| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry (J.E.M., R.J.U., C.S.F., X.L.,), Uniformed Services University of the Health Sciences, Bethesda, Maryland; the Deployment Health Clinical Center (X.L.), Walter Reed Army Medical Center, Washington, DC; and the Department of Psychiatry (A.L.), Jefferson Medical College, Philadelphia, Pennsylvania.
Address reprint requests to: James E. McCarroll, Department of Psychiatry, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814-4799. Email: jmccarroll{at}usuhs.mil
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: We studied the pre-post responses of 352 military men and women who worked in the mortuary that received the dead from the Persian Gulf War (Operation Desert Storm) in 1990 to 1991. Symptoms of somatization were measured before and after exposure to the dead. The respondents were volunteers and nonvolunteers for assignment to the mortuary; some had prior experience in handling the dead and some did not. Four groups of participants were examined based on the degree of exposure to remains. Age, sex, volunteer status, prior experience handling remains, and preexposure measures of depression and mutilation fear were statistically controlled.
RESULTS: Postexposure somatic symptoms increased significantly over preexposure levels for the two groups with the most exposure to the dead.
CONCLUSIONS: These results provide additional evidence that exposure to the dead is related to somatic distress.
Key Words: military, • psychiatry, • somatization, • death, • volunteers, • war.
Abbreviations: PTSD = posttraumatic stress disorder;; BSI = Brief Symptom Inventory;; MQ = Mutilation Questionnaire.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
High levels of somatic complaints were found in Israeli combat veterans with combat stress reactions and PTSD (6). Israeli combat veterans with PTSD reported significantly more somatic symptoms than combat veterans without PTSD (controls), but did not differ from controls on their physical examination (7). In a population of U.S. Vietnam War veterans, PTSD symptom severity and somatization were significantly related to self-report of health problems, but only PTSD symptom severity was related to physician-rated health (8). Patients with PTSD in a health-screening clinic for U.S. Persian Gulf War veterans reported more combat exposure and a greater number of physical symptoms than nontreatment-seeking veterans on active duty (9). Among the veterans with PTSD, the most commonly reported symptoms were fatigue, nausea, muscle aches, dizziness, back pain, stomachache, and numbness. Somatic symptoms may have a particular relationship to the specific stressor or to exposure to death and the dead. Increased reports of somatic symptoms in mortuary workers may persist for months (2). Reports of traumatic events in Gulf War veterans were associated with both PTSD and somatoform diagnoses (10). Veterans who handled dead bodies had a three-fold increased risk of receiving a somatoform diagnosis.
Physical symptoms seem to be common after many severe or recurrent traumatic stressors (11). A prospective study of victims of a natural disaster in Puerto Rico found that exposure to the disaster was related to a higher prevalence of medically unexplained physical symptoms, particularly gastrointestinal and pseudoneurological ones (12). In a prospective community prevalence study of the relationship between persons with a history of PTSD and the development of somatic symptoms, a history of PTSD was associated with significantly more somatic symptoms (13). Persons with PTSD were also more likely to report somatic symptoms compared with those persons with other psychiatric disorders. Additionally, the odds ratios for pain and conversion symptoms were increased in persons with a history of PTSD relative to those persons with no psychiatric disorder, and PTSD increased the risk of somatic symptoms beyond that expected by the presence of comorbid psychiatric disorders.
When the military forces of the United States and its allies began deployment to the Persian Gulf for Operations Desert Shield and Desert Storm, large numbers of casualties were expected in the initial operations of the war. Some of this expectation was based on the estimate of Iraqi possession and possible use of chemical and biological weapons on the battlefield. As a result, the U.S. military operated a mortuary between January and March 1991 at Dover Air Force Base on the East Coast of the United States to receive American war casualties of all services. A total of 383 were officially reported to have been killed in Operations Desert Shield and Desert Storm.
Military personnel from all four U.S. military services (U.S. Army, Air Force, Navy, and Marine Corps) volunteered or were assigned to the Dover mortuary for the period of the war. The mortuary staff was required to identify the dead and prepare them for shipment to their home and subsequent burial. Some service members were volunteers for mortuary duty whereas others had simply been assigned; some had previous experience in handling the dead whereas others had no such experience. As part of a research and consultation team, we were invited to collect research data that might be useful in understanding the stress of exposure to the dead. We collected measures of symptom levels of those assigned to the mortuary both before and after exposure to the dead and after the conclusion of mortuary activities, and categorized the degree of exposure of each individual to the dead.
