August 2010  << Back  

  Can't open the PDF? Click here for help.



Biological correlates of direct exposure to terrorism several years postdisaster

Phebe Tucker, MD

Department of Psychiatry and Behavioral Sciences, University of Oklahoma College of Medicine, Oklahoma City, OK, USA

Betty Pfefferbaum, MD, JD

Department of Psychiatry and Behavioral Sciences, University of Oklahoma College of Medicine, Oklahoma City, OK, USA

Carol S. North, MD, MPE

VA North Texas Health Care System, Departments of Psychiatry and Surgery, Division of Emergency Medicine, University of Texas Southwestern Medical Center, North Texas VA Health Care System, Dallas, TX, USA

Adrian Kent, JD, PhD

Department of Psychiatry, University of Oklahoma College of Medicine, Oklahoma City, OK, USA

Haekyung Jeon-Slaughter, PhD

Department of Psychiatry, University of Oklahoma College of Medicine, Oklahoma City, OK, USA

Don E. Parker, PhD

General Clinical Research Center, University of Oklahoma College of Medicine, Oklahoma City, OK, USA

BACKGROUND: It is important to understand long-term biological and psychiatric correlates of intense exposure to terrorism.

METHODS: We assessed psychiatric diagnoses and biological stress measures in 50 healthy, highly exposed Oklahoma City bombing survivors recruited from a bombing registry 6½ to 7 years postdisaster, comparing them with demographically matched, nonexposed community members. The Diagnostic Interview Schedule (DIS) determined Axis I psychiatric diagnoses. Participants’ salivary cortisol levels were obtained at 8 am, and physiologic assessment measured participants’ heart rate and blood pressure responses to a bombing-related interview.

RESULTS: Eleven survivors with posttraumatic stress disorder (PTSD) had significantly higher cortisol levels than did both non-PTSD survivors and controls. Survivors with and without PTSD did not differ in any autonomic reactivity measure, whereas the total survivor group had significantly higher reactivity than controls in all measures. Positive correlations occurred between several autonomic reactivity measures, but none between cortisol and autonomic measures.

CONCLUSIONS: Results differentiate the autonomic and cortisol systems relative to terrorism exposure. Findings support research associating PTSD with hypothalamic-pituitary-adrenal (HPA) axis changes, whereas autonomic reactivity appeared to be a more generalized trauma response. Correlation statistics substantiated a lack of connection between the 2 biological systems. Follow-up could elucidate the long-term course of these stress systems and eventual health status in survivors, in view of the medical morbidity noted in PTSD studies.

KEYWORDS: cortisol, Oklahoma City bombing, psychophysiologic assessment, PTSD, terrorism, trauma



Exposure to terrorism can have numerous psychological sequelae, including work impairment1 and major depression, posttraumatic stress disorder (PTSD), and other anxiety disorders.2 Survivors of terrorist acts worldwide have an estimated 28% prevalence of PTSD at varying times postdisaster,3 with 34% of Oklahoma City bombing survivors having PTSD within the first 6 months,2 and almost 40% of direct survivors of a terrorist bombing in Nairobi developing PTSD 8 to 10 months later.4 Terrorism thus compares with other trauma types that lead to high rates of PTSD diagnosis and symptoms,5 with higher risks noted among those survivors who had early PTSD symptoms and elevated heart rates,6 who sustained injuries,2,7-10 those who were female,2 and those who had prior psychiatric disorders.2

Biological abnormalities demonstrated in PTSD include differences in cortisol levels and other hypothalamic-pituitary-adrenal (HPA) axis measures.11 Investigations of cortisol levels in PTSD—which vary in length of time elapsed between index trauma and cortisol assessments from several months to more than 50 years—have differed in their findings. Several have found that individuals with PTSD have lower cortisol levels11-15 and hypersuppression of cortisol15 compared with controls. However, some studies have revealed higher cortisol levels in individuals with PTSD,16,17 and others have shown no differences in cortisol levels between controls and participants with PTSD.18-20 Studies have assessed salivary, serum, and urinary cortisol levels at various times during the day in participants with PTSD arising from both combat and noncombat trauma, such as childhood sexual abuse, motor vehicle accidents, cancer, and mixed trauma types.

Another biological abnormality seen in PTSD and after trauma is physiologic reactivity to trauma reminders, with increases noted in heart rate, blood pressure, and other autonomic measures.18,21-24

We identified only 2 studies that explored relationships between autonomic reactivity and cortisol in PTSD, with both failing to find connections between the 2 systems.18,25 One study showed that patients with comorbid PTSD and depression had a reduction in autonomic reactivity but no change in cortisol secretion after pharmacotherapy.18 The second found no correlation among physiologic reactivity, cortisol, and catecholamine systems.25 No studies were identified that compared these systems in terrorism survivors.

