| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Portland VA Medical Center (D.S., K.A.C.) and the Center for Research on Occupational and Environmental Toxicology (D.S., L.M., W.K.A., D.S.R., C.A.K.), Department of Psychiatry (K.A.C.), and Department of Neurology (L.M.B.), Oregon Health Sciences University, Portland, OR.
Address reprint requests to: Daniel Storzbach, PhD, Center for Research on Occupational and Environmental Toxicology, 3181 SW Sam Jackson Park Road, L606, Oregon Health Sciences University, Portland, OR 97201. Email: storzbach{at}home.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: An epidemiological survey focusing on exposures and symptoms was mailed to a random sample of Gulf War veterans from Oregon and southwestern Washington. Volunteers were recruited from survey respondents who agreed to undergo a thorough medical examination and psychological and neurobehavioral assessment. Persistent symptoms with no medical explanation associated with service in the Persian Gulf (eg, fatigue, muscle pain, and memory deficits) that began during or after the war qualified respondents as cases. The 241 veterans with unexplained symptoms were classified as case subjects, and the 113 veterans without symptoms were classified as control subjects. All veterans completed a battery of computerized assessment tests consisting of 12 psychosocial and 6 neurobehavioral tests. Differences between case and control subjects on neurobehavioral and psychological variables were assessed with univariate and multivariate statistical comparisons.
RESULTS: Case subjects differed substantially and consistently from control subjects on diverse psychological tests in the direction of increased distress and psychiatric symptoms. Case subjects had small but statistically significant deficits relative to control subjects on some neurobehavioral tests of memory, attention, and response speed. A logistic regression model consisting of four psychological variables but no neurobehavioral variables classified case and control subjects with 86% accuracy.
CONCLUSIONS: Our results revealed that Gulf War veterans who report symptoms associated with that conflict differed on multiple psychological measures in the direction of increased distress and performed more poorly on neurobehavioral measures when compared with control subjects who did not report symptoms. This suggests that psychological differences have a prominent role in investigation of possible explanations of Gulf War symptoms.
Key Words: psychological assessment • neurobehavioral assessment, • veterans • Persian Gulf War, • health symptoms • clinical evaluation.
Abbreviations: BAI = Beck Anxiety Inventory; BARS = Behavioral Assessment and Research System; BDI = Beck Depression Inventory; CES-R = Combat Exposure Scale, Operation Desert Storm Revision; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th edition; GW = Gulf War; HSS = Health Screening System; LES = Life Experiences Scale; Mississippi PTSD scale = Mississippi Scale for Combat-Related Posttraumatic Stress Disorder; MMPI-2 = Minnesota Multiphasic Personality Inventory-2; ODTP = Oregon Dual Task Procedure; PANAS = Positive and Negative Affect Schedule; PCL-C = Posttraumatic Stress Disorder Checklist–Civilian Version; Penn inventory = Penn Inventory for Posttraumatic Stress Disorder; PTSD = posttraumatic stress disorder; SASSI-2 = Substance Abuse Subtle Screening Inventory-2; SCIS = Subjective Cognitive Impairment Scale; SCL-90-R = Symptom Checklist-90-Revised; SF-36 = Health Status Questionnaire, short form.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The absence of clear causes of GW-associated symptoms has led researchers to propose several alternative explanations. Previous research findings on war-related illness have led some investigators to suggest that GW-associated symptoms result from psychological responses to war-related stress (4). Stress responses of this sort would be consistent with numerous research findings. Acute and chronic life stressors associated with increased somatic symptoms (5) have been documented in veterans of every military conflict since the US Civil War (2). In particular, PTSD (as defined in DSM-IV) is associated with increased risk of negative health outcomes (6, 7), although most studies have found lower rates of PTSD among GW veterans than among veterans of previous wars (8, 9). Research using either clinical patient samples (10, 11) or samples from specific military units (12–14) has found evidence of psychological distress among GW veterans with unexplained symptoms. The only population-based study of GW veterans published to date also found more symptoms of depression, PTSD, anxiety, and alcohol abuse among GW-deployed respondents than among non–GW-deployed control subjects (3). The data for this study, however, were obtained from telephone survey interviews and did not include clinical evaluation of participants. Interim results from our center (15) demonstrated that among the first 101 GW veterans studied of the 354 reported here, those with symptoms differed substantially from those without symptoms on a broad range of psychological tests indicative of increased distress.
