Military Medicine: Lifetime and 12-Month Prevalence of Psychiatric Disorders in 8,169 Male Vietnam W

Objective: This study reports the prevalence of psychiatric disorders among a nationally distributed sample of Vietnam Era veterans assessed using standardized psychiatric interviewing methods. Methods: In 1992, the National Institute of Mental Health Diagnostic Interview Schedule was administered by telephone to 8,169 middle-aged males who served in the military during the Vietnam era (1965-1975). Results: Approximately 72% of respondents reported a lifetime history and 36% reported a 12-month history of at least one psychiatric disorder. The most prevalent psychiatric disorders included alcohol abuse and/or dependence (54% lifetime, 17% 12 month), nicotine dependence (48% lifetime, 22% 12 month), and posttraumatic stress disorder (10% lifetime, 4.5% 12 month). Conclusions: Because of possible participation bias, these results likely represent conservative estimates of psychiatric disorder prevalences among the more than eight million Vietnam Era veterans and reinforces the major public health challenge of preventing, identifying, and treating psychiatric illness in American veterans.

Introduction

Studies of the prevalence and associations of psychiatric disorders in populations can provide the basis for developing hypotheses about the etiology of the disorders and assist in formulating priorities for prevention and treatment. Several psychiatric epidemiological studies published during the last 20 years used large sample sizes and reliable diagnostic methods and sampling techniques to assess the prevalence of psychiatric disorders, the extent of comorbidity, and demographic correlates of psychiatric illnesses in different populations.1-5

To date, the primary sources of information on the prevalence of psychiatric disorders in the United States are the Epidemiological Catchment Area Study (ECA)5 and the National Comorbidity Study (NCS).2 In the ECA, more than 20,000 respondents in five communities were administered a structured interview to determine the prevalence of psychiatric disorders as defined primarily by the Diagnostic and Statistical Manual of Mental Disorders, Third Edition (DSM-III).6 The NCS administered the University of Michigan-Composite International Diagnostic Interview (UM-CIDI) to a representative U.S. sample of 8,098 respondents to diagnose psychiatric disorders according to Diagnostic and Statistical Manual of Mental Disorders, Third Edition Revised (DSM-III-R).7

Two large-scale studies have examined the prevalence of psychiatric disorders among veterans who served in the military during the Vietnam Era: the Centers for Disease Control Vietnam Experience Study (VES), a multidimensional psychological assessment on a random subsample of 4,462 Vietnam veterans;8 and the National Vietnam Veterans Readjustment Study (NWRS), which assessed the prevalence of psychiatric disorders in 2,348 Vietnam veterans.9-11 The present study examines the prevalence of 20 DSM-III-R psychiatric disorders among 8,169 middle-aged male members of the Vietnam Era Twin (VET) Registry, a national database of twin pairs, both of which were in service during the Vietnam Era. In 1992, this nationally distributed sample, identified without regard to psychiatric status or other health characteristics, was administered the National Institute of Mental Health Diagnostic Interview Schedule Version 3, Revised (DIS-HI-R)12 to obtain data on the prevalence of DSM-III-R criteria psychiatric disorders. Because the NWRS and VES used different sampling techniques than the present study, used DSM-III (rather than DSM-III-R) criteria for most diagnoses, and did not report posttraumatic stress disorder (PTSD) based on DSM criteria, our analyses will contribute to the existing literature on veteran mental health. Comparisons to the NWRS and VES will be discussed.

Materials and Methods

Sample

The VET Registry was developed from a Department of Defense computer file of approximately 5.5 million Vietnam Era veterans born between 1939 and 1957 who served on active military duty during the Vietnam War (May 1965 to August 1975). Using an algorithm that matched database entries for same last name, different first name, same date of birth, and similar social security numbers, a total of 7,369 twin pairs were identified. A complete description of the registry's construction and method of zygosity determination is available elsewhere.13-15

The basis for the present report is a structured psychiatric interview that was conducted in 1992. All VET Registry twin pairs in which at least one member responded to a survey conducted in 1987 and his cotwin had not refused to participate in future studies and was not known to be deceased were eligible for the 1992 interview. Of these 10,300 eligible individuals (5,129 twin pairs and 42 singletons), 47 were subsequently determined to be deceased or significantly incapacitated. Of the remaining 10,253 individuals, 8,169 were interviewed (3,516 twin pairs and 1,137 singletons).

