Veterans of the first Gulf War (Operations Desert Shield and Desert Storm) have reported medical symptoms and complaints leading some investigators to propose the existence of a “Gulf War syndrome.” Reported symptoms range in severity from skin rash to neurologic complaints and cognitive impairment.1–3 The results of 4 large population-based studies suggest there is no syndrome unique to Gulf War veterans.4–7 Nonetheless, these complaints are frequently reported by Gulf War veterans and, to some extent, remain unexplained.
We have conducted a large population-based telephone survey of Gulf War veterans and a comparison group of nondeployed military personnel.8 We found that Gulf War veterans had much higher rates of self-reported symptoms of cognitive dysfunction, chronic fatigue, bronchitis, asthma, fibromyalgia, depression, anxiety, posttraumatic stress disorder, alcohol abuse, and sexual discomfort, similar to other large-scale studies.5,9,10 Many of the symptoms reported by Gulf War veterans were consistent with known anxiety syndromes such as panic disorder, generalized anxiety disorder, and posttraumatic stress disorder.5,9–13 In some cases, anxiety-based syndromes appeared to have contributed to the veteran being hospitalized.14,15
Given the frequency of anxiety symptoms among veterans and the overlap among the various anxiety disorders, we felt that further analysis of the data set was warranted. We hypothesized that anxiety conditions could potentially explain many of the physical symptoms reported by ill military personnel returning from the Gulf.16 We further hypothesized that the prevalence of anxiety symptoms overall, as well as 3 specific conditions (panic disorder, generalized anxiety disorder, and posttraumatic stress disorder) would be strongly related to Gulf deployment, and to combat specifically. Another goal was to explore the effect of predeployment physical and emotional health on the self-reporting of anxiety disorders. Posttraumatic stress disorder and its associated symptoms have been partially explored elsewhere.17
The sampling design and survey methods have been detailed elsewhere.16 The Statistical Laboratory Survey Section of the Department of Statistics at Iowa State University conducted the telephone interviews, which took place between September 1995 and May 1996.8,16 Briefly, 4886 subjects were randomly selected from Gulf War regular military, Gulf War National Guard/Reserve, non-Gulf War regular military, and non-Gulf War National Guard/Reserve. Military personnel were eligible if Iowa was listed as the home of record on their initial military record and if the person had served in a regular military or in an activated National Guard or Reserve unit sometime between August 2, 1990, and July 31, 1991. This time period included Operations Desert Shield and Desert Storm. (Non-Gulf War military personnel were on active duty at the time of the first Gulf War but were not deployed to the Gulf.) For National Guard/Reserve personnel to be eligible for the study, they had to be classified as National Guard or United States Reserve personnel at some time during the same time period as the Gulf War.
Study procedures and instruments were approved by the Institutional Review Board of the University of Iowa College of Medicine. Verbal consent of study participants was documented by the interviewer.
We used a structured telephone interview to assess a broad range of health concerns and to determine the prevalence of symptoms suggestive of specific conditions.8,16 The PRIME-MD,18 based on the fourth edition of the Diagnostic and Statistic Manual of Mental Disorders (DSM-IV),19 was the source of questions and definitions for major depression, panic disorder, and generalized anxiety disorder. This structured instrument was developed for use in primary care settings to identify current psychiatric disorders. Panic disorder required evidence of one or more spontaneous attacks accompanied by impairment during the past month. Generalized anxiety disorder required 3 or more symptoms of anxiety for the past year and anxiety or worry more days than not during the past 6 months. The instrument shows satisfactory sensitivity and excellent specificity for both diagnoses. Its overall diagnostic accuracy is 96% for panic disorder and 94% generalized anxiety disorder, respectively.19
Formal reliability studies involving telephone administration of the PRIME-MD have not been done, but the study by Kobak et al.20 is relevant. These investigators compared telephone administration by trained raters with a computer-generated form using interactive voice response technology. Both forms produced roughly equivalent levels of sensitivity and specificity, and showed an overall kappa of 0.67, suggesting that telephone administration produces acceptable results.