Studies of biological differences after trauma generally focus on individuals with PTSD. Because many individuals exposed to trauma do not develop PTSD, additional studies are warranted that explore relationships between biological stress measures in traumatized individuals who do not develop PTSD, as well as those who do.

Long-term health implications of such biological changes remain speculative. Studies have shown that PTSD is associated with increased health care utilization,26 somatic and physical symptoms,27 medical illness,28 and all-cause mortality.29 Specifically, cardiovascular morbidity has been noted among veterans with PTSD,29-31 even after controlling for depressive symptoms.32 A plethora of factors may contribute to medical morbidity and mortality in PTSD, including alterations in HPA axis, neuroimmune functioning, and autonomic reactivity,33 as well as preexisting behaviors, psychological characteristics, other medical and psychiatric conditions, genetic risks, and lifestyle behaviors (eg, use of drugs, alcohol, or tobacco).

Studies of biological abnormalities in survivors of terrorism in particular are sparse. Such studies are difficult to conduct due to challenges in accessing sufficient numbers of individuals exposed to the same or similar terrorist acts within similar time periods postdisaster. An Israeli study of cortisol assessments of survivors of terrorism and other trauma found that those with and without PTSD did not differ in plasma or salivary cortisol levels,34 whereas treatment-seeking World Trade Center (WTC) survivors had lower urinary cortisol levels.35 In a study of WTC survivors who were pregnant at the time of the terrorist attack, mothers with PTSD and their babies had lower salivary cortisol levels36 than did non-PTSD WTC survivors and their babies.36 Assessments of heart rate variability of cardiac patients hospitalized in the New York City area during the week after September 11, 2001 showed decreased parasympathetic tone.37 Victims of sarin gas attacks in the Tokyo subway had increased prefrontal cortex activity, whereas only those with PTSD had elevated skin conductance during exposure to trauma stimuli.38

Oklahoma City’s 1995 terrorist bombing offered a unique opportunity to study long-term biological correlates of terrorism, since many staying within the community have been willing to participate in research, and because a random sample of directly exposed survivors was systematically assessed for psychiatric disorders approximately 6 months and 1 year after the bombing.2,39

An earlier publication24 described greater physiologic reactivity among 60 Oklahoma City bombing survivors than community comparison participants 6½ to 7 years postdisaster, regardless of diagnosis of PTSD. The current study addresses a gap in the literature by exploring patterns of salivary cortisol secretion, physiologic reactivity, and psychiatric assessments for a healthy subset of 50 of these terrorism survivors. Questions our study sought to answer were:

1. Will survivors with PTSD differ from survivors without PTSD in am cortisol levels and autonomic reactivity to trauma reminders?

2. Will survivors differ from unexposed comparison participants in these 2 biological measures?

3. Will cortisol and physiologic reactivity measures correlate with each other?



Of 182 survivors of the Oklahoma City bombing who participated in a study by North and colleagues2 conducted approximately 6 months postdisaster, 113 participated in a follow-up study at 6½ to 7 years postbombing. Potential study participants were enlisted from the follow-up sample and contacted by letter and/or telephone call, with assessments starting in late October 2001. Comparison participants, matched for gender and age, were also recruited, initially by asking survivors to identify an unexposed peer from the community, and later through mechanisms such as flyers or word of mouth.

Of the 113 survivors in the follow-up study, 71 participated in the study’s physiologic and cortisol assessments. Of the 42 survivors who did not participate in the physiologic and cortisol assessments, 27 did not participate for unknown reasons and 15 were excluded due to medical illness or use of prohibited medications. Of the total of 71 directly exposed survivors who underwent both biological assessments, complete cortisol data were available for only 51 survivors due to an insufficient quantity of am saliva or exclusion of data by the laboratory. Fifty-one comparison participants were recruited for these selected 51 survivors. Later, we deleted one pair of survivor and comparison participants because they were not matched by gender, leaving a final sample size for this study of 50 survivors and 50 comparison participants.


Written informed consent was obtained for all participants in accordance with the Declaration of Helsinki and requirements of the Institutional Review Boards (IRBs) of both the University of Oklahoma Health Sciences Center and Washington University School of Medicine, the academic institution of one investigator at the time. All participants were paid $150 for their participation in the study.

Salivary cortisol measures

To prevent the possibility that administering psychometric batteries and physiologic assessments first might induce biological responses that could affect cortisol measures, salivary am cortisol samples were collected first, immediately after informed consent was obtained, followed by physiologic and then psychometric assessments.

Prior studies have indicated that salivary and plasma cortisol levels are highly correlated,40 with salivary levels lower than serum levels. Samples for cortisol levels were obtained in Salivette tubes (Sarstedt, Nümbrecht, Germany) from all participants at 8 am. Participants were instructed to rinse their mouths with water, not to eat or drink 15 minutes before samples were collected, and to abstain from caffeine and nicotine before the visit. No participants spontaneously reported trauma reminders the day of cortisol collection. Samples were centrifuged for 2 minutes and stored at –20°C, with salivary cortisol levels later measured by radioimmunoassay methods (Salivary Cortisol Kit; Salimetrics, LLC, State College, PA).