Psychiatric Assessment and Diagnosis

In 1992, respondents were interviewed via telephone by trained and experienced interviewers using a computerized version of the DIS-III-R.16 The DIS-III-R is a research diagnostic interview designed to be administered by trained lay interviewers to assess psychiatric disorders according to DSM-III-R criteria.7 Before the interview, potential participants were sent a letter that explained the purpose of the study. One to 2 weeks later, an interviewer phoned the subject, reviewed the research project, and solicited verbal informed consent to participate.

The DIS-III-R is the most widely used assessment instrument for psychiatric diagnosis in community samples. It is designed to elicit psychiatric diagnostic elements including symptoms, symptom severity, frequency, distribution over time, and whether or not symptoms are due to physical illness, use of drugs or alcohol, or the presence of another psychiatric disorder. The following DSM-III-R diagnoses were assessed: alcohol abuse and dependence, nicotine (dependence only), major depressive episode, major depression, bipolar disorder, manic episode, dysthymia, panic disorder, generalized anxiety disorder, PTSD, antisocial personality disorder (ASPD), conduct disorder, pathological gambling, and illicit drug abuse and dependence (amphetamine, cannabis, cocaine, opioids, phencyclidine, lysergic acid diethylamide (LSD), and sedatives). Both lifetime and 12-month prevalence of psychiatric disorders were calculated. Lifetime prevalence is the proportion of the sample who had ever experienced the disorder, while 12-month prevalence is the proportion who experienced the disorder at some time during the 12 months before the interview. Because of unavailability of data, 12-month prevalence rates were not computed for conduct disorder and pathological gambling.

Reliability

To assess reliability, a subset of 146 participants were reinterviewed by an interviewer other than the one who performed the original interview. The mean time between interviews was 466 days (SD, ±50.5; range, 357-601). Kappa statistics for test-retest reliability17 of lifetime prevalence ranged from 0.54 to 0.60 for affective disorders, from 0.25 to 0.32 for anxiety disorders, from 0.26 to 0.76 for substance use disorders, and from 0.34 to 0.41 for "other disorders" (ASPD, conduct disorder, and pathological gambling). The κ statistic for any disorder (excluding nicotine dependence) was 0.78. Overall, the test-retest reliability for most of the DSM-III-R psychiatric diagnoses was satisfactory, especially considering the low prevalence of many of the disorders and the mean duration between interviews of nearly 1.3 years. Reliability was poor (

Data Analysis

We included all 8,169 veterans (twin pairs and singletons combined) in our analyses. We created computer algorithms to derive the lifetime and 12-month diagnostic variables for all 20 psychiatric disorders, the proportion of the sample with comorbid lifetime and 12-month disorders, and the age-adjusted odds ratios (ORs) for the likelihood of lifetime and 12-month psychiatric disorders as a function of demographic variable categories. Additionally, summary algorithms were derived for three categories of psychiatric disorders (Table I: any affective disorder, any anxiety disorder, and any illicit drug abuse and/or dependence) and for three summary variables for all of the psychiatric disorders (any disorder, any disorder except alcohol abuse and/or dependence, and any disorder except alcohol abuse and/or dependence and nicotine dependence). Because 12-month estimates of the prevalence of conduct disorder and pathological gambling could not be calculated for this sample, we did not include these two disorders in the three summary estimates for lifetime and 12-month prevalence of any psychiatric disorder.

Because data derived from twin siblings are not independent observations, proportions, ORs, and 95% confidence intervals (CI) for lifetime and 12-month prevalence and comorbidity estimates were calculated using Stata (Version 5).18 Appropriate estimates of 95% CI of the sample with particular psychiatric disorders, adjusted for the clustered nature of the twin data, were generated based on Huber's robust variance estimator. We identified the twin pair as the primary sampling unit, thereby correcting for the nonindependence of the data. Age-adjusted ORs for the likelihood of particular psychiatric disorders by category of individual demographic variables were also computed using STATA.