To screen for current (past month) posttraumatic stress disorder, we used the Posttraumatic Stress Disorder Checklist,21 which is widely accepted and has excellent reliability and validity. Its cutoff value of ≥50 has been established using a receiver operator characteristic curve compared with a structured psychiatric interview.
Current mood disorders were assessed using the PRIME-MD. Major depression in the past month required 5 or more depressive symptoms. Minor depression was defined as feeling depressed or hopeless, or expressing little interest or pleasure in life, and having 2 to 4 additional depressive symptoms rated moderate to severe during the past month. Dysthymia required depressive symptoms that have caused impairment for 2 years or more. This definition corresponds to the DSM-IV diagnosis but does not fully overlap. Current (past month) alcohol abuse was defined from items taken from the PRIME-MD18 and the CAGE Questionnaire.22 These instruments have been shown to be both sensitive and specific in detecting alcoholism.18,23,24 The psychiatric conditions assessed were not mutually exclusive.
Two items taken from the Marlow-Crowne Social Desirability Scale25 describe antisocial-like behavior: 1) “getting even” rather than forgiving and forgetting, and 2) having times when the person felt like “smashing things.” The questionnaire also asked about incarceration, which provided further information about antisocial behavior.
To examine the role of prior medical problems, we asked subjects about 53 physical conditions, whether these conditions were present in the year before the telephone survey and whether they first appeared before August 1990. For analysis, we trichotomized the number of physical conditions (0, 1–4, 5+).
Level of military preparedness at the time of deployment was based on the sum of responses to 6 questions.26 These questions asked how prepared or trained the veterans felt they were to do their job. For analysis, military preparedness was dichotomized to “less/moderately prepared” (a “yes” answer on up to 5 of the items) versus “most prepared” (a “yes” answer on all 6).
All descriptive and univariate statistical analyses used SUDAAN (Research Triangle Institute, Research Triangle Park, NC) to account for the complex sampling design. Prevalence rates of overall anxiety and specific anxiety disorders are population estimates using the sampling weights inherent in the survey design. We computed prevalence rates and standard errors (SEs) for each of the 4 military groups using the cross-tabulation procedure. SUDAAN linear regression was used to compare continuous variables for veterans with and without anxiety. Logistic regression analysis was performed to compare prevalence rates of overall anxiety and specific anxiety disorders among each of the 4 military groups and to examine risk factors for an anxiety condition among those deployed to the Gulf. All logistic regression models included the study design stratification variables (age, sex, race, rank, branch of service, and military status). Adjusted models included the study design stratification variables and also adjusted for pre-Gulf War mental health conditions (posttraumatic stress disorder, depression, or anxiety). We reported adjusted odds ratios (ORs) and 95% confidence intervals (CIs).
Because SUDAAN does not provide variable selection technique, SAS (SAS Institute, Cary, NC) stepwise regression was used as the first step to identify independent risk factors for anxiety disorders in Gulf War veterans. The resulting multiple logistic regression model was then evaluated in SUDAAN, using backward elimination procedure to find predictors significant at alpha = 0.05 (2-tailed). All stratification variables were left in the model.
The sample included 4886 eligible participants. Overall, 3695 subjects completed the interview (76% of those eligible and 91% of those for whom a valid telephone number was identified). Most participants were married men 25 years or younger. Nearly all were white, reflecting the general population of Iowa. Most had at most achieved a high school education. Military personnel were mainly enlisted, not officers. The Army was the most heavily represented branch of service and the Air Force the least,8 reflecting the overall military population at that time.
The estimated prevalence of current anxiety conditions is shown in Table 1. A total of 163 (3.7%) subjects overall met criteria for at least one of the 3 specified anxiety conditions. Of these, 76 (47%) met criteria for panic disorder, 108 (66%) met criteria for generalized anxiety disorder, and 53 (33%) met criteria for posttraumatic stress disorder. There was substantial overlap among the conditions. Veterans deployed to the Gulf had approximately twice the rates for all anxiety conditions compared with non-Gulf War military personnel.