Physiologic assessment

Physiologic assessments were performed next, measuring autonomic reactivity to a trauma interview, following procedures described previously.24 A Lablink Modular Instrument System (Biopac Systems Inc., 1999) recorded heart rate via standard ECG lead II placement. A Critikon Dinamap Vital Signs Monitor (1846 SX) (GE Healthcare, Milwaukee, WI) recorded systolic, diastolic, and mean arterial blood pressure every 30 seconds, with a Compaq computer (Hewlett-Packard, Palo Alto, CA) interfacing with this equipment. Methods to elicit traumatic responses were adapted from those of Pitman’s group.21 Blood pressure and heart rate reactivity were measured in 3 phases of 4 minutes each—at rest (pretest phase), during a semistructured interview about the bombing (test phase), and at rest after the interview was completed (posttest phase). Our interviewer asked each participant to describe how he or she felt after learning about the bombing, if he or she knew anyone injured or killed, and if there was anything that jumped into his or her mind when thinking about the bombing.

Psychometric instruments

All participants were administered the Diagnostic Interview Schedule (DIS), DSM-IV version,41 by a trained interviewer to assess presence of lifetime, predisaster, postdisaster, and current Axis I conditions (bombing-related PTSD, major depression, generalized anxiety disorder [GAD], panic disorder, and alcohol disorder). The DIS Disaster Supplement Questionnaire (DSQ),42 adapted from the DIS Disaster Supplement,43 obtained information about demographics, disaster exposure, level of functioning, and physical and mental health status. The Disaster Supplement has been used in studies of more than 3000 survivors of 15 disasters, and the DSQ has been used in studies of hundreds of survivors.

Data analysis

Chi-square tests compared all categorical demographic variables and the presence of any predisaster or postdisaster Axis I disorders at index to determine whether the 50 participants differed from the 132 nonparticipants from the index sample of 182 on diagnoses or dichotomous demographic variables. Independent sample t tests compared group differences in age and years of education; α was set at .05.

As data for am cortisol levels showed significant departures from normality, Wilcoxon rank sum tests were used to compare survivor and comparison groups and survivors with and without PTSD.

For each physiologic variable—heart rate and systolic, diastolic, and mean arterial blood pressure—means and standard deviations were calculated for both groups for the pretest, test, and posttest phases. Autonomic reactivity scores, calculated by subtracting pretest from test scores for physiologic measures, were obtained for the 50 direct survivors with complete cortisol and physiologic data and for their 50 matched community comparison participants. Wilcoxon rank sum tests were used to analyze all autonomic reactivity measures.

Spearman correlations were used to examine the association between means of am cortisol levels and autonomic reactivity scores (test minus pretest values for autonomic variables).


TABLE 1 shows demographic variables for survivors and comparison participants, which are quite similar across groups. Among the survivors, 36 (84%) reported physical injury from the bombing, 5 (12%) reported no injuries, and no injury data were available for 2 (4%).

Current (within the preceding 30 days) comorbid Axis I diagnoses (at 6½ to 7 years postdisaster) were examined for survivors and controls, and are listed in TABLE 1. Of 11 bombing survivors who had current bombing-related PTSD, 4 had one or more current comorbid Axis I diagnoses. All 11 also had bombing-related PTSD at 6 months postbombing, and 5 of the 11 also had prior PTSD from earlier life traumas.

Survivor participants (n = 50) did not differ from survivor nonparticipants (n = 132) in gender, age, race (Caucasian vs non-Caucasian), employment status, marital status (married vs nonmarried), years of education, or presence of any prebombing or postdisaster Axis I disorders.

In comparing the 39 survivors without current PTSD and the 11 with current PTSD at nearly 7 years after the bombing (FIGURE 1), Wilcoxon rank sum tests showed that those with PTSD had significantly higher am cortisol levels (PTSD mean = 0.34, SD = 0.12; non-PTSD mean = 0.23, SD = 0.18; P = .02). In addition, survivors with PTSD had significantly higher am cortisol levels than controls (control mean = 0.24, SD = 0.16; P = .01) (FIGURE 1). Finally, survivors without PTSD did not differ in cortisol levels from comparison participants (P = .66).

To explore possible effects of smoking status44 on cortisol levels, post hoc chi-square analyses compared 3 survivor smokers and 45 survivor nonsmokers, 9 control smokers and 38 control nonsmokers, and 12 total smokers and 83 total nonsmokers. Smoking status was unknown for 2 survivors and 3 controls. No group differences were found in cortisol levels according to smoking status.