Previous studies of these same mortuary personnel before the arrival of human remains (14, 15) focused on the stress of anticipation of exposure to the remains. Persons who anticipate working in a mortuary have higher levels of distress symptoms even before exposure compared to workers who do not anticipate working with the dead. In addition, nonvolunteers, females, inexperienced mortuary workers, and those with high mutilation fear reported more distress.
Previous findings on the relationship between psychological stress and the development of somatic symptoms after exposure to human remains are limited in terms of the small number of studies, the populations available for study, and the time course for which participants can be followed. We wished to determine if there were pre-post differences in somatization scores among those who were exposed to the dead in the mortuary. Given that we could collect preexposure data and had a relatively large sample, we were able to control for many relevant co-variates including the level of symptoms before exposure to the dead.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A total of 358 people, 300 men and 58 women, participated in the study. The mean age of all participants was 31.8 years (N = 355, SD = 8.4). The mean age for men was (N = 31.8, SD = 8.6); for women, it was 31.6 (N = 57, SD = 7.2). The educational level was high: 26% were high school graduates, 55% had attended college, 12% were college graduates, and over 6% had a master’s or doctoral degree. Forty-two percent were married, 30% were single, 14% were divorced and remarried or divorced and living with someone, and 13% were separated or divorced. Sixty-seven percent were white, 22% were black, 6% were Hispanic, 3% were "other race," and 1% were Asian.
Participants were asked to indicate if they had worked with dead bodies before coming to the mortuary. Those who responded positively were called "experienced" (N = 128), whereas those who responded negatively were called "inexperienced" (N = 228). The distinction between volunteers (N = 220) and nonvolunteers (N = 136) was based on whether a person had specifically volunteered for this assignment. Although they may not have volunteered for duty at the mortuary, all participants were volunteers for the research. After complete description of the study to the subjects, written informed consent was obtained.
A variable of exposure to bodies was based on responses to the postquestionnaire. Four independent exposure categories were created and ordered from the most exposure to no exposure to the remains. The first group (N = 68), body handlers, were persons who assisted with the processing of the bodies through the identification phase and who reported that they had personally handled one or more bodies. The second group (N = 155), exposed by duty, had been assigned to a duty which placed them in the mortuary (such as x-ray or handling personal effects) where it was almost certain that they had seen a large number of bodies, but who reported that they had personally handled none. The third group (N = 90), observed bodies, were assigned to a duty outside the mortuary, but reported that they had observed one or more bodies being removed from the transfer cases at the entrance to the mortuary. The third group was thought to be different from the second group in that the quantity of their exposure was less than that of group two because their full-time job was outside the mortuary. The last group (N = 40), unexposed, was made up of persons who were assigned to the mortuary, but not assigned to mortuary duty, did not report handling any bodies, and did not report seeing any bodies removed from transfer cases. The median number of days between administration of the premortuary and postmortuary duty questionnaires was 23.5 days.
There was a statistically significant difference in the sex distribution of the four exposure groups, 
2 = 10.69, df = 3, p = .014 (Table 1). Men were over-represented in the body handler group (95%) and under-represented in the unexposed group (70%) compared with the overall percentage of men in the study, 84%. Women were under-represented in the body handler group (5%) and over-represented in the unexposed group (30%) compared with the overall percentage of women (16%).
|
2 = 28.99, df = 3, p = .0001 (Table 1). There were more volunteers (79%) than nonvolunteers (21%) in the body handler group and in the exposed by duty group (67% volunteers, 33% nonvolunteers. The differences in prior experience in handling bodies were not statistically different among the four groups (Table 1).
Instruments
Somatic symptoms.
Somatic symptoms were measured with the BSI (16), a 53-item self-report inventory, a shorter version of the SCL-90-R (17). The BSI consists of nine clinical scales measuring primary symptom dimensions and three global indices of distress. Each item is scored on a 0 to 4 Likert scale from "not at all" to "extremely." We used the somatization subscale to measure psychological distress because of past findings of its association with posttraumatic symptoms. Participants were asked to respond according to the way they felt during the past week, including that day. The somatization scale contains seven items that measure somatic complaints: faintness, pains in the heart or chest, nausea or upset stomach, trouble getting breath, hot or cold spells, numbness or tingling, and feeling weak.
Mutilation questionnaire.
The MQ consists of 30 items in a true-false response format (18). The intent of the scale is to measure fear associated with gruesome or dangerous tasks or events where bodily injury is possible. Representative items are "using sharp knives makes me nervous" and "medical odors make me tense and uncomfortable." The MQ showed good internal consistency across several samples (18). In a study of self-reports of fainting and fear of blood and injuries, Kleinknecht (19) found that women reported more blood and injury fear than men and that the MQ was an independent predictor of fainting. Kleinknecht and Thorndike (20) used the MQ to assess the verbal-cognitive component of blood/injury fear involving revulsion to blood and injury stimuli and fear of bodily damage. Blood phobics who were high on the MQ reported greater empathic distress, fear and insecurity, and nightmares than nonphobics.