Results

In 1992, the mean age of the 8,169 respondents was 42.0 years (SD, 2.8; range, 34-53 years). Most were Caucasian (93.5%), 6.1% were African American,

Prevalence of Psychiatric Disorders

Table I presents the DSM-III-R psychiatric disorder prevalence estimates (percentages and 95% CI) of the 20 lifetime and eighteen 12-month disorders for the 8,169 veterans. Psychiatric disorders are presented in five major categories: affective disorders (major depressive episode, major depression, bipolar disorder, manic episode, and dysthymia), anxiety disorders (panic disorder, generalized anxiety disorder, and PTSD), licit (alcohol abuse and/or dependence, nicotine dependence) and illicit (amphetamine, cannabis, cocaine, opioid, phencyclidine/LSD, sedative abuse and/or dependence) substance use disorders, and "other" disorders (ASPD, conduct disorder, and pathological gambling).

Of the disorders assessed, the prevalence estimates of alcohol abuse and/or dependence and nicotine dependence were particularly high for lifetime (54.6% and 47.6%, respectively) and 12-month (17.3% and 23.3%, respectively) periods. Lifetime prevalence was similar (between 10.2% and 11.9%) for the categories of any affective disorder, any anxiety disorder, and any illicit drug abuse and/or dependence. For 12-month prevalence categories, any anxiety disorder was the highest (5.7%), whereas the prevalence of any affective disorder and any illicit drug abuse and/or dependence were much lower (3.5% and 1.3%, respectively).

Table I demonstrates that about 7 of every 10 respondents reported a lifetime history of at least one psychiatric disorder, while more than 1 of every 3 respondents reported a 12-month histoiy of at least one disorder. If alcohol abuse and/or dependence and nicotine dependence are not included in the category "any disorder," then about 3 in 10 respondents reported a lifetime history of at least one psychiatric disorder and about 1 in 10 respondents had a 12-month history of at least one psychiatric disorder. The prevalence of most of the psyehiatric disorders in the past 12 months was low. Twelve-month prevalence estimates greater than 1.5% were observed only for nicotine dependence, alcohol abuse and/or dependence, PTSD, major depressive episode, and major depression.

Psychiatric Comorbidity

Many respondents met criteria for more than one psychiatric disorder. As shown in Table II, 12.1% of the respondents met criteria for two lifetime disorders (not including nicotine dependence) and 11.2% had three or more lifetime disorders. Approximately 3% had two 12-month disorders and 1.6% had three or more 12-month disorders. Of those with at least one lifetime disorder, 19.9% had two lifetime disorders and 18.4% had three or more lifetime disorders. Of those with at least one 12-month psychiatric disorder, 13.6% had two 12-month disorders, while 6.9% had three or more.

Association of Demographic Characteristics and Psychiatric Disorders

Age-adjusted OR of selected DSM-III-R lifetime psychiatric disorders by race, marital status, education, annual household income, and present employment are reported for five psychiatric disorders (alcohol abuse and/or dependence, PTSD, major depression, any illicit drug abuse and/or dependence, nicotine dependence) in Table III. For all five psychiatric disorders, the risk of having the disorder in one's lifetime was significantly higher for those who were widowed, separated, or divorced (vs. married), who earned less than $20,000/year (vs. those earning $40,000/year or more), and who were unemployed vs. employed. Being never married (vs. married) was statistically significantly associated with an increased likelihood of meeting lifetime criteria for major depression (OR, 1.46; 95% CI, 1.10, 1.94) and any illicit drug abuse and/or dependence (OR, 1.74; 95% CI, 1.35, 2.25). Respondents who were never married were significantly less likely to meet lifetime criteria for nicotine dependence compared with those who were married (OR, 0.78; 95% CI, 0.66, 0.93).

Compared with those with more than a high school education, those with less than a high school education were statistically significantly more likely to meet criteria for alcohol abuse and/or dependence (OR, 1.68; 95% CI, 1.28, 2.21), PTSD (OR, 1.67; 95% CI, 1.17, 2.38), and nicotine dependence (OR, 2.85; 95% CI, 2.18, 3.73). African Americans were statistically significantly less likely than Caucasians to meet the criteria for alcohol abuse and/or dependence (OR, 0.61; 95% CI, 0.50, 0.74) and for nicotine dependence (OR, 0.59; 95% CI, 0.48, 0.72).

Age-adjusted ORs for the associations of demographic factors with the same five psychiatric disorders for the previous 12 months were also calculated (Table IV). Statistical significance was found between each of the five disorders and marital status (widowed/separated/divorced vs. married), low annual household income, and unemployment. Those who were never married (vs. married) were statistically significantly more likely to meet the criteria for alcohol abuse and/or dependence (OR, 1.39; 95% CI, 1.12, 1.72), major depression (OR, 1.80; 95% CI, 1.16, 2.79), and any illicit drug abuse and/or dependence (OR, 3.21; 95% CI, 1.86, 5.54).

A lower level of education was statistically significantly associated with PTSD (OR, 2.57; 95% CI, 1.66, 3.98), major depression (OR, 2.27; 95% CI, 1.34, 3.86), and nicotine dependence (OR, 2.71; 95% CI, 2.08, 3.52), but was not statistically significantly associated with alcohol abuse and/or dependence or any illicit drug abuse or dependence. There was no statistically significant relationship between race and any of these five psychiatric disorders. The association of greatest magnitude was for the likelihood of any illicit drug abuse and/or dependence among those with low (vs. high) income (OR, 4.12; 95% CI, 2.19, 7.74). After adjusting for age, subjects who had an annual income of $20,000 or less were 4.2 times more likely to meet lifetime criteria for any illicit drug abuse and/or dependence.

Discussion

This study demonstrates that a large proportion of middleaged American males who served in the military during the Vietnam Era have been affected by at least one psychiatric disorder in their lifetime. Of 19 psychiatric disorders assessed (i.e., excluding nicotine dependence), 61.1% of respondents had a lifetime history of at least one disorder and 23.4% met the criteria for at least one disorder in the 12 months before the interview. Even after excluding alcohol abuse and/or dependence, about 1 in 4 had a lifetime history and about 1 in 10 had a 12-month history of at least one psychiatric disorder.

The strengths of this study include the large sample size, the national distribution ol respondents, the selection of the sample without regard to psychiatric status, the standardized data gathering methodology, and the assessment of psychiatric disorders according to DSM-III-R criteria. Our conclusions are likely to be applicable to all nonclinical samples of Vietnam Era male veterans, because twin studies of psychiatric illness are generalizable to nontwin populations.19 The primary limitation of the study results from the potential problem of nonresponse bias. Of the 10,300 members of the VET Registry who were targeted for the 1992 study, 8,169 were interviewed. The remaining 2,136 VET Registry members (20.7%) did not participate because they refused (n = 1,990), were dead (n = 43), or administrative problems prevented us from locating or contacting them (n = 98). To address the nonresponse bias issue, responders and nonresponders to the 1992 survey were compared on the following variables: race and education at entry into military service, score on the Armed Forces Qualification Test (AFQT, a test of aptitude and intelligence), and deployment to Vietnam. A greater proportion of Caucasians than non-Caucasians responded (80.1% vs. 70.2%, respectively, p

Because individuals were required to pass a battery of psychological assessments before induction, individuals with severe disorders, such as schizophrenia, were probably excluded from military service. This likely resulted in a mentally healthier sample as compared with civilian populations.21,22 Lastly, our results are limited to middle-aged men and may not generalize to women or to older or younger age cohorts.

Prevalence

Because of differences in sample selection and data collection methodology, caution must be used when comparing the present study to the NWRS, VES, EGA, and NCS studies. For example, the NWRS included Vietnam veterans regardless of service branch, enlistment status, or number of tours of duty; assessed psychiatric disorders according to DSM-III criteria (with the exception of PTSD); conducted face-to-face interviews; and defined "current" as the previous 6 months. The diagnosis of PTSD by the NWRS was constructed from responses to the Mississippi Combat-Related PTSD Scale, the PTSD component of the Structured Clinical Interview for DSM-III,20 and the PTSD Scale of the Minnesota Multiphasic Personality Inventory.10,11 The VES limited participants to those who had at least 16 weeks of active service but only one tour of duty, earned a military occupational specialty other than trainee or duty soldier, and had a pay grade no higher than E-5 (sergeant) at discharge; used DSM-III criteria, conducted face-to-face interviews, and defined "current" as the previous month.8 The EGA sampled the U.S. general population ages 18 years and older and used face-to-face interviews to assess prevalence of 1-month, 6-month, 1-year, and lifetime DSM-III psychiatric disorders.5,23 The NCS sampled the noninstitutionalized civilian population ages 15 to 54, and used the UM-CIDI in face-to-face interviews to assess the 12-month and lifetime prevalence of DSM-III-R psychiatric disorders. In the present study, although veterans were eligible regardless of the number of tours of duty, an honorable discharge was required; we used DSM-III-R criteria for all diagnoses, conducted telephone interviews, and defined "current" as the previous 12 months.