Wilcoxon rank sum tests found no statistically significant differences in any autonomic reactivity measure between survivors with PTSD and those without PTSD (heart rate: P = .19; systolic blood pressure: P = .22; diastolic blood pressure: P = .17; mean arterial pressure: P = .34). As seen in TABLE 2 and FIGURE 2, comparisons of the total survivor and comparison groups showed that survivors had significantly greater autonomic reactivity in all measures.

Pearson correlations showed no positive correlations between survivors’ cortisol levels and any physiologic reactivity measure or any baseline (pretest) autonomic measure (TABLE 3). Several correlations occurred between various physiologic measures, including pretest heart rate and heart rate reactivity, and heart rate reactivity and systolic blood pressure reactivity, as shown in TABLE 3.


Demographics and current AXIS I diagnoses

  Survivors (n = 50) Controls (n = 50)
Age, y, mean (SD) 47 (10) 46 (10)
Education, y, mean (SD) 15 (3) 14 (2)
Gender, n (%)    
  Males 23 (46%) 23 (46%)
  Females 27 (54%) 27 (54%)
Race, n (%)    
  Caucasian 43 (86%) 42 (84%)
  African American 6 (14%) 5 (10%)
  Native American 0 1 (2%)
  Other 1 (2%) 2 (4%)
Current Axis I diagnoses (DIS), n (%)    
PTSD (4 with comorbid disorders, below) 11 (22%) 1 (2%)
  Panic disorder 1 (2%) 0
  GAD 1 (2%) 0
  Major depression + panic disorder 1 (2%) 0
  Major depression + GAD 1 (2%) 0
Alcohol abuse/dependence 1 (2%) 0
Panic disorder 1 (2%) 0
GAD 0 0
Major depression 0 0
DIS: Diagnostic Interview Schedule; GAD: generalized anxiety disorder; PTSD: posttraumatic stress disorder.

FIGURE 1 AM cortisol levels by diagnostic group

PTSD: posttraumatic stress disorder.

*Significantly different from both non-PTSD survivors and controls within 5% significance level.

Means, μg/dL

Survivors with current bomb-related PTSD had significantly higher am cortisol levels than both non-PTSD survivors (P = .02) and controls (P = .01).


Survivor and control group differences in cortisol and physiologic reactivity

  Survivors (n = 50) Controls (n = 50)  
  Mean (SD) Mean (SD) P valuea
AM cortisol 0.26 (0.17) 0.24 (0.16) .54
  Pretest 72.13 (11.95) 68.24 (10.35) .07
  Test 77.83 (12.03) 71.22 (10.82) .0025
  Posttest 72.52 (11.02) 68.05 (14.89) .07
  Difference pretest and test 5.70 (6.08) 2.98 (3.37) .02
  Pretest 120.08 (14.98) 118.89 (14.38) .68
  Test 135.15 (18.46) 123.57 (17.79) .0015
  Posttest 125.32 (17.68) 115.90 (22.14) .08
  Difference pretest and test 15.07 (9.57) 4.68 (7.88) <.0001
  Pretest 74.78 (9.75) 72.30 (10.08) .25
  Test 83.25 (10.73) 76.83 (9.99) .0035
  Posttest 79.43 (16.52) 70.63 (13.03) .0055
  Difference pretest and test 8.47 (5.19) 4.53 (4.70) <.0001
  Pretest 92.07 (10.79) 89.10 (10.99) .16
  Test 103.81 (12.83) 94.75 (12.94) .0004
  Posttest 95.61 (12.29) 87.00 (15.81) .0041
  Difference pretest and test 11.74 (6.76) 5.65 (5.67) <.0001
DBP: diastolic blood pressure; HR: heart rate; MAP: mean arterial pressure; SBP: systolic blood pressure.
aP value from Wilcoxon rank sum test.

FIGURE 2 Physiologic measures by test phase

DBP: diastolic blood pressure; HR: heart rate; MAP: mean arterial pressure; PTSD: posttraumatic stress disorder; SBP: systolic blood pressure.


Correlation matrix

  Pretest HR HR reactivity Pretest SBP SBP reactivity Pretest DBP DBP reactivity Pretest MAP MAP reactivity
Cortisol AM 0.25 0.11 –0.09 –0.27 –0.12 –0.28 –0.08 –0.25
Pretest HR 1.00 –0.29a 0.26 –0.24 0.25 –0.18 0.29a –0.17
HR reactivity   1.00 –0.18 0.29a –0.08 0.25 –0.07 0.24
Pretest SBP     1.00 0.09 0.72a –0.06 0.86a –0.02
SBP reactivity       1.00 0.31a 0.63a 0.28 0.72a
Pretest DBP         1.00 0.07 0.92a 0.08
DBP reactivity           1.00 –0.01 0.88a
Pretest MAP             1.00 0.02
MAP reactivity               1.00
DBP: diastolic blood pressure; HR: heart rate; MAP: mean arterial pressure; SBP: systolic blood pressure.
aP < .05.