Prominent alternatives to the stress-response hypothesis are various toxic exposure hypotheses. In the wake of recent reports of release of organophosphate nerve gas during the GW, it has been proposed that exposure to low doses of organophosphates could explain some of the symptoms reported by GW veterans (16). Possible exposure to other chemicals, including multiple potentially synergistic or enabling chemicals or environmental elements, have also been proposed (17). Although not conclusive, neurobehavioral deficits in this population would be consistent with an association between chemical exposure and GW symptoms. Findings of previous neurobehavioral assessments have been mixed, with some investigators reporting significant abnormalities (18, 19) and others reporting none (11, 14). Interim results from our laboratory (15) previously identified a small subgroup with neurobehavioral deficits among the first 101 participants of the final sample of 354 veterans described in this article.
The purpose of this article is to present findings from comprehensive psychological and neurobehavioral evaluations of 354 GW veterans drawn from a population-based sample to address the following questions: 1) Are there differences in psychological functioning between symptomatic veterans and those with no symptoms (control subjects)? 2) Are there differences in neurobehavioral functioning between symptomatic veterans and control subjects? 3) Among those measures for which there are significant differences between symptomatic veterans and control subjects, which are most strongly associated with GW symptoms?
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
A working case definition was developed before the mail survey was sent; the case definition was based on symptoms reported nationally from 1992 through 1995 by veterans of the GW (21–23) and specific review of the most common (>10%) symptoms reported by veterans in the Portland component of the Department of Veterans Affairs Persian Gulf War Registry. Potential case subjects for the case-control study were selected from all questionnaire respondents reporting at least one of the following symptoms: 1) musculoskeletal pain; 2) cognitive-psychological changes, including memory loss, confusion, inability to concentrate, mood swings, or somnolence; 3) gastrointestinal complaints; 4) skin or mucous membrane lesions; or 5) unexplained fatigue. Veterans reporting unexplained fatigue had to have four or more health symptoms related to chronic fatigue (24). Three requirements for the symptoms had to be met: 1) onset during or after deployment to the Persian Gulf, 2) symptoms persisting for 1 month or longer, and 3) occurrence of symptoms during the 3-month period preceding recruitment into the case-control study. Potential control subjects were selected from all questionnaire respondents who did not report any symptoms included in the case definition. Some potential control subjects reported symptoms not included in the case definition, such as sinusitis, allergies, and hypertension.
We received 1119 completed surveys for a response rate of 55% of all survey returns. Of these, 155 were ineligible for the case-control clinical evaluation because they were originally nonrespondents who subsequently agreed to limited participation in the study by completing the telephone interview (N = 116) or the shorter mail exposure/health questionnaire (N = 39). An additional 165 were ineligible because they refused further contact (N = 25), were a Vietnam veteran (N = 26), had symptoms predating the GW (N = 17), lived >100 miles from the test site (N = 37), had exclusionary diagnoses (N = 32), or for other reasons (N = 28). The 32 exclusionary diagnoses included epileptic seizures, malignancy, schizophrenia, hepatitis, infection with human immunodeficiency virus, malaria, diabetes, and other diseases. Of the remaining 799 survey responders, 69 could not be contacted by phone. Ultimately, 517 potential case subjects and 213 potential control subjects were contacted by telephone and asked to participate in the case-control clinical evaluation. It was not possible to assess the health status of persons who did not respond to the survey; however, nonrespondents did not differ from the respondents in age, race, gender, reserve or active status, or branch of service. Ten percent of the total sample had enrolled previously in the Portland Persian Gulf War Registry, and 13% of the survey respondents had enrolled in the registry, suggesting that nonrespondents may be healthier than respondents overall.
Potential case and control subjects were invited to participate in the 6- to 8-hour clinical evaluation, which included a physical examination with emphasis on the neurological and musculoskeletal systems, health history specific to the symptoms reported by the participant, blood and urine analyses, and computerized neurobehavioral and psychological tests. All examiners were initially blinded to the exposure history and the potential case or control status of participants. However, physicians of necessity became unblinded once informed of health histories specific to participants’ symptoms. Informed consent was obtained after the nature and possible consequences of participation were explained. Participants who completed the evaluation received $50.