Anxious Gulf War veterans were slightly older than nonanxious veterans (26.7 [SE = 0.7] vs. 25.8 [SE = 0.8]) and had substantially less time in the military (mean difference in months = -11.2; SE = 5.5), and less time on active duty (mean difference in months = -12.5; SE = 5.4).
Table 2 compares Gulf War and non-Gulf War military personnel for the specified anxiety conditions. Gulf War veterans had approximately twice the odds for panic disorder, generalized anxiety disorder, posttraumatic stress disorder, and any anxiety disorder compared with non-Gulf War military personnel. Among those serving in the Gulf, National Guard/Reserve personnel were at lower risk of symptoms (especially panic disorder) than regular military personnel. Overall, anxiety conditions were more common among Gulf War veterans than those not deployed to the Gulf (OR = 2.1; 95% CI = 1.3–3.1). Additional adjustment for predeployment diagnoses (posttraumatic stress disorder, depression, and anxiety) had little effect on the results.
The remainder of the analyses focuses on Gulf War veterans. Table 3 shows the psychiatric and medical comorbidity of Gulf War veterans with and without anxiety. Both medical and psychiatric health conditions were highly prevalent among veterans with anxiety. These other conditions included symptoms of cognitive dysfunction (85%), any depression (80%), symptoms of fibromyalgia (54%), dysthymia (47%), reported injuries (37%), and symptoms of asthma (29%).
Table 4 displays the analysis of potential risk factors among those deployed to the Gulf (demographic, military, personality, and illness) for those who satisfied criteria for an anxiety condition compared with those who did not. These ORs are adjusted for age, sex, race, branch of military service, and rank. Several factors are associated with an anxiety condition, including service in the regular military, service specifically in the Army, and service as an enlistee. Other important risk factors included income <$30,000 yearly, current cigarette smoking, active duty status in 1995, lower levels of preparedness for Gulf deployment, ever incarcerated, and the presence of antisocial traits.
Participation in combat emerged as an important risk factor for the development of anxiety. This variable was strongly related to all 3 anxiety conditions: posttraumatic stress disorder (OR = 2.1; CI = 1.7–4.2), panic disorder (6.2; 3.0–13.0), and generalized anxiety disorder (2.1; 1.2–3.5). Gulf War veterans remaining on active duty at the time of the interview (1995/1996) were less likely to have an anxiety condition, as were those who reported higher levels of military preparedness before Operations Desert Shield and Desert Storm.
Those with an anxiety disorder were more likely to have seen a mental health professional, taken psychotropic medication, ever been psychiatrically hospitalized, and had a preexisting condition of depression, anxiety, or a physical disorder before deployment. Subjects with 5 or more physical disorders before deployment had an odds ratio for anxiety of nearly 25.
Table 5 shows the results of the multivariate logistic regression. The variables in Table 4 were all included, except that pre-Gulf War posttraumatic stress disorder, depression, and anxiety were collapsed into a single variable, and psychiatric hospitalization was omitted because of its low prevalence rate among deployed veterans with anxiety. The model was well calibrated (c-stat = 0.824). All variables except military branch were strongly associated with the development of an anxiety disorder.