Considered as a group, directly exposed, physically healthy survivors of the Oklahoma City bombing were distinguished from nonexposed community comparison participants in reporting more current Axis I conditions. Current bombing-related PTSD was diagnosed in 11 survivors (22%), of whom 4 had one or more comorbid Axis I disorders; all 11 had bombing-related PTSD, and 5 had prior PTSD when assessed 6 months after the bombing.

Biological measures of morning salivary cortisol levels and physiologic reactivity showed distinct patterns in comparing survivors with and without PTSD and in comparing survivors with unexposed controls. Survivors with bombing-related PTSD differed both from survivors without PTSD and nonexposed comparison participants in having significantly higher morning cortisol levels. At the same time, survivors without PTSD did not differ from comparison participants in am cortisol levels. Thus, direct exposure to terrorism without developing PTSD was not associated with altered HPA axis functioning (cortisol), although experiencing current PTSD from this trauma exposure was associated with HPA axis differences. Our higher morning cortisol results for survivors with PTSD are consistent with studies by Lemieux and Liberzon’s groups,16,17 but differed from other studies showing lower cortisol levels in individuals with PTSD compared with controls.11-15

There are several possible explanations for our different cortisol findings and for inconsistent findings in general in the trauma literature. Although we took care to sample cortisol levels before conducting psychometric and physiologic assessments to prevent their affecting HPA axis functioning, we described to participants the planned procedures before obtaining their informed consent, to be consistent with IRB requirements. Thus, it is possible that anticipation of psychometric and physiologic assessments (and of the trauma memories these assessments might awaken) may have increased survivors’ stress levels and affected cortisol measures. This phenomenon has been described in adult survivors of childhood sexual abuse with PTSD, who had increased salivary cortisol levels before and during exposure to trauma reminders, with levels dropping to those of a non-PTSD group during recovery.45 The authors attributed this to heightened HPA axis activity from participants’ reported anticipatory anxiety about the trauma scripts, which they had been informed about before the procedure. In our study as well, if anticipation of accessing traumatic memories affected am cortisol secretion, it did so selectively for those with PTSD, as other non-PTSD survivors did not differ from comparison participants. We did not perform repeated cortisol sampling to determine if levels stabilized later, possibly decreasing to levels below controls, as found in some other studies.11-15

The disparity in cortisol levels in various studies of PTSD populations may also arise from differences in timing of samples during the day, as cortisol is secreted in a diurnal pattern that may be affected by changes in sleep patterns. Other factors potentially complicating cortisol assessments are participants’ intervening stressors or trauma reminders, concomitant medications or medical conditions, and comorbid mental conditions such as depression. Studies have also differed, as mentioned previously, in types of trauma, length of time since index trauma, age and gender of participants, and methods of measuring cortisol. Or, as Liberzon’s group has suggested, individuals may differ in how neurobiological systems are affected in trauma and PTSD.17

Considering another neurobiological measure, physiologic assessment showed a different pattern. Survivors with PTSD and survivors without PTSD did not differ in any autonomic reactivity measure, whereas the total survivor group had greater autonomic reactivity than the comparison group in all measures. These findings differ from previous studies of PTSD populations that have demonstrated greater autonomic reactivity in individuals with PTSD vs comparison participants who were exposed21,22 and controls who were not exposed to the same traumatic stressor.18,23-24 Thus, as in our previous assessment,24 and in contrast to many previous studies, our investigation found that the experience of direct exposure to terrorism—not necessarily the development of PTSD—was sufficient for survivors to demonstrate a robust physiologic response several years later. The fact that this autonomic response in terrorism survivors, regardless of PTSD status, lasted for almost 7 years speaks to its enduring nature.

As noted previously, only 2 other studies18,25 were identified that examined relationships between autonomic reactivity and the HPA axis among traumatized individuals. Both found no associations between the 2 systems in PTSD populations. In our study also, no significant correlations existed between cortisol level and any pretest or autonomic reactivity measure. Overall, our results showed no evidence for reciprocal interactions between the different biological measures of sympathetic nervous system reactivity and HPA axis functioning (cortisol). More refined, in-depth measures of HPA axis functioning could perhaps further elucidate this area, either through low-dose dexamethasone suppression tests or measures of corticotropin-releasing factor (CRF), considered important in regulating stress’s autonomic, immune, and behavioral effects.46

Our findings further differentiate the autonomic and cortisol stress systems relevant to trauma in the pathophysiology of PTSD; the former system seems to be sensitive to exposure to trauma, while the latter system seems to be associated with development of PTSD from that trauma.