The lifetime and current prevalence of alcohol abuse and/or dependence found by our study and the other studies of Vietnam Era veterans is generally higher than that reported for civilian male populations. We found that 55% of our respondents had a lifetime prevalence of alcohol abuse and/or dependence and approximately 17% had a 12-month history; the NWRS reported that 39.2% of veterans had a lifetime prevalence of alcohol abuse and/or dependence and 11.2% had a 6-month history;9 and the VES reported lifetime alcohol abuse and/or dependence estimates of 50.6% for theater veterans and 41.8% for nonveterans and 1-month prevalences of 13.7% for theater and 9.2% for nontheater veterans. By contrast, the civilian (control) component of the NVVRS reported lifetime and 6-month prevalences of alcohol abuse and/or dependence of 25.2% and 7.0%, respectively;9 the male respondents in the NCS between ages 15 and 54 reported lifetime and 12-month prevalences of alcohol abuse and/or dependence of 32.6% and 14.1%, respectively;2 and the ECA generated lifetime and 12-month prevalences of alcohol abuse and/or dependence for men in the general population of 13.8% and 6.8%, respectively.5 These results strongly suggest that alcohol abuse and/or dependence is a serious problem among Vietnam Era veterans and is more common among Vietnam Era veterans than nonveterans.

The present study also suggests that the prevalence of PTSD is higher among Vietnam Era veterans than nonveterans. Ten percent of VET Registry members had a lifetime history of PTSD, and almost 5% had a 12-month history of this disorder. Although it is difficult to compare our study findings to published results from other veteran populations because of differences in assessment methodology, the NWRS reported a 30.9% lifetime prevalence of PTSD for male theater veterans10 and a 6-month prevalence of 15.2%,11 and the VES reported a lifetime prevalence of PTSD of 14.7%.8 In contrast, only 6% of male NCS respondents ages 15 to 54 satisfied criteria for having had PTSD at some time in their lifetime.24 The higher prevalence of lifetime and current PTSD among Vietnam veterans compared with the general population is not surprising. In addition, our data suggest that the lifetime prevalence of PTSD is higher among Vietnam theater veterans (15.0%) than non-theater veterans (6.1%).

Approximately 10% of VET Registry members had a lifetime history of illicit drug abuse and/or dependence and 1,3% satisfied the 12-month criteria. This pattern is similar to prevalences reported by the NWRS (5.7% lifetime for theater and 7.0% for nontheater; 1.8% 6-month for theater and 1.5% for nontheater)14 and VES studies (approximately 14% lifetime, 0.4% current). The ECA reported 6.1% lifetime and 2.0% 6-month prevalence of any drug abuse and/or dependence. The lifetime and 12-month prevalences reported by the NCS for its male respondents ages 15 to 54 were 15% and 5%, respectively.

Approximately 11% of VET Registry members had a lifetime history of an affective disorder and 3.5% had a 12-month history. These results are similar to the civilian cohort in the NCS study2 in which 14.7% of 15- to 54-year-old males had a lifetime history and 8.5% had a 12-month history. The lifetime and 12-month prevalences of generalized anxiety disorder we observed, 1.5% and 1.2%, respectively, are less than the 3.6% and 2.0% reported for male respondents ages 15 to 54 by the NCS study.2,25 This lower prevalence of generalized anxiety disorder among Vietnam veterans is unexplained and has not been reported by other studies of Vietnam veterans.

Comorbidity

Although some studies have indicated that Vietnam veterans have higher prevalences of comorbid psychiatric disorders by comparison to the general population,26 our results do not support this. We observed that among Vietnam veterans who had at least one lifetime psychiatric disorder, 38% had at least a second lifetime disorder other than nicotine dependence (i.e., 19.9% + 18.4% in Table II), and among those with a current disorder, more than 20% had a second current disorder (13.6% + 6.9% in Table II). This comorbid prevalence contrasts with the EGA study, which found that 60% of those with at least one lifetime psychiatric disorder had a second lifetime disorder,5 and the NCS, which found that 79% of respondents with one lifetime psychiatric disorder had a second lifetime disorder.2

Differences in assessments and sample populations might explain the observed lower prevalence of comorbid disorders among VET Registry members as compared with the ECA and NCS studies. First, disorders that were commonly comorbid in the EGA study, such as somatization disorder, schizophrenia, agoraphobia, and obsessive-compulsive disorder, were not assessed in the present study. Had we assessed these disorders, the prevalence of psychiatric comorbidity may have been closer to estimates from the ECA. Our assessments were limited to men while the EGA and NCS studies included women and men. Women are more likely than men to have more than one psychiatric disorder during their lifetime.2 The ECA and NCS also included men and women as young as 18 years of age. These younger subjects are more likely to report more than one psychiatric disorder during their lifetime.2 In addition, the health requirements for military service would have excluded those subjects with severe comorbidity (e.g., those with comorbid severe depression and chronic alcoholism). Lastly, comorbidity in this community sample of veterans is expected to be lower than that found in more severely affected clinical veteran populations.

Sociodemographic Associations

Marital status, employment, and income were most strongly and consistently associated with psychopathology. Those who were widowed, separated, or divorced, unemployed, or reported lower annual incomes had higher rates in all categories of lifetime and current disorders than those who were married or employed or had higher incomes. Our results are consistent with the ECA study that found the rates of lifetime psychiatric disorders were high among subjects who had been separated or divorced, subjects who were not working full-time, and subjects who were working in unskilled jobs.5

Conclusions

The high lifetime and current prevalence of psychiatric disorders among this nationally distributed, nonclinical sample of male Vietnam Era veterans reinforces the major public health challenge of mental illness among veterans. More effective methods must be developed to prevent, identify, and treat psychiatric illness, especially alcohol abuse and dependence and FTSD in Vietnam Era veterans.

Acknowledgments

The U.S. Department of Veteran Affairs (DVA) has provided financial support for the development and maintenance of the VET Registry. Numerous organizations have provided invaluable assistance in the conduct of this study, including the Department of Defense; National Personnel Records Center, National Archives and Records Administration: the Internal Revenue Service; National Opinion Research Center; National Research Council, National Academy of Sciences; and Institute for Survey Research, Temple University. Most importantly, the authors gratefully acknowledge the continued cooperation and participation of the members of the VET Registry and their families. Without their contribution this research would not have been possible.

This study was supported by National Institute of Drug Abuse Grant R01-DA04604.

References

1. Dohrcnwend BP: A psychosocial perspective on the past and future of psychiatric epidemiology. AmJ Epidemiol 1998; 147: 222-31.

2. Kassier RC, McGonagle KA, Zhao S, el al: Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in (he United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 1994; 51: 8-19.

3. Lee CK, Kwak YS, Yamamoto J, el al: Psychiatric epidemiology in Korea. Part II: urban and rural differences. J Nerv Ment Dis 1990; 178: 247-52.

4. Pickles A: Psychiatric epidemiology. Stat Methods Med Res 1998; 7: 235-51.

5. Robins LN, Regier DA (ed): Psychiatric Disorders in America: The Epidemiologie Catchment Area Study. New York, The Free Press, 1991.

6. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Ed 3. Washington DC, American Psychiatric Association, 1980.

7. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Revised Ed 3. Washington DC, American Psychiatric Association, 1987.

8. Centers for Disease Control Vietnam experience study: health status of Vietnam veterans. I. Psychosocial characteristics. JAMA 1988:259: 2701-17.

9. Jordan BK, Schlenger WE, Hough R, et al: Lifetime and current prevalence of specific psychiatric disorders among Vietnam veterans and controls. Arch Gen Psychiatry 1991; 48: 207-15.

10. Kulka RA, Schlenger WE, Fairbank JA, et al: Trauma and the Vietnam War Generation: Report of findings from the National Vietnam Veterans Readjustment Study. Phychosoeial Stress Series No. 18. Edited by Figley CR. New York, Brunner/Mazel Publishers, 199Oa.

11. Kulka RA, Schlenger WE, Fairbank JA, et al: The National Vietnam Veterans Readjustment Study: Tables of Findings and Technical Appendices. Psychosocial Stress Series No. 20. Edited by Figley CR. New York, Brunner/Mazel, Publishers, 199Ob.

12. Robins LN, Heizer JE, Croughan J, et al: National Institute of Mental Health Diagnostic Interview Schedule: ils history, characteristics and validity. Arch Gen Psychiatry 1981; 38: 381-9.

13. Eisen S, True W, Goldberg J, et al: The Vietnam Era Twin (VET) Registry: method of construction. Acta Genet Med Gemellol, 1987; 36: 61-6.

14. Eisen S, Neunian R, Goldberg J, et al: Determining zygosity in the Vietnam Era Twin Registry: an approach using questionnaires. Clin Genet 1989; 35: 423-34.

15. Henderson WG, Eisen S, Goldberg J, et al: The Vietnam Era Twin Registry: a resource for medical research. Public Health Rep 1990; 105: 368-73.

16. Robins LN, Heizer JE, Cottier L, et al: National Institute of Mental Health Diagnostic Interview Schedule Version III Revised, 1989.

17. Fleiss JL: Statistical Methods for Rates and Proportions, Ed 2. New York, Wiley, 1981.

18. Stata Corporation: Stata Statistical Software: Release 5.0. College Station TX, Stata Corp, 1997.

19. Kendler KS: Twin studies of psychiatric illness: current status and future directions. Arch Gen Psychiatry 1993; 50: 905-15.

20. Spitzer R, Williams J: Structured Clinical Interview for DSM-III. Non-patient version (SCID-NP-11-1-86). New York, Biometrics Research Department, 1986.

21. Kang HK, Bullman TA: Mortality among U.S. veterans of the Persian Gulf War. N Engl J Med 1996; 335: 1498-1504.

22. Rothberg JM, Bartone PT, Holloway UC, et al: Life and death in the U.S. Army: in corpore sano. JAMA 1990; 264: 2241.

23. Regier D, Burke J Jr: Comorbidity of affective and anxiety disorders in the NIMH Epidemiologie Catchment Area Program. In: Comorbidity of Mood and Anxiety Disorders, pp 113-22. Edited by Maser J, Cloninger C. Washington DC, American Psychiatric Press, 1990.

24. Kessler RC, Sonnega A, Bromet E, et al: Posllraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry 1995; 52: 1048-60.

25. Wittchen HU, Zhao S, Kessler RC, et al; DSM-III-R generalized anxiety disorder in the National Comorbidity Survey. Arch Gen Psychiatry 1994; 51: 355-64.

26. U.S. Department of Health and Human Services: Health status of Vietnam veterans. Psychological and Neurological Evaluation, Vietnam Experience Study, VoI IV. Atlanta. GA, Centers for Disease Control. 1989.

Guarantor: Seth A. Eisen, MD

Contributors: Seth A. Eisen, MD*[double dagger]; Kristin H. Griffith, MA[dagger]§; Hong Xian, PhD[dagger]§; Jeffrey F. Scherrer, MA[dagger]¶; Irene D. Fischer, MA[dagger]§; Sunanta Chantarujikapong, MD[dagger]§; Joyce Hunter, MSN[dagger]; William R. True, PhD[dagger]¶; Michael J. Lyons, PhD||#; MingT. Tsuang, MD**[dagger][dagger]

* Research and Medical Service and fResearch Service, Department of Veteran Affairs Medical Center, St. Louis, MO 63106.

[double dagger] Division of General Medical Sciences, Departments of Internal Medicine and Psychiatry, Washington University School of Medicine, St. Louis, MO 63110.

§ Department of Internal Medicine, Washington University School of Medicine, St. Louis, MO 63110.

¶ School of Public Health, Saint Louis University, St. Louis, MO 63104.

|| Department of Psychology, Boston University, Boston, MA 02215.

# Harvard Institute of Psychiatric Epidemiology and Genetics, Boston, MA 02115.

** Department of Psychiatry, Institute of Behavioral Genomics, University of California, San Diego, San Diego, CA.

[dagger][dagger] Departments of Epidemiology and Psychiatry, Harvard Institute of Psychiatric Epidemiology and Genetics, Boston, MA 02115.

This manuscript was received for review in September 2003. The revised manuscript was accepted for publication in December 2003.