Analyses
The hypothesis tested was whether somatic symptoms increased from preexposure levels after exposure to the dead. We predicted that those who were exposed to remains would have greater somatization symptoms. The SAS Mixed Model (Proc Mixed) with repeated measures was used to compute statistical significance of the pre-post differences (21). It is a linear model procedure based on maximum likelihood estimate solutions to determine statistical significance. The mixed model was chosen because it uses both fixed and random effects and in the estimation process it accounts for the correlation between the scores at time 1 and time 2. The fixed effect factors for this model were exposure group, volunteer status, experience, sex, age, depression, and the MQ score. The random effects were subjects because the inference is to be made to the entire populations of subjects who could have been in this study. The unit of analysis was time. Missing data do not cause serious problems for this model in the same way as in other repeated measures analysis of variance designs because the basic theory on which the mixed model is based holds even with unbalanced and missing data. The mixed model produces least squares means, adjusting the means for the effects of the controlled variables. Sex was controlled due to the unequal distribution of men and women in the exposure categories; depression was controlled due its relation to posttraumatic and somatic symptoms.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
2 = 105.77, df = 8, p < .0001) (Figure 1). When the individual exposure groups were compared on somatization symptoms, there were significant pre-post differences between somatization symptoms before and after exposure for the body handlers and the exposed by duty groups, but not for the group that observed bodies or the unexposed group (Table 2). The greatest increase was for the body handlers followed by the group that was exposed by duty. Preexposure mortuary somatization symptom levels were significantly associated with postexposure somatization symptoms (r = 0.26, p < .0001).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The meaning of the increased somatic symptoms immediately after working in the mortuary is not clear. We could not determine the long-term time course of these symptoms because this study was not able to follow participants beyond their termination of duty at the Dover mortuary. Among the possibilities are that some of the participants were actually sick at the end of the period of time. However, we were not aware of any outbreaks of illness and it is unlikely that two separate groups that worked in different areas would be sick whereas other participants working in the same area were not. A second possible explanation is that the somatic symptoms were a result of fatigue (or injury) due to the long hours of work, but it is also unlikely that the groups were different in this respect. In addition, the items of the somatization scale do not seem to be related to fatigue or injury. Why exposure to traumatic death may increase the reporting of somatic symptoms is not known. This may represent an increased awareness of one’s own body or concern about usually dismissed physical symptoms or the somatic consequences of psychological stress. Somatic symptoms may also be the language of distress in some subgroups of those who are exposed to trauma.
There are several clinical implications to these findings. Those working with individuals exposed to death and the dead should be alert to the fact that symptoms are present in anticipation of exposure, before any actual exposure. Therefore, individuals who report somatic symptoms even before exposure may be at increased risk after exposure. Although a physical basis of any somatic complaint must be carefully examined, it is important to keep in mind that those persons with somatic symptoms that seem to be psychophysiological may be persons exposed to stressful events. Finally, symptoms that seem early after exposure to death and the dead should be followed up by concerned persons, including the individual, to assess the time course of the return to baseline health and to facilitate the early detection of any continuing distress, illness, or disability.
There are several limitations to this study. The population of mortuary workers sampled is healthy and educated by virtue of the screening that occurs before and during military service. Reports of symptoms are not clinician-based, but are self-reports. As a result, it may be easier to detect symptoms in a population with a minimal level of symptoms whereas a community population of mortuary workers may have a higher level of baseline (background) symptoms. Future studies should examine and compare possible differences in these two types of reports and populations.
Received for publication November 14, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Z. Solomon, R. Berger, and K. Ginzburg Resilience of Israeli Body Handlers: Implications of Repressive Coping Style Traumatology, December 1, 2007; 13(4): 64 - 74. [Abstract] [PDF]
|
|
|
|
 |
|
 C. S. Fullerton, R. J. Ursano, and L. Wang Acute Stress Disorder, Posttraumatic Stress Disorder, and Depression in Disaster or Rescue Workers Am J Psychiatry, August 1, 2004; 161(8): 1370 - 1376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Reissman, E. A. S. Whitney, T. H. Taylor Jr, J. A. Hayslett, P. M. Dull, I. Arias, D. A. Ashford, E. A. Bresnitz, C. Tan, N. Rosenstein, et al. One-Year Health Assessment of Adult Survivors of Bacillus anthracis Infection JAMA, April 28, 2004; 291(16): 1994 - 1998. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