An important, yet unanswered, question is whether robust physiologic reactivity to trauma reminders in terrorism survivors and HPA axis changes in survivors with PTSD may be relevant to their future general health. Our survivor group was selected to exclude those with major medical illnesses and medication use to avoid confounding interpretation of biological assessments; as such, participants were physically healthy, and the majority (74%) also lacked current Axis I disorders. It is possible that continued overreactivity of the sympathetic nervous system in healthy terrorism survivors over an even longer period of time (>7 years posttrauma) may contribute to cardiovascular morbidity; this may be a causal factor for survivors with PTSD, a condition noted for both physiologic hyperarousal and for cardiac vulnerability,29-32 and could occur also in non-PTSD terrorism survivors.

One might speculate that dysregulation of the HPA axis in chronic PTSD may also affect immune functioning and general health; PTSD has been associated with increased medical illness,28 physical symptoms,27 and all-cause mortality,29 as described. Research has shown that the HPA axis, particularly cortisol, is important in stress and has a reciprocal effect on the immune system.47,48 This important link is reinforced by a recent 15-year study of combat veterans that found that all-cause disease mortality was associated with lifetime or current PTSD, high inflammatory marker levels, and high cortisol/DHEA (dehydroepiandrosterone sulfate) ratios at follow-up.49 Following our Oklahoma City terrorism survivors over a longer time period might elucidate longer-term relationships between future health status and exposure to terrorism among aging survivors with and without PTSD.

Alternatively, we have previously suggested that long-term physiologic reactivity to trauma reminders in healthy survivors may help prepare them to take action in future catastrophes.24 It is also possible that increased cortisol secretion in anticipation of trauma reminders in PTSD might be a pathologic amplification of a system that enhances the body’s stress response lasting long beyond its need.

Among the limitations of our study, it was not a random sample of survivor and comparison participants, and sample size was limited due to missing data and sampling restrictions imposed by biological assessments. However, the 50 participants did not differ from the 132 nonparticipants from the index study in demographic variables or the presence of any Axis I diagnoses, including PTSD, before or after the bombing.

Another limitation is that we did not explore either previous psychiatric treatment or family psychiatric histories, which might portend a genetic vulnerability to stress. Although we excluded survivors who suffered minor head injuries that could have affected their memories of trauma, survivors’ recall may have been affected by the passage of time. Finally, as our study began 1 month after the events of September 11, 2001, it is not possible to exclude potential effects of this disaster on our participants, possibly exaggerating survivors’ biological responses.


Our study is unique in documenting both that terrorism exposure in individuals—regardless of PTSD status—was associated with robust autonomic reactivity, and that PTSD from terrorism was associated with changes in cortisol secretion among a large group of physically healthy survivors of the same terrorist event when assessed several years later. These findings offer some evidence differentiating biological correlates of PTSD and trauma exposure, perhaps further elucidating the pathophysiology of PTSD and nonpathologic responses. The issue of autonomic reactivity in survivors without PTSD warrants further exploration to determine the long-term effects of trauma on this system among apparently mentally healthy survivors. Follow-up of these survivors over an even longer time period could determine whether these neurobiological changes are sustained and whether they are eventually associated with development of medical morbidity or onset of psychiatric morbidity as survivors age.

ACKNOWLEDGEMENTS: Supported under Award Number MIPT106-113-2000-020 from the National Memorial Institute for the Prevention of Terrorism (MIPT) and the Office for State and Local Government Coordination and Preparedness, US Department of Homeland Security. Funding for this study supported design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript. Points of view in this document are those of the authors and do not necessarily represent the official position of MIPT or the US Department of Homeland Security.

Preparation of this manuscript has been supported in part by the National Institutes of Health, National Center for Research Resources, General Clinical Research Center Grant M01 RR-14467, and by the National Institute of Mental Health Grant MH40025 supplement. This support included management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.

Dr. North acknowledges support from the VA and NIMH grants R01 040025 and R01 068853. The contents of this article represent the views of the authors and do not represent the views of the Department of Veterans Affairs or the United States Government.

The principal investigator takes responsibility for the integrity of the data and accuracy of data analysis, and all authors had full access to all the data in the study.

The authors wish to acknowledge Brian Maynard, PhD, Akm Hossain, MD, and Dorothy B. Wyatt, RN, for their assistance with this manuscript and study.

DISCLOSURES: Within the past 5 years, only the principal author, Dr. Tucker has conducted separate clinical drug trials not related to the content of this study. These trials were funded by AstraZeneca, Bristol-Myers Squibb, Otsuka Pharmaceuticals, GlaxoSmithKline, and Cephalon, Inc. Drs. Pfefferbaum, North, Kent, Jeon-Slaughter, and Parker report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.


