Q J Med 2000; 93: 283-289
© 2000 Association of Physicians
Fatigue in Cambodia veterans
From the Departments of Medical Psychology, and 1 Internal Medicine, University Hospital Nijmegen, The Netherlands
Received 21 December 1999 and in revised form 14 March 2000
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
In 1992 and 1993, Dutch military personnel were deployed in the peace operation UNTAC in Cambodia. Since returning, Cambodia veterans have reported health complaints which they perceive to be related to their service. Their symptoms strikingly resemble health problems reported by Gulf War veterans. Four years post-return, a cross-sectional survey on health symptoms in Cambodia veterans was initiated. Questionnaires were sent to all Cambodia veterans and four comparison groups. Forgetfulness, difficulty concentrating and fatigue were the symptoms most commonly endorsed. An operational case definition was constructed using a validated fatigue severity questionnaire. Cases were not uniquely found in Cambodia veterans (17%). In Rwanda and Bosnia veterans, respectively, 28% and 11% also met our case definition. Fatigue severity level was predicted by pre-mission, during-mission and post-mission variables, of which retrospective recollection of side-effects of vaccines and causal attributions also have been shown to be relevant in studies on Gulf-related illness.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
In the last few years, there has been considerable public concern and medical interest in symptoms in veterans who have served in the Persian Gulf War, and veterans who have participated in other wars and peace operations. In the Netherlands, symptoms have been reported in (ex)-servicemen who were deployed in the peace operation United Nations Transitional Authority for Cambodia (UNTAC).
In 1992 and 1993, approximately 2616 Dutch servicemen were sent to Cambodia in three battalions, each for a 6-month period. UNTAC was established in March 1992 after the ending of the civil war in Cambodia that had lasted two decades. Over 30 different nationalities, including French, German and Australian military personnel, have participated. The major tasks of UNTAC were supervising the administration of the country, disarming local factions, guiding free elections, guarding human rights, repatriating Cambodian refugees from refugee camps in Thailand, and mine clearance. The majority of Dutch military personnel in Cambodia were blue helmets. They were stationed in the north-western province of Bantey Meanchey.
Since returning, Cambodia veterans have reported health problems, which they attribute to their service in Cambodia. Their complaints, as voiced in the Dutch media, are severe fatigue, cognitive problems and headaches, which they perceive to be related to the malaria chemoprophylaxis mefloquine (Lariam®) or the vaccines they were exposed to. According to symptomatic Cambodia veterans, symptoms had started in Cambodia or just after return.
On face of it, symptoms in Cambodia veterans strikingly resemble symptoms reported in military personnel with alleged Gulf War syndrome: the same main symptoms are reported, causal attributions to chemical factors also have been made, and the topic was discussed in the media.1,2 Recent studies on Gulf War syndrome were not able to identify a unique Gulf-related illness, or to prove dose-effect relationships between exposures and physical or psychological symptoms.3–5 However, the nature and aetiology of Gulf War syndrome remain controversial.
In 1997, we initiated a comprehensive, independent study on symptoms in Cambodia veterans. To our knowledge, no scientific studies have been conducted on Cambodia veterans in other countries. This article reports the first phase of our study, in which we sent a cross-sectional postal survey to all Cambodia veterans whose particulars were known to the Ministry of Defence. The findings of our study will be compared with recently published findings relating to Gulf War syndrome. We investigated the prevalence of symptoms in Cambodia veterans, whether these symptoms were specific to Cambodia veterans, and what factors predicted symptoms in these Cambodia veterans.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
According to the latest figures of the Dutch Ministry of Defence, an entire sample of 2616 servicemen were deployed in Cambodia. Data were available for 2548 Cambodia veterans. Sixty-eight names and addresses (2.6%) could not be retrieved from the databases of the Ministry of Defence, probably because these files had been deleted after servicemen had left service. Thus, the subjects were 2548 Cambodia veterans whose addresses were known to the Ministry of Defence. To control for symptoms in soldiers who have been deployed elsewhere in tropical and non-tropical countries, soldiers who have not taken mefloquine but were administered vaccines, and soldiers who have never been deployed, four comparison groups were included: 120 (ex)-servicemen who were deployed in UNAMIR (United Nations Assistance Mission for Rwanda: 1994–1996; Rwanda group), 203 servicemen who were deployed in UNPROFOR (United Nations Protection Force in former Yugoslavia: 1993–1995; Bosnia group), 200 military personnel who were on stand-by for UNAMIR in 1994 (notice group), and 187 never-deployed marines (no deployment group).