Veterans of the first Gulf War are at high risk for anxiety and nearly twice as likely as Gulf War-era control subjects to report symptoms of panic disorder, generalized anxiety disorder, and posttraumatic stress disorder. Rates among era control subjects for the 3 relatively common disorders were near the expected prevalence in the general population.27,28 These figures are for current disorders, assessed at the time of the 1995–1996 interview, and are not meant to reflect the subject's condition after Operations Desert Shield and Desert Storm. Gulf deployment was associated with these anxiety disorders, as was combat itself, dispelling the view that the only important psychologic consequence of wartime involvement is posttraumatic stress disorder.29
Gulf deployment led some persons to experience life-threatening situations, and so the higher rates of posttraumatic stress disorder among Gulf War veterans are not unexpected. In fact, posttraumatic stress disorder is one of the few disorders in the DSM-IV for which exposure to traumatic events is required to make the diagnosis.19
More remarkable were the increased rates of panic disorder and generalized anxiety disorder. These disorders are not generally considered consequences of wartime exposures. However, this study, as well as earlier work,30,31 suggests that these conditions frequently occur in veterans exposed to combat, and they should be recognized as potential complications of wartime situations. Mellman et al.30 studied World War II, Korean War, and Vietnam War era combat veterans and found high lifetime rates of posttraumatic stress disorder, depression, and other anxiety disorders. Those authors conclude that the “emergence of generalized anxiety disorder suggests that symptoms of the disorder, like posttraumatic stress disorder, may represent a primary response to trauma.”30(p.1573) We believe the same could be said for panic disorder.
It is well accepted that persons with posttraumatic stress disorder have elevated rates of other anxiety disorders.32,33 Our study is one of the first to show that rates of other anxiety disorders are increased even in combat veterans without posttraumatic stress disorder. Because the interviews took place nearly 4 years after the Gulf War, and reflect conditions as of 1995–1996, we cannot exclude the possibility that veterans with panic disorder or generalized anxiety disorder had experienced posttraumatic stress disorder in the past but no longer had symptoms at the time of the interview.
From a historic perspective, many symptoms of the “war syndromes” discussed by Hyams et al.34 are fully consistent with anxiety disorders. Using medical records dating to the Civil War, Hyams and colleagues showed convincingly that war veterans from all eras have developed similar war-related symptoms. It is unlikely that the increased rates of anxiety disorders seen in our veterans are specific to Gulf deployment, but probably reflects the consequences of wartime experiences in general. Although all Gulf War veterans would have experienced stressful situations, not all would have been directly threatened or exposed to combat. The intensity of the stressor, interacting with the subject's personality and predeployment emotional health, could lead him or her to develop a particular type of anxiety disorder. For example, veterans exposed to intensive combat might develop posttraumatic stress disorder, whereas those with less intensive exposures might develop panic disorder or generalized anxiety disorder.
Interestingly, officers were less likely than enlisted personnel to develop anxiety disorders. Officers are more highly educated and could be better trained; for that reason, they could be better prepared for military service and combat. Possibly, officers are less likely to develop symptoms than enlisted personnel when faced with traumatic circumstances. Level of military preparedness and other military factors that vary by branch could partially account for Army personnel having an increased likelihood of developing anxiety. Because branch of service was not an independent risk factor for anxiety in our multivariate analysis, it could be that other factors mediated this univariate association. The fact that individuals remaining on active duty as of 1995–1996 had lower rates of anxiety suggests that persons developing symptoms could have left or been discharged from the service. On the other hand, it could be that those dedicated to a military career are better trained and prepared, and therefore less likely to develop anxiety.
Although the cause of anxiety disorders is not fully known, Barlow35 has suggested a model that considers both biologic and psychologic vulnerability. He proposes that a vulnerable person subjected to a traumatic situation could develop posttraumatic stress disorder. Data show that persons with posttraumatic stress disorder often have preexisting psychiatric disorders that create an emotional vulnerability.36 These individuals are also reported to have biologic vulnerabilities as well, including a dysregulation of the hypothalamic-pituitary-adrenal axis.37 With panic disorder, Barlow proposes that the vulnerable person experiences (in response to stressful situations) a “false alarm,” or spontaneous panic attack. In contrast, with generalized anxiety disorder, stress leads to generalized anxiety without the false alarms. These theories are consistent with the view that panic disorder and generalized anxiety disorder are genetically mediated in part, and that each is associated with biologic and experiential alterations, such as CO2 sensitivity, that could lead to a constitutional vulnerability.38,39
Although we have no data on biologic vulnerability factors (eg, family–genetic information) among personnel involved in this study, veterans who developed an anxiety disorder were more likely than control subjects to show evidence of emotional vulnerability. Some of the most important predevelopment factors included prior treatment by a mental health professional, having taken psychotropic medication, having been psychiatrically hospitalized, or having prior diagnoses of depression, anxiety, or some medical condition. Self-reported predeployment medical and psychiatric illness appears to be among the most robust correlates of post-Gulf War anxiety syndromes.
The 3 anxiety disorders were associated with considerable psychiatric and medical comorbidity, including other anxiety disorders, mood disorders, and substance abuse and dependence. The presence of one condition could increase the likelihood that another will be found. These results are consistent with clinical and epidemiologic data on anxiety disorders confirming their relationship with comorbid disorders.40–42 Furthermore, anxiety conditions are often associated with medical complaints.43 Anxious patients manifest somatic symptoms such as palpitations, shortness of breath, and dizziness. These data should alert military planners that anxious veterans are likely to meet criteria for multiple syndromes, all of which need to be considered by medical personnel.
Cigarette smoking emerged as an important correlate of anxiety. Several large epidemiologic studies44,45 have linked smoking with anxiety and other mental disorders. It remains unclear whether cigarette smoking itself induces the symptoms, or the behavior reflects an underlying predisposition to the development of anxiety or other syndromes.
The association of antisocial characteristics and behavior with anxiety disorders could reflect the fact that persons with antisocial or criminal tendencies frequently develop comorbid psychiatric and physical complaints.46,47
There are several potential methodologic limitations. First, recall bias can substantially alter the frequency of reports of symptoms, particularly in retrospective studies in which people are interviewed at a point remote from some event.48,49 Media interest in Operations Desert Shield and Desert Storm has been intense. Persons sensitized by the media (or medical professionals) about the relationship between their symptoms and Gulf War experiences might report more symptoms. Our cross-sectional study design limits our ability to make causal statements. The study was limited to Iowans at the time of enlistment, and whether these results are generalizable to other populations will need to be demonstrated. For example, although the odds ratio for developing anxiety was nearly 2 in nonwhites, our power to detect differences between racial groups is limited. However, previous results from our study have been comparable to other large population-based studies, which argues the present results are likely generalizable as well.4–7 Although the medical and psychiatric conditions are based exclusively on self-report, we used validated instruments for their assessment. In particular, we believe diagnoses of posttraumatic stress disorder, panic disorder, and generalized anxiety disorder in this study can be viewed with confidence. The PRIME-MD shows excellent diagnostic accuracy for the anxiety conditions studied here,18 as does the posttraumatic stress disorder checklist.21 We are currently gathering data through face-to-face interviews to investigate more fully the validity of these measures in our subjects.
These data show that veterans of the first Gulf War are at risk for the development of posttraumatic stress disorder, generalized anxiety disorder, and panic disorder. Exposure to combat was independently associated with each condition, suggesting that wartime situations can contribute to the development of several anxiety disorders, not just posttraumatic stress disorder. Factors suggestive of prior psychiatric difficulties were strongly correlated with these anxiety syndromes. Other predeployment factors, including medical morbidity, level of preparedness, and cigarette smoking, were also important correlates of anxiety. Because panic disorder, generalized anxiety disorder, and posttraumatic stress disorder are treatable,50–56 it is important that healthcare professionals be alert to their presence in veteran populations.
We appreciate the support and participation of investigators at the University of Iowa, Iowa Department of Public Health, Centers for Disease Control and Prevention, and Iowa State University Statistical Laboratory. We also acknowledge Mike Dove and the Defense Manpower Data Center for their assistance in making data available to draw the sample. We appreciate the secretarial support provided by Liz Smothers.
The Scientific Advisory Committee provided advice, guidance, and critique throughout all phases of this investigation. The Public Advisory Committee established an open dialogue between the study personnel and veterans, veteran organizations, and affected individuals.
1. NIH Technology Assessment Workshop Panel. The Persian Gulf experience and health. JAMA.
2. Persian Gulf Veterans Coordinating Board. Unexplained illnesses among desert storm veterans: a search for causes, treatment, and cooperation. Arch Intern Med.