  1. Stein BD, Elliott MN, Jaycox LH, et al. A national longitudinal study of the psychological consequences of the September 11, 2001 terrorist attacks: reactions, impairment, and help-seeking. Psychiatry. 2004;67:105–117.
  2. North CS, Nixon SJ, Shariat S, et al. Psychiatric disorders among survivors of the Oklahoma City bombing. JAMA. 1999;282:755–762.
  3. Gidron Y. Posttraumatic stress disorder after terrorist attacks: a review. J Nerv Ment Dis. 2002;190:118–121.
  4. North CS, Pfefferbaum B, Narayanan P, et al. Comparison of post-disaster psychiatric disorders after terrorist bombings in Nairobi and Oklahoma City. Br J Psychiatry. 2005;86:487–493.
  5. Kessler RC, Sonnega A, Bromet E, et al. Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psych. 1995;52:1048–1060.
  6. Shalev AY, Freedman S. PTSD following terrorist attacks: a prospective evaluation. Am J Psychiatry. 2005;162:1188–1191.
  7. Abenhaim L, Dab W, Salmi LR. Study of civilian victims of terrorist attacks (France 1982-1987). J Clin Epidemiol. 1992;45:103–109.
  8. Verger P, Dab W, Lamping DL, et al. The psychological impact of terrorism: an epidemiologic study of posttraumatic stress disorder and associated factors in victims of the 1995-1996 bombings in France. Am J Psychiatry. 2004;161:1384–1389.
  9. Tucker P, Pfefferbaum B, Nixon SJ, et al. Predictors of post-traumatic stress symptoms in Oklahoma City: exposure, social support, peri-traumatic responses. J Behav Health Serv Res. 2000;27:406–416.
  10. Wilson FC, Poole AD, Trew K. Psychological distress in police officers following critical incidents. Irish Journal of Psychology. 1997;18:321–340.
  11. Yehuda R. Biology of posttraumatic stress disorder. J Clin Psychiatry. 2001;62(suppl 17):41–46.
  12. Goenjian AK, Yehuda R, Pynoos RS, et al. Basal cortisol, dexamethasone suppression of cortisol and MHPG in adolescents after the 1988 earthquake in Armenia. Am J Psychiatry. 1996;153:929–934.
  13. Leucken LJ, Dausch B, Gulla V, et al. Alterations in morning cortisol associated with PTSD in women with breast cancer. J Psychosom Res. 2004;56:13–15.
  14. Bremner JD, Vythilingam M, Anderson G, et al. Assessment of the hypothalamic-pituitary-adrenal axis over a 24-hour diurnal period and in response to neuroendocrine challenges in women with and without childhood sexual abuse and posttraumatic stress disorder. Biol Psychiatry. 2004;54:710–718.
  15. Stein MB, Yehuda R, Koverola C, et al. Enhanced dexamethasone suppression of plasma cortisol in adult women traumatized by childhood sexual abuse. Biol Psychiatry. 1997;42:680–686.
  16. Lemieux AM, Coe CL. Abuse-related posttraumatic stress disorder: evidence for chronic neuroendocrine activation in women. Psychosom Med. 1995;57:105–115.
  17. Liberzon I, Abelson JL, Flagel SB, et al. Neuroendocrine and psychophysiologic responses in PTSD: a symptom provocation study. Neuropsychopharmacology. 1999;21:40–50.
  18. Tucker P, Beebe KL, Burgin C, et al. Paroxetine treatment of depression with PTSD: effects on autonomic reactivity and cortisol secretion. J Clin Psychopharm. 2004;24:131–140.
  19. Tucker P, Ruwe WD, Masters BK, et al. Neuroimmune and cortisol changes in selective serotonin reuptake inhibitor and placebo treatment of chronic posttraumatic stress disorder. Biol Psychiatry. 2004;56:121–128.
  20. Young EA, Breslau N. Cortisol and catecholamines in posttraumatic stress disorder: an epidemiologic community study. Arch Gen Psychiatry. 2004;61:394–401.
  21. Pitman RK, Orr SP, Forgue DF, et al. Psychophysiological assessment of posttraumatic stress disorder imagery in Vietnam combat veterans. Arch Gen Psychiatry. 1987;44:970–975.
  22. McFall ME, Murburg MM, Ko GN, et al. Autonomic responses to stress in Vietnam combat veterans with posttraumatic stress disorder. Biol Psychiatry. 1990;27:1165–1175.
  23. Orr SP, Metzger LJ, Lasko NB, et al. Physiologic responses to sudden, loud tones in monozygotic twins discordant for combat exposure: association with posttraumatic stress disorder. Arch Gen Psychiatry. 2003;60:283–288.
  24. Tucker PM, Pfefferbaum B, North CS, et al. Physiologic reactivity despite emotional resilience several years after direct exposure to terrorism. Am J Psychiatry. 2007;164:230–235.
  25. Liberzon I, Abelson JL, Flagel SB, et al. Neuroendocrine and psychophysiologic responses in PTSD: a symptom provocation study. Neuropsychopharmacol. 1999;19:40–50.