Material
The questionnaire collected information regarding:
Fatigue severity, using the Checklist Individual Strength (CIS).6,7 The CIS measures different aspects of fatigue (fatigue severity, concentration problems, and physical activity level) over the last 14 days. Psychometric properties are excellent. Scores on the 8-item fatigue severity scale range from 8 to 56, higher scores indicate more fatigue and related problems. The CIS has been used in all studies of the Netherlands Fatigue Research Group, so data from different comparison groups as well as healthy controls are available.
Self-reported symptoms, using a questionnaire determining the presence of 33 symptoms which were selected on the basis of literature on Gulf War syndrome, chronic fatigue syndrome, side-effects of mefloquine and vaccinations, and tropical infectious diseases.
The period from notice until departure (pre-mission period) by including four questions on satisfaction with the preparation offered by the Ministry of Defence; attitude to the peace operation (3 items); length of time between notice and departure (1 item); side-effects of vaccines (1 item); and earlier experience in the tropics (1 item).
The time in Cambodia by including four questions on subjective experience of the peace operation; satisfaction with living conditions in Cambodia (3 items); problems at the home front (1 item); length of stay in Cambodia (1 item); repatriation (1 item); worries about one's health (1 item); and side-effects of mefloquine (1 item).
The period from return until the moment of study (post-mission period) by including a four-item self-efficacy questionnaire that measures perceived control over symptoms;5 a 6-item questionnaire measuring consequences of the peace operation on relationships;8 four questions measuring causal attributions (my complaints are related to mefloquine use; my complaints are related to traumatic events; my complaints are related to vaccines that were administered before departure; my complaints are related to disease caught in Cambodia); adaptation (5 items); problems at the home front (1 item); and satisfaction with military career (1 item).
The questionnaire was tested in a pilot study. Control groups received an adjusted version of the questionnaire to fit their personal situation.
Procedure
Given the independent nature of our study, we were bound by privacy regulations. The Ministry of Defence was not allowed to pass on their address database to the researchers, and vice versa. Thus, we did not possess the complete address database for all 2548 Cambodia veterans, and we were not allowed to inquire about the addresses of non-responders. Unfortunately, therefore, we were not able to investigate non-responders further.
The State Secretary for Defence announced our independent study in a letter directed to all Cambodia veterans. Two weeks later, the Ministry of Defence sent the questionnaires, including a postage-paid return envelope addressed directly to the research group. A reminder was sent another 3 weeks later. Respondents who returned the questionnaire and wrote down their particulars were recorded in an address database. All respondents were offered a gift voucher of 25 Dutch guilders, which they could donate to charity or which they could receive themselves. The ethics committee of our hospital approved the study.
Case definition
Based on news reports on Cambodia issues, literature on Gulf War syndrome and preliminary interviews with military personnel, fatigue, difficulty concentrating and forgetfulness were assumed to be frequent symptoms in Cambodia veterans. In the absence of a set case definition, the CIS-fatigue severity scale was used to produce an operational case definition. Respondents with CIS-fatigue severity scores 
37 were defined as cases. The cut-off score of 37 was derived from the mean score of a group of healthy controls used in another study by adding two standard deviations (mean 17.3, SD 10.1) (95%CI 14.5–20.2).9
Statistics
Data were entered twice in Dbase IV to check for errors in data entry. Data analysis was performed using SPSS (version 8.0). For reasons of homogeneity, only men were included. Univariate analyses were used to test differences between groups. The alpha level was set at p=0.05. We used exploratory factor analysis with varimax rotation to identify underlying factors in questions that collected information regarding the pre-mission period, the time in Cambodia, and the post-mission period. An eigenvalue >1.0 was chosen as the extraction criterion. Single factors were collected in a factor if Cronbach's alpha reliability coefficient was >0.60.