After medical examinations and any diagnostic referrals were completed, a clinical case determination committee with representatives from the disciplines of neurology, rheumatology, internal medicine, neuropsychology, and epidemiology reviewed all potential case and control subjects. Potential cases with explainable diagnoses, exclusionary diagnoses or conditions, or who denied case definition symptoms at the time of clinical examination were excluded from the study. Fifty-four potential control subjects with nonexclusionary explained symptoms (eg, mechanical back pain) were retained as control subjects. Clinical test results of potential control subjects were reviewed for the possibility of unrecognized or unreported illness that would result in exclusion. We removed six potential control subjects who had exclusionary diagnoses not revealed on the questionnaire. One potential control subject was found to have unexplained musculoskeletal symptoms and became a case subject. We also screened potential control subjects for the previous presence of case symptoms, but none were identified. Two participants with very poor scores on the ODTP (<61% correct), a computerized variant of the Portland Digit Recognition Test of motivation to perform poorly on memory tests (25, 26), were also excluded. Three case subjects were determined to have severe cognition problems suggestive of probable dementia. These case subjects were included in all analyses of physical and laboratory findings (to be reported elsewhere), but we deemed it inappropriate to include these three cases because their cognition deficits rendered their self-reported data unreliable for case classification.
Participants
A total of 443 veterans were evaluated in our study, and 361 participants met our case or control criteria (241 case and 113 control subjects). The overall sample was 17% female, with 42 (17%) female case subjects and 17 (16%) female control subjects. Mean age at the time of assessment was 31.9 years (SD = 6.9 years) for case subjects and 34.3 years (SD = 9.3 years) for control subjects; the difference between groups (2.4 years) was statistically significant (p < .01). Mean years of education was 13.3 (SD = 1.7) for case subjects and 14.0 (SD = 2.2) for control subjects; this difference (0.7 years) was also statistically significant (p < .001). Mean score on the Armed Forces Qualifying Test, a self-administered test of mental aptitude taken by all military personnel at induction (27), was 60.6 (SD = 19.3) for case subjects and 66.1 (SD = 19.5) for control subjects, a trend that did not quite reach statistical significance (p < .07). Data from the Armed Forces Qualifying Test were available for only 64.7% of case subjects and 54.9% of control subjects.
Eighty-seven percent (212 of 244) of symptomatic veterans were found to have unexplained cognitive or psychological symptoms; 38%, unexplained musculoskeletal symptoms; and 42%, unexplained fatigue. Only 15% of symptomatic veterans were determined to have unexplained gastrointestinal complaints, and only 5% had nondiagnosable skin lesions (percentages do not add to 100% because >48% reported symptoms in two or more categories; see below). In only three cases were gastrointestinal or dermatological symptoms the only types of unexplained symptoms reported by the veteran. Based on this finding, we adjusted the original working case definition from the original five types of symptoms to only three types of symptoms (ie, cognitive/psychological, fatigue, and musculoskeletal). The specific complaints included in each symptom category are shown on Table 1.
|
Participants had served in all military branches and included veterans who were both on active or activated reserve status at the time of deployment. Only 14% of questionnaire respondents and 19% of clinical evaluation participants reported that they had previously participated in examinations offered by the Department of Veterans Affairs or Department of Defense as part of the GW registry. This is similar to the national rate of 15% of the 700,000 eligible veterans who were self-referred for registry examinations (4).
Psychological Testing
Psychological and neurobehavioral functioning were assessed with a 4-hour battery of 18 tests that included 12 psychological questionnaires and 6 neurobehavioral tests providing objective performance measures of memory or concentration loss, complex cognitive processing, and response speed. The psychological questionnaires were administered using the computerized HSS (28). Neurobehavioral tests were drawn from the BARS (29), a computer-implemented battery of measures sensitive to neurotoxic insult (30). Both the BARS and HSS test batteries have several features in common, including the DataSled response input unit and principles of test instruction and presentation (28, 29). These subject-friendly features include the DataSled’s simplified nine-key input and other aspects designed to simplify the test-taking process. The 12 HSS and 6 BARS tests administered in this study have previously been demonstrated to have adequate test-retest reliability and comparability with tests administered in their original formats (31).