  26. Boscarino JA. Posttraumatic stress disorder and physical illness: results from clinical and epidemiologic studies. Ann N Y Acad Sci. 2004;1032:141–153.
  27. Hoge CW, Terhakopian A, Castro CA, et al. Association of posttraumatic stress disorder with somatic symptoms, health care visits and absenteeism among Iraq war veterans. Am J Psychiatry. 2007;164:150–153.
  28. Schnurr PP, Green BL. Understanding relationships among trauma, post-traumatic stress disorder, and health outcomes. Adv Mind Body Med. 2004;20:18–29.
  29. Boscarino JA. Posttraumatic stress disorder and mortality among U.S. Army veterans 30 years after military service. Ann Epidemiol. 2006;16:248–256.
  30. Kang HK, Bullman TA, Taylor JW. Risk of selected cardiovascular diseases and posttraumatic stress disorder among former World War II prisoners of war. Ann Epidemiol. 2006;16:381–386.
  31. Boscarino JA, Chang J. Electrocardiogram abnormalities among men with stress-related psychiatric disorders: implications for coronary heart disease and clinical research. Ann Behav Med. 1999;21:227–234.
  32. Kubzansky LD, Koenen KC, Spiro A, et al. Prospective study of posttraumatic stress disorder symptoms and coronary heart disease in the normative aging study. Arch Gen Psychiatry. 2007;64:109–116.
  33. Doerfler LA, Paraskos JA. Anxiety, posttraumatic stress disorder, and depression in patients with coronary heart disease: a practical review for cardiac rehabilitation professionals. J Cardiopulm Rehab. 2004;24:414–421.
  34. Shalev AY, Videlock EJ, Peleg T, et al. Stress hormones and posttraumatic stress disorder in civilian trauma victims: a longitudinal study. Part I: HPA axis responses. Int J Neuropsychopharmacol. 2008;11:365–372.
  35. Bierer LM, Tischler L, Labinsky E, et al. Clinical correlates of 24-h cortisol and norepinephrine excretion among subjects seeking treatment following the World Trade Center attacks on 9/11. Ann N Y Acad Sci. 2006;1071:514–520.
  36. Yehuda R, Engel SM, Brand SR, et al. Transgenerational effects of posttraumatic stress disorder in babies of mothers exposed to the World Trade Center attacks during pregnancy. J Clin Endocrinol Metab. 2005;90:4115–4118.
  37. Lampert R, Baron SJ, McPherson CA, et al. Heart rate variability during the week of September 11, 2001. JAMA. 2002;288:575.
  38. Matsuo K, Kato T, Taneichi K, et al. Activation of the prefrontal cortex to trauma-related stimuli measured by near-infrared spectroscopy in posttraumatic stress disorder due to terrorism. Psychophysiology. 2003;49:492–500.
  39. North CS. The course of post-traumatic stress disorder after the Oklahoma City bombing. Mil Med. 2001;166(suppl 12):51–52.
  40. Aardal E, Holm AC. Cortisol in saliva—reference ranges and relation to cortisol in serum. Eur J Clin Chem Clin Biochem. 1995;33:927–932.
  41. Robins LN, Cottler LB, Compton WM, et al. Diagnostic Interview Schedule. Version 4.2. St. Louis, MO: Washington University Department of Psychiatry; 1998.
  42. North CS, Pfefferbaum B. The Diagnostic Interview Schedule/Disaster Supplement Questionnaire. St. Louis, MO: Washington University; 2002.
  43. Robins LN, Smith EM. The Diagnostic Interview Schedule/Disaster Supplement. St. Louis, MO: Washington University; 1992.
  44. Badrick E, Kirschbaum C, Kumari M. The relationship between smoking status and cortisol secretion. J Clin Endocrinol Metab. 2007;92:819–824.
  45. Elzinga BM, Schmahl CG, Vermetten E, et al. Higher cortisol levels following exposure to traumatic reminders in abuse-related PTSD. Neuropsychopharmacology. 2003;28:1656–1665.
  46. Nemeroff CB, Bremner JD, Foa EB, et al. Posttraumatic stress disorder: a state-of-the-science review. J Psychiatric Res. 2006;40:1–21.
  47. Maes M, Bosmans E, Meltzer HY. Immunoendocrine aspects of major depression. Relationships between plasma interleukin-6 and soluble interleukin-2 receptor, prolactin and cortisol. Eur Arch Psychiatry Clin Neurosci. 1995;245:172–178.
  48. Plata-Salamán CR. Immunoregulators in the nervous system. Neurosci Biobehav Rev. 1991;152:185–215.
  49. Boscarino JA. Psychobiologic predictors of disease mortality after psychological trauma: implications for research and clinical surveillance. J Nerv Ment Dis. 2008;196.

CORRESPONDENCE: Phebe Tucker, MD, Department of Psychiatry and Behavioral Sciences, University of Oklahoma College of Medicine, WP 3440, PO Box 26901, Oklahoma City, OK 73190 USA E-MAIL: