), the
of smoking history involved the inability to provide a
National Institute of Drug Abuse (through a supplement to
robust measure of extent of cigarette smoking. The NCS
R01 MH/DA46376), and the W. T. Grant Foundation (Grant
included information about respondents’ age of earliest
90135190). The original collector of the data, the Univer-
regular smoking, age of last regular smoking, and greatest
sity of Michigan Interuniversity Consortium for Political
number of cigarettes smokes per day at the time of great-
and Social Research, and the relevant funding agency, bear
Vol 34, Winter 2009
129

---

HYPERTENSION IN RELATION TO PTSD
no responsibility for uses of this collection or for interpreta-
14. Kubzansky LD, Koenen KC, Spiro A, Vokonas PS, Sparrow D.
tions or inferences based on such uses.
Prospective study of posttraumatic stress disorder symptoms
and coronary heart disease in the normative aging sample.
Arch Gen Psychiatry. 2007;64:109–116.
NOTE
15. Blanchard EB. Elevated basal levels of cardiovascular respons-
For comments and further information, address corre-
es in Vietnam veterans with PTSD: a health problem in the
spondence to Dr Jeffrey L. Kibler, Center for Psychological
making. J Anxiety Disord. 1990;4:233–237.
Studies, Nova Southeastern University, 3301 College Ave-
16. Buckley TC, Kaloupek DG. A meta-analytic examination of
nue, Ft. Lauderdale, FL 33314, USA (e-mail: kibler@nova
basal cardiovascular activity in posttraumatic stress disorder.
Psychosom Med. 2001;63:585–594.
.edu).
17. Filakovic P, Barkic J, Kadoic D, et al. Biological parameters of
posttraumatic stress disorder. Psychiatr Danub. 1997; 9:207–211.
REFERENCES
18. McFall ME, Murburg MM, Ko GN, Veith RC. Autonomic
1. Boscarino JA, Chang J. Electrocardiogram abnormalities
responses to stress in Vietnam combat veterans with posttrau-
among men with stress-related psychiatric disorders: impli-
matic stress disorder. Biol Psychiatry. 1990;27:1165–1175.
cations for coronary heart disease and clinical research. Ann
19. McFall ME, Veith RC, Murburg MM. Basal sympathoadrenal
Behav Med. 1999;21:227–234.
function in posttraumatic stress disorder. Biol Psychiatry.
2. Hovens JE, Op den Velde W, Falger PRJ, de Groen JHM, Van
1992;31:1050–1056.
Duijn H, Aarts PGH. Reported physical health in resistance
20. Muraoka MY, Carlson JG, Chemtob CM. Twenty-four hour
veterans from World War II. Psychol Rep. 1998;82:987–996.
ambulatory blood pressure and heart rate monitoring in com-
3. McFarlane AC, Achison M, Rafalowicz E, Papay P. Physical
bat-related posttraumatic stress disorder. J Trauma Stress.
symptoms in post-traumatic stress disorder. J Psychosom Res.
1998;11:473–484.
1994;38:715–726.
21. Murburg MM, McFall ME, Ko GN, Veith RC. Stress-induced
4. Schnurr PP, Spiro A, Paris AH. Physician-diagnosed medical
alterations in plasma catecholamines and sympathetic nervous
disorders in relation to PTSD symptoms in older male military
system function in PTSD. In: Murburg MM, ed. Catechol-
veterans. Health Psychol. 2000;19:91–97.
amine Function in Posttraumatic Stress Disorder: Emerg-
5. Davidson JRT, Hughes D, Blazer DG, George LK. Post-trau-
ing Concepts. Washington, DC: American Psychiatric Press;
matic stress disorder in the community: an epidemiological
1994:189–202.
study. Psychol Med. 1991;21:713–721.
22. Friedman MJ, McEwen BS. Posttraumatic stress disorder,
6. Keane TM, Wolfe J. Comorbidity in post-traumatic stress dis-
allostatic load, and medical illness. In: Schnurr PP, Green BL,
order: an analysis of community and clinical studies. J Appl
eds. Trauma and Health: Physical Health Consequences of
Soc Psychol. 1990;20:1776–1788.
Exposure to Extreme Stress. Washington, DC: American Psy-
7. Kessler RC, Sonnega A, Bromet E, Nelson CB. Posttraumatic
chological Association; 2004:157–188.
stress disorder in the National Comorbidity Survey. Arch Gen
23. McEwen BS. The neurobiology of stress: from serendipity to
Psychiatry. 1995;52:1048–1060.
clinical relevance. Brain Res. 2000;886:172–189.
8. Breslau N, Davis GC, Peterson EL, Schultz LR. A second
24. McEwen BS. The neurobiology and neuroendocrinology
look at comorbidity in victims of trauma: the posttraumatic
of stress: implications for post-traumatic stress disorder
stress disorder?major depression connection. Biol Psychiatry.
from a basic science perspective. Psychiatr Clin North Am.
2000;48:902–909.
2002;25:469–494.
9. Anda R, Williamson D, Jones D, et al. Depressed affect, hope-
25. McEwen BS. Stressed or stressed out: what is the difference?
lessness, and the risk of ischemic heart disease in a cohort of
J Psychiatry Neurosci. 2005;30:315–318.
US adults. Epidemiology. 1993;4:285–294.
26. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of
10. Barefoot JC, Schroll M. Symptoms of depression, acute myo-
childhood abuse and household dysfunction to many of the
cardial infarction, and total mortality in a community sample.
leading causes of death in adults: the Adverse Childhood
Circulation. 1996;93:1976–1980.
Experiences (ACE) Study. Am J Prev Med. 1998;14:245–258.
11. Booth-Kewley S, Friedman HS. Psychological predictors
27. Surtees P, Wainwright N, Day N, Brayne C, Luben R, Khaw K.
of heart disease: a quantitative review. Psychol Bull.
Adverse experience in childhood as a developmental risk fac-
1987;101:343–362.
tor for altered immune status in adulthood. Int J Behav Med.
12. Ford DE, Mead LA, Chang PP, Levine DM, Klag MJ. Depres-
2003;10:251–268.
sion predicts cardiovascular disease in men: the Precursors
28. Kessler RC. The National Comorbidity Survey of the United
Study. Circulation. 1994;90:614.
States. Int Rev Psychiatry. 1994;6:365–376.
13. Pratt LA, Ford DE, Crum RM, Armenian HK, Gallo JJ, Eaton
29. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-
WW. Depression, psychotropic medication, and risk of myo-
month prevalence of DSM-III-R psychiatric disorders in the
cardial infarction: prospective data from the Baltimore ECA
United States: results from the National Comorbidity Survey.
follow-up study. Circulation. 1996;94:3123–3129.
Arch Gen Psychiatry. 1994;51:8–19.
130
Behavioral Medicine

---

KIBLER ET AL
30. Goodwin RD, Stein MB. Association between childhood trau-
37. World Health Organization. Composite International Diag-
ma and physical disorders among adults in the United States.
nostic Interview (CIDI; Version1.0) [data file]. Geneva, Swit-
Psychol Med. 2004;34:509–520.
zerland: World Health Organization; 1990.
31. Sachs-Ericsson N, Blazer D, Plant EA, Arnow B. Childhood
38. Robins LN, Helzer JE, Cottler L. The Diagnostic Interview
sexual and physical abuse and the one-year prevalence of
Schedule, Version III-R. St. Louis, MO: Washington Univer-
medical problems in the National Comorbidity Study. Health
sity; 1989.
Psychol. 2005;24:32–40.
39. Beckham JC, Gehrman PR, McClemon FJ, Collie CF, Feldman
32. Beckham JC, Moore SD, Feldman ME, Hertzberg MA, Kirby
ME. Cigarette smoking, ambulatory cardiovascular monitoring
AC, Fairbank JA. Health status, somatization, and sever-
and mood in Vietnam veterans with and without chronic post-
ity of posttraumatic stress disorder in Vietnam combat vet-
traumatic stress disorder. Addict Behav. 2004;29:1579–1593.
erans with posttraumatic stress disorder. Am J Psychiatry.
40. Beckham JC, Kirby AC, Feldman ME, et al. Prevalence and
1998;155:1565–1569.
correlates of heavy smoking in Vietnam veterans with chronic
33. Butterfield MI, Forneris CA, Feldman ME, Beckham JC.
posttraumatic stress disorder. Addict Behav. 1997;22:637–647.
Hostility and functional health status in women veterans with
41. Beckham JC, Roodman AA, Shipley RH, et al. Smoking in
and without posttraumatic stress disorder. J Trauma Stress.
Vietnam veterans with posttraumatic stress disorder. J Trauma
2000;13:735–741.
Stress. 1995;8:461–472.
34. Schnurr PP, Janikowski MK. Physical health and post-trau-
42. Kagan BL, Leskin G, Haas B, Wilkins J, Foy D. Elevated lipid
matic stress disorder: review and synthesis. Semin Clin Neuro-
levels in Vietnam veterans with chronic posttraumatic stress
psychiatry. 1999;4:295–304.
disorder. Biol Psychiatry. 1999;45:374–377.
35. Kessler RC. National Comorbidity Survey, 1990–1992 [com-
43. Ockene IS, Houston-Miller N, for the American Heart Associa-
puter file]. Conducted by University of Michigan, Survey
tion Task Force on Risk Reduction. Cigarette smoking, cardio-
Research Center. 2nd ICPSR ed. Ann Arbor, MI: Inter-univer-
vascular disease, and stroke. Circulation. 1997;96:3243–3247.
sity Consortium for Political and Social Research [producer
44. Southwick SM, Yehuda R, Charney DS. Neurobiological alter-
and distributor]; 2002.
ations in PTSD: review of the clinical literature. In: Fullerton
36. American Psychiatric Association. DSM-III-R: Diag-
CS, Ursano RJ, eds. Posttraumatic Stress Disorder: Acute and
nostic and Statistical Manual of Mental Disorders. 3rd ed.,
Long-term Responses to Trauma and Disaster. Progress in
revised. Washington, DC: American Psychiatric Associa-
Psychiatry Series, No. 51. Washington, DC: American Psychi-
tion; 1987.
atric Press; 1997:241–266.
Vol 34, Winter 2009
131

---

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

---