The 6 BARS neurobehavioral tests were 1) Simple Reaction Time, a test of response speed measuring the latency between the appearance of a stimulus on the computer screen and a button press (32); 2) the Selective Attention Test, which assesses attention by presenting small dots on a variable interstimulus interval schedule inside or outside two squares and requiring a differential response ("3" or "7" button) for dots inside the respective squares and withholding of a response for dots outside the squares (29), scored as the percentage of correct responses; 3) Digit Span, a test of attention and memory (33) in which a series of numbers are presented sequentially on the computer screen and the participant is required to reproduce the sequence of numbers by typing them in the same sequence (forward) or, in the second part of the test, in reverse sequence (backward) (number sequences were presented in increasing length starting from 3 (forward) or 3 (backward) numbers, and the test terminated when the participant responded incorrectly on two consecutive trials with the same number of digits or when the participant completed correctly a span of nine); 4) Symbol Digit, a coding test in which 5-mm-tall digits in the top row of a two-row matrix are paired with symbols in the bottom row of the matrix and the participant is required to type the numbers that correspond with the respective symbols in the lower row (which is empty) of a second matrix (34); 5) Serial Digit Learning, a test of learning in which the same nine-digit series is repeatedly presented one number at a time until the participant, after the presentation is completed, correctly types the nine-digit sequence two times or until 12 trials are completed without two consecutive correct responses (35); and 6) ODTP, a new version of a test of motivation, attention, and memory (25) that involves presentation of a five-digit number followed by a vigilance task for 5- to 25-second intervals followed by a forced-choice between the original five-digit number and a different (incorrect) alternative.
The 12 HSS psychological tests were the 1) Mississippi PTSD scale, a 39-item scale reporting war-related military PTSD symptoms (36); 2) Penn inventory, a 26-item scale that assesses frequency of complaints in relation to stressful experiences (37); 3) PCL-C, a 17-item questionnaire tapping DSM-IV PTSD symptoms (38); 4) SF-36, a 36-item questionnaire that assesses functional (somatic) impairment and symptoms due to medical health problems (39, 40); 5) BAI, a 21 -item questionnaire that assesses subjective, somatic, and panic-related symptoms of anxiety (41); 6) BDI, a 21-item questionnaire that assesses the intensity of cognitive, affective, somatic, and performance-related symptoms with regard to depression (42); 7) SASSI-2, an 88-item questionnaire that samples a broad range of alcohol and drug abuse patterns, including subtle attributes of abuse and denial (43, 44); 8) SCL-90-R, a 90-item questionnaire that provides a direct assessment of psychological symptoms and a limited range of health disorders (45) and includes the SCIS (46), a subscale derived for this study from SCL-90-R items chosen by two board-certified neuropsychologists that assess subjective experience of cognitive impairment; 9) MMPI-2, a 370-item (basic form) questionnaire that assesses basic personality structure and function (47); 10) PANAS, a 20-item questionnaire that assesses the "emotional style" a person uses to cope with life/world events (48); 11) LES, a 57-item questionnaire that records life events within the past year and rates those events for positive and negative impact and perceived control of the events (49); 12) CES-R, a 50-item questionnaire assessing war-zone exposure to violence, wounding, wounding or death of others, threat of severe injury or death, leadership failures, abusive violence, and prisoner-of-war captivity developed in our laboratory by modifying previously published scales (50, 51).
These multiple tests sampling a broad range of psychological and neurobehavioral functioning were selected to provide comprehensive assessment of factors potentially relevant to unexplained illness, including symptoms of posttraumatic stress symptoms, functional health or mental health impairment, personality vulnerability (MMPI-2 and PANAS), psychosocial factors, specific disorder symptoms (eg, substance abuse, depression, and anxiety), and war zone experience. Because of the known relationships between combat and PTSD and research demonstrating long- and short-term health effects of chronic posttraumatic conditions (6), a major focus of our research proposal was posttraumatic stress symptoms. In keeping with recommendations that evaluators rely on more than one test, multiple measures of posttraumatic stress symptoms were selected: a measure of combat-related PTSD (Mississippi PTSD scale) comparable to that used in many prior studies of veteran populations; a measure of non–combat-related trauma (Penn inventory); and a brief measure of symptoms of posttraumatic stress that had been used to classify participants who met formal DSM-IV criteria for PTSD (PCL-C). Our investigation has found that these measures, although moderately highly correlated with one another, do seem to sample some divergent features of the presentation and etiology of posttraumatic stress symptoms.
Statistical Analysis
All variables were evaluated for normality and homogeneity of variance. For variables that violated normality assumptions (the majority), we compared performance between case and control subjects using the Mann-Whitney U statistic. For variables for which normality assumptions were ascertained, we compared performance between case and control subjects on all tests using two-tailed t tests. To assess the effects of the significant group differences for age and education, the influence of these variables were also evaluated with analysis of covariance for normally distributed variables and with multivariate logistic regression for variables with distributions that violated normality assumptions. Because the results of these supplementary covariance analyses were essentially identical to those of the univariate analyses, results reported are from the univariate Mann-Whitney and t test analyses. Bonferroni procedures were used to determine levels of statistical significance appropriate for the large number of comparisons. To assess the relative univariate strength of group differences, the effect size statistic g was computed for all variables according to the methodology of Hedges and Olkin (52). To determine the variables most predictive of caseness, logistic regression was performed using a forward stepwise procedure based on the likelihood ratio test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
0.80). As shown in Table 4, there were significant differences between case and control subjects on all SF-36 scales of self-rated physical and mental health and health-related function, with case subjects uniformly reporting themselves as less healthy and more negatively affected by their health status. On the CES-R, case subjects reported significantly greater combat exposure than control subjects. On the LES questionnaire of recent life events, case subjects reported more events within the previous year as having significantly greater negative impact and over which they felt less control (see Table 4). No significant differences were found for ratings of positive impact and positive control of life events.
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Case subjects exhibited deficient performance compared with control subjects on neurobehavioral performance tests. Differences on three measures, Simple Reaction Time, ODTP forced-choice correct, and ODTP forced-choice latency, were statistically significant after correction for multiple comparisons. These tests assess memory, attention, and response speed. Although mean differences between case and control subjects on neurobehavioral measures were consistently in the direction of more deficient performance in the case group, neurobehavioral effect sizes were uniformly small according to Cohen’s (54) classification system. Previous interim findings from about one-third of the present sample suggested that observed neurobehavioral differences were largely associated with a subgroup of about 20% of case subjects (15). We do not offer an explanation for these neurobehavioral differences based on our analyses, although others (16, 18, 19) have ascribed similar deficits to chemical exposures based on what we view as insufficient objective evidence of exposures.
A logistic regression model comprised of four psychological measures classified >86% of the study’s case and control subjects successfully. The four measures, MMPI-2 Hs scale, Mississippi PTSD scale, SCIS, and SF-36 general health scale, assessed somatic complaints associated with psychological distress, posttraumatic symptoms, subjective appraisal of cognitive impairment, and perceived ill health, respectively. Consistent with their small effect sizes calculated during univariate analyses, no neurobehavioral variables were entered into the equation. The high degree of accurate classification based on psychological tests highlights the relative prominence of the psychological measures, suggesting that they discriminate veterans with health symptoms from those without symptoms and should be retained as part of future batteries used to assess GW veterans.
Case subjects reported significantly greater combat experience than control subjects. They also reported greater impact and less control over postwar negative life events during the past year. Negative effects of war-related posttraumatic stress symptoms can result in compromise of adaptive coping, consequently increasing the likelihood of negative life events. Also, those experiencing greater levels of psychological distress are prone to view the world around them more negatively and are more sensitive in noticing and remembering negative life events. This may explain why one study reported the number of "combat-related" experiences recalled by some (but not all) GW veterans with diagnosed PTSD increased over time (59).
There does not yet exist a generally accepted definition of syndrome or disease for cases of GW unexplained illness (60). To date only classes or categories of symptoms have been presented without clear systematic relationships between them. Our case definition consists of symptoms that have been reported as frequent and unexplained in GW veterans (21–23). Given this definition, it is not surprising that case subjects represented themselves as generally less healthy than control subjects on all SF-36 measures of self-reported health and health-related functional impairment. Consistent with the presence of their unexplained symptoms, their mean scores on all subscales of this test rank them within the lowest 25% of self-reported functioning in comparison with other 25- to 44-year-old persons in a general US population (61). Because there is a significant prevalence of unexplained symptoms in general populations (62), it may be that our control subjects are more physically healthy than is representative of the overall population of veterans. There was some support for this assertion because control group scores on seven of eight SF-36 subscales were better than mean values reported for a general nonclinical normative sample of equivalent age (61) and because MMPI-2 scores of the control group were slightly lower than the means for the MMPI-2 standardization sample (47).
There were small but statistically significant differences in age and education between case and control subjects. However, the results of supplementary covariance analyses were essentially identical to those of the univariate analyses, suggesting that the influence of these demographic variables was minimal. Thus, although being older and better educated were statistically associated with a slightly lower risk for symptoms, the etiological and clinical importance of these demographic factors remains unclear. The difference between case and control subjects on the Armed Forces Qualifying Test, a mental aptitude test, did not reach statistical significance.
There was no evidence that motivation significantly influenced our results. The ODTP is a variant of the Portland Digit Recognition Test, a measure of motivation to perform poorly on memory tests (25, 26). Case subjects made significantly more errors than control subjects on the main measure of motivation, the forced-choice task. This difference between groups lacked clinical relevance because both case and control subjects performed at high levels of overall accuracy. Only two potential case subjects were excluded because of an ODTP score suggesting poor motivation. The mean number of correct responses for case subjects was 98%, compared with 99% for control subjects. These results suggest that both case and control subjects were well motivated on this measure and, by inference, other test battery measures.
In summary, case subjects differed significantly and consistently from control subjects on diverse psychological tests in the direction of increased distress. Four psychological measures, assessing somatization of psychological distress, posttraumatic symptoms, perceived cognitive impairment, and perceived physical health function, accurately classified most participants as case or control subjects. There were statistically significant differences on neurobehavioral performance tests of memory, attention, and response speed, but the effect sizes were small when compared with the large psychological differences. This suggests that the psychological differences will have a prominent role in investigations of possible explanations of GW symptoms. These results are also consistent with the suggestion that psychological approaches to symptom management may be helpful, a possibility that is currently under investigation (63). However, because the present data do not show whether stress was a cause of GW symptoms or a consequence of those symptoms, the explanation of GW symptoms remains to be determined.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication February 26, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. A. Swoboda Negotiating the diagnostic uncertainty of contested illnesses: physician practices and paradigms Health (London) , October 1, 2008; 12(4): 453 - 478. [Abstract] [PDF]
|
|
|
|
 |
|
 J. R. Rundell Somatoform-Spectrum Diagnoses Among Medically Evacuated "Operation Enduring Freedom" and "Operation Iraqi Freedom" Personnel Psychosomatics, April 1, 2007; 48(2): 149 - 153. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Spinhoven and M. Verschuur Predictors of Fatigue in Rescue Workers and Residents in the Aftermath of an Aviation Disaster: A Longitudinal Study Psychosom Med, July 1, 2006; 68(4): 605 - 612. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J Vasterling and J. D. Bremner The impact of the 1991 Gulf War on the mind and brain: findings from neuropsychological and neuroimaging research Phil Trans R Soc B, April 29, 2006; 361(1468): 593 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Wessely and L. Freedman Reflections on Gulf War illness Phil Trans R Soc B, April 29, 2006; 361(1468): 721 - 730. [Full Text] [PDF]
|
|
|
|
 |
|
 B. van den Berg, L. Grievink, J. Yzermans, and E. Lebret Medically Unexplained Physical Symptoms in the Aftermath of Disasters Epidemiol. Rev., July 1, 2005; 27(1): 92 - 106. [Full Text] [PDF]
|
|
|
|
|
|
N. Fiedler, N. Giardino, B. Natelson, J. E. Ottenweller, C. Weisel, P. Lioy, P. Lehrer, P. Ohman-Strickland, K. Kelly-McNeil, and H. Kipen Responses to Controlled Diesel Vapor Exposure Among Chemically Sensitive Gulf War Veterans Psychosom Med, July 1, 2004; 66(4): 588 - 598. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Gray, R. J. Reed, K. S. Kaiser, T. C. Smith, and V. M. Gastanaga Self-reported Symptoms and Medical Conditions among 11,868 Gulf War-era Veterans : The Seabee Health Study Am. J. Epidemiol., June 1, 2002; 155(11): 1033 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Abou-Donia, A. M. Dechkovskaia, L. B. Goldstein, S. L. Bullman, and W. A. Khan Sensorimotor Deficit and Cholinergic Changes following Coexposure with Pyridostigmine Bromide and Sarin in Rats Toxicol. Sci., March 1, 2002; 66(1): 148 - 158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Ford, K. A. Campbell, D. Storzbach, L. M. Binder, W. K. Anger, and D. S. Rohlman Posttraumatic Stress Symptomatology Is Associated With Unexplained Illness Attributed to Persian Gulf War Military Service Psychosom Med, September 1, 2001; 63(5): 842 - 849. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