3. Committee to Review the Health Consequences of Service During the Persian Gulf War. Health Consequences of Service During the Persian Gulf War: Initial Findings and Recommendations for Immediate Action
. Washington, DC: Academic Press; 1994.
4. Doebbeling BN, Clarke WR, Watson D, et al. Is there a Persian Gulf War syndrome? Evidence from a large population-based survey of veterans and non-deployed controls. Am J Med
5. Fukuda K, Nisenbaum R, Stewart G, et al. Chronic multisymptom illness affecting Air Force veterans of the Gulf War. JAMA
6. Ismail K, Everitt B, Blatchley N, et al. Is there a Gulf War syndrome? Lancet
7. Knoke JD, Smith TC, Gray GC. Factor analysis of self-reported symptoms: does it identify a Gulf War syndrome? Am J Epidemiol
8. Iowa Persian Gulf Study Group.Self-reported illness and health status among Gulf War veterans: a population-based study. JAMA
9. Unwin C, Blatchley N, Coker W, et al. Health of UK servicemen who served in the Persian Gulf War. Lancet
10. Perconte ST, Wilson AT, Pontius EB, et al. Psychological and war stress symptoms among deployed and non-deployed reservists following the Persian Gulf war. Milit Med
11. McCarroll JE, Ursano RJ, Fullerton CS. Symptoms of posttraumatic stress disorder following recovery of war dead: 13–15 months following. Am J Psychiatry
12. Gray GC, Kaiser KS, Hawksworth AW, et al. Increased postwar symptoms and psychological morbidity among US Navy Gulf War veterans. Am J Trop Med Hyg
13. Baker DG, Mendenhall CL, Simbartl LA. Relationship between posttraumatic stress disorder and self-reported physical symptoms in Persian Gulf War veterans. Arch Intern Med
14. Dlugosz L, Hocter WJ, Kaiser KS, et al. Risk factors for hospitalization after the Persian Gulf War: US Armed Forces, June 1, 1991–September 30, 1993. J Clin Epidemiol
15. Gray GC, Coate BD, Anderson CM, et al. The postwar hospitalization experience of US veterans of the Persian Gulf War. N Engl J Med
16. Doebbeling BN, Jones MF, Hall DB, et al. Methodological issues in a population-based health survey of Gulf War veterans. J Clin Epidemiol
17. Barrett DH, Doebbeling CC, Schwartz DA, et al. Posttraumatic stress disorder and self-reported physical health status among US military personnel serving during the Gulf War Period: A population-based study. Psychosomatics.
18. Spitzer RL, Williams JBW, Kroenke K, et al. Utility of a new procedure for diagnosing mental disorders in primary care: the PRIME-MD 1000 study. JAMA
19. Diagnostic and Statistical Manual of Mental Disorders
, 4th ed. Washington, DC: American Psychiatric Press; 1994.
20. Kobak KA, Taylor VH, Dohi SL, et al. A computer-administered telephone interview to identify mental disorders. JAMA
21. Weathers F, Litz BT, Heran DS, et al. Posttraumatic stress disorder checklist (PCL): reliability, validity, and diagnostic utility.
Paper presented at a meeting of the International Society of Traumatic Stress Studies; San Antonio, TX; October 1993.
22. Ewing JA. Detecting alcoholism: The CAGE questionnaire. JAMA
23. Staab JP, Datto CJ, Weinrieb RM, et al. Detection and diagnosis of psychiatric disorders in primary medical care settings. Med Clin North Am
24. Escobar F, Espi F, Canteras M. Diagnostic tests for alcoholism in primary health care: compared efficacy of different instruments. Drug Alcohol Depend
25. Strahan R, Gerbasi KC. Short, homogenous version of the Marlow-Crowne Social Desirability Scale. J Clin Psychol
26. Voelker MD, Saag KG, Schwartz DA, et al. Health-related quality of life among Gulf War era veterans. Am J Epidemiol.
27. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States—results from the National Comorbidity Survey. Arch Gen Psychiatry
28. Kessler RC, Sonnega A, Bromet E, et al. Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry
29. Haley RW. Is Gulf War syndrome due to stress? The evidence reexamined. Am J Epidemiol
30. Mellman TA, Randolph CA, Brawman-Mintzer O, et al. Phenomenology and course of psychiatric disorders associated with combat-related posttraumatic stress disorder. Am J Psychiatry
31. Centers for Disease Control Vietnam Experience Study. Health status of Vietnam veterans. I. Psychosocial characteristics. JAMA.
32. Breslau N, Davis GC, Andreski P, et al. Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry
33. North C, Nixon S, Shariat S, et al. Psychiatric disorders among survivors of the Oklahoma City bombing. JAMA
34. Hyams KC, Wignall S, Roswell R. War syndromes and their evaluation: from the US Civil War to the Persian Gulf War. Ann Intern Med
35. Barlow DH. Anxiety and Its Disorders
, 2nd ed. New York: Guilford Press; 2002.
36. Smith EM, North CS, McCool RE, et al. Acute post-disaster psychiatric disorders: identification of persons at risk. Am J Psychiatry
37. Yehuda R. Biology of posttraumatic stress disorder. J Clin Psychiatry
. 2001;62(suppl 17):41–46.
38. Johnson MR, Lydiard RB. The neurobiology of anxiety disorders. Psychiatr Clin North Am
39. Jetty PV, Charney DS, Goddard AW. Neurobiology of generalized anxiety disorder. Psychiatr Clin North Am
40. Noyes R. Comorbidity in generalized anxiety disorder. Psychiatr Clin North Am
41. Massion AO, Warshaw MG, Keller MB. Quality of life and psychiatric morbidity in panic disorder and generalized anxiety disorder. Am J Psychiatry
42. Breslau N. The epidemiology of posttraumatic stress disorder: what is the extent of the problem? J Clin Psychiatry
. 2001;2(suppl 17):16–22.
43. Katon WJ, Von Korff M, Lin E. Panic disorder: relationship to high medical utilization. Am J Med.
44. Breslau N, Kilbey M, Andreski P. Nicotine dependence, major depression, and anxiety in young adults. Arch Gen Psychiatry
45. Black DW, Zimmerman M, Coryell WH. Cigarette smoking and psychiatric disorder in a community sample. Ann Clin Psychiatry
46. Robins LN. Deviant Children Grown Up
. Baltimore: Williams and Wilkins; 1966.
47. Black DW. Bad Boys, Bad Men—Confronting Antisocial Personality Disorder
. New York: Oxford University Press; 1999.
48. Coughlin SS. Recall bias in epidemiologic studies. J Clin Epidemiol
49. Neugebauer R, Ng S. Differential recall as a source of bias in epidemiologic research. J Clin Epidemiol
50. American Psychiatric Association.Practice guidelines for the treatment of patients with panic disorder. Am J Psychiatry
. 1998;155(suppl 5):1–34.
51. Foa EB, Davidson JRT, Frances A, eds. Expert Consensus Guideline Series—treatment of posttraumatic stress disorder. J Clin Psychiatry
. 1999;60(suppl 60):1–76.
52. Falsetti SA, David J. The nonpharmacologic treatment of generalized anxiety disorder. Psychiatr Clin North Am
53. Brawman-Mintzer O. Pharmacologic treatment of generalized anxiety disorder. Psychiatr Clin North Am
54. Marks I, Lovell K, Noshirvani H, et al. Treatment of posttraumatic stress disorder by exposure and/or cognitive restructuring—a controlled study. Arch Gen Psychiatry
55. Brady K, Pearlstein T, Asnis GM, et al. Efficacy and safety of sertraline treatment of posttraumatic stress disorder: a randomized controlled trial. JAMA.
56. Davidson JRT, DuPont RL, Heges D, et al. Efficacy, safety, and tolerability of venlafaxine extended release and buspirone in outpatients with generalized anxiety disorder. J Clin Psychiatry