Prediction of outcome measures was done by linear regression analyses. The dependent variable was CIS-fatigue severity score. The independent variables were the factors and single items that related to the periods mentioned above. Factor and regression analyses could only be performed on Cambodia veterans who reported present symptoms or indicated that they had suffered symptoms during or after the peace operation, since the questions on perceived control over symptoms and causal attributions do not make sense to non-symptomatic respondents. This group of 672 respondents did not differ from the total group in age, marital status and service branch. The groups only differed in educational level, the former group being overall less educated (
2, df=1, p<0.01). Respondents with missing values were not included in the analyses.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
A completed questionnaire was returned by 1721 male Cambodia veterans (68%). In the Rwanda, Bosnia, notice and no-deployment groups, the numbers of male respondents were 58 (48%), 75 (37%), 77 (39%), and 91 (49%), respectively. Biographical data are shown in Table 1
. Cambodia veterans and comparison groups differed significantly in age (F(4,2003)=92.8, p<0.001), marital status (2, df=4, p<0.001), educational level (2, df=4, p<0.001), and rank (2, df=4, p<0.05).
|
Symptoms
Cambodia veterans reported a mean of 3.0 current symptoms (SD 3.9, 95%CI 2.8–3.2). As expected, the main symptoms reported were forgetfulness (35%), difficulty concentrating (28%) and fatigue (24%), followed by feeling unrefreshed by sleep (21%), flatulence (20%), joint pain (15%), sight problems (14%), headache (14%), tension or nervousness (13%) and rashes (11%). Table 2
shows the respective percentages in the comparison groups.
|
Between groups, significant differences were found on the total number of symptoms reported. Cambodia veterans reported more symptoms than the notice and no deployment group (F(4,2021)=9.8, p<0.001). Results remained significant after correction for age, rank and education. No significant differences were found between Cambodia, Rwanda and Bosnia veterans.
Cases and specificity
A total of 1716 Cambodia veterans completed the CIS-fatigue severity scale. Their mean score was 21.9 (SD 12.6). Elevated CIS-fatigue severity scores (37) were found in 288 respondents. Thus, 16.8% of Cambodia veterans were marked as cases (Table 3).
|
Between groups, significant differences were found on the CIS-fatigue severity scale (F(4,2012)=19.2, p<0.01). Cambodia veterans scored significantly higher than the notice and no deployment group and significantly lower than Rwanda veterans (Table 3
). Results remained significant after correction for age, rank and education. Respondents in the comparison groups also met the case definition: 27.3% of Rwanda veterans, 10.8% of Bosnia veterans, 3.9% of the notice group and 2.2% of the never deployed marines had a CIS-fatigue severity score 37 (Table 3). Thus, cases were not found in Cambodia veterans exclusively.
Factor analyses
For the pre-mission period, two factors (eigenvalue >1.0) were identified. Cronbach's alpha was >0.60 in only one factor, which we called ‘satisfaction with the preparation offered by the Ministry of Defence’ (4 items, % of variance=23, 
=0.64).
For the time in Cambodia, four factors (eigenvalue >1.0) were identified. Again, Cronbach's alpha was >0.60 in only one factor: ‘subjective experience of the peace operation’ (4 items, % of variance=24, =0.80).
In the post-mission period, four factors (eigenvalue >1.0) were identified. Cronbach's alpha reliability coefficient was >0.60 in two factors: ‘adaptation after return’ (4 items, % of variance=22, =0.61) and ‘causal attributions to medication’ (2 items, % of variance=14, =0.68).
Table 4 gives an overview of all factors and single items that were entered as independent variables in regression analysis.
|
Predictors of severe fatigue and fatigue related problems
For the pre-mission period, higher current CIS-fatigue severity scores were predicted by less satisfaction with the preparation offered by the Ministry of Defence and more experienced side-effects of vaccines (Table 5).
|
For the time in Cambodia, higher current CIS-fatigue severity scores were predicted by a greater impact of the mission, more problems at the home front and more worries about one's health (Table 6
).
|
For the post-mission period, higher current CIS-fatigue severity scores were predicted by lower self-efficacy, more problems in re-adjusted on return, less satisfaction with military career, stronger causal attributions to medication, more problems at the home front and a stronger causal attribution to disease caught in Cambodia (Table 7
). Taking the pre-mission, mission and post-mission predictors together, the same predictors come out as in the post-mission period.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Main conclusions
In the present study, the prevalence rate of health symptoms in Cambodia veterans was 17%. These symptoms were specific to Cambodia: in groups of Rwanda and Bosnia veterans, 27% and 10%, respectively, also met our case definition. Fatigue in Cambodia veterans was predicted by less satisfaction with the preparation offered by the Ministry of Defence in the pre-mission period and more side-effects of vaccines, as retrospectively recollected. Further, fatigue was predicted by a more poignant, threatening, aggravating and powerless subjective experience of the mission, more problems at the home front and more worries about one's health status during the mission. Finally, fatigue was predicted by less perceived control over symptoms, more adjustment problems after the mission, less satisfaction with military career, stronger causal attributions to mefloquine and vaccines, more problems at the home front after the mission and a stronger causal attribution to disease caught in Cambodia.
Limitations of the study
The present study has some limitations. Firstly, for reasons of privacy, data on non-responders are not available. Although the response rate among Cambodia veterans was quite high, information on non-responders would gain more insight into the representativeness of the results. Secondly, the sample sizes of the comparison groups are relatively small. Thirdly, and a major limitation of the study, is its retrospective nature. Since we relied on self-report of the Cambodia veterans, recall bias may have been induced, particularly in respondents with more health symptoms. Therefore, it can be questioned whether the factors found to be predictive of fatigue are initiating factors or mere attributions that contribute to the maintenance of the symptoms. Finally, no data are available on the pre-mission level of physical and psychological functioning of the Cambodia veterans. These data are considered important in establishing the onset of the symptoms. On basis of our survey, it cannot be ruled out that symptoms in some respondents already were present before the mission in Cambodia, or have started years after their return.
Comparison with Gulf War syndrome
Symptoms in Cambodia veterans strikingly resemble Gulf War syndrome, seen in military personnel who were deployed in the Persian Gulf War. In various studies on American, British and Danish Gulf War veterans, using different designs and methods of selecting subjects, the same main symptoms are reported as in our study, namely fatigue, difficulty concentrating and forgetfulness.3,10–14 However, it should be noted that the prevalences reported in US and British studies are higher than those in the Danish and our present study. This might be explained by the use of different case definitions. In the absence of a valid and agreed case definition, we used a validated fatigue severity scale to produce an operational case definition. A cut-off score of 37 was chosen. If we had used other cut-off scores that also lay beyond the scope of military personnel who have never been deployed or who have been on stand-by for UNAMIR, e.g. 33 (mean+2SD for the never-deployed group) and 26 (mean+2SD for the standby group), these symptom percentages would have been much higher, at 23% and 35%, respectively.
As in our study, Gulf War veterans report the same health problems as comparison groups, although at higher levels.3 They further experience more psychological distress, and they more often fulfil set case definitions such as CDC multisymptom syndrome criteria.3,5
In the present study, some of the Rwanda and Bosnia veterans also fulfilled our case-definition, in contrast with servicemen who have been on stand-by for Rwanda and never-deployed marines. However, Unwin et al. found that a comparison group of Bosnia veterans did not resemble Gulf War veterans, but rather resembled the non-deployed soldiers.3 A first explanation for these conflicting results may be the large difference in sample size between the British and Dutch Bosnia groups (2620 and 75, respectively). Further, it could be suggested that the performance of the Dutch in Srebrenica could account for this difference. However, our data did not show an association between symptom reporting and deployment in Srebrenica. Seventeen percent (n=13) of the Bosnia veterans had been deployed in Srebrenica during the fall of the enclave. Fatigue severity scores did not differ between Srebrenica veterans and veterans who were deployed elsewhere in Bosnia during the period 1993–1995 (t=0.5, p>0.05). Two Srebrenica and six other Bosnia veterans met our case definition.
In the present study, fatigue was predicted, among others, by more retrospectively recollected side-effects of vaccines, a stronger causal attribution to disease caught in Cambodia and a more poignant, threatening, aggravating and powerless subjective experience of the mission. Gulf War veterans also show a positive relationship between retrospective recollection of side-effects of vaccines and report of current symptoms.3 Further, causal attributions are also considered important in Gulf War syndrome. Several studies have shown a relationship between more self-reported exposures or stronger opinions on being exposed to noxious agents and more health problems or lower health perception.3,11
Veterans syndromes
In the literature, it is generally concluded that there is no new or discrete syndrome accounting for symptoms in Gulf War veterans.3,11,14,15 A common factor in symptoms in veterans is further suggested by the resemblance to other symptoms after acts of war and peace operations, and the inability to find relations between specific exposures and specific symptoms.1,2,14,16 In our opinion, findings on Gulf War syndrome are highly relevant to findings in Cambodia veterans and vice versa. Therefore we question specific names such as ‘Gulf War syndrome’, and advocate a broader nomenclature, namely ‘veterans syndromes’. However, more research is needed to clarify exceptions. Further, to address problems on recall bias, cause and effect, initiating and maintaining factors, and baseline levels of functioning, prospective longitudinal studies are needed. This is especially so since participation in humanitarian operations is increasing, and symptoms in veterans are a profound problem for military forces.
|
Acknowledgments |
|---|
We thank SCM van Esch, HM van der Ploeg and I Bramsen, Department of Medical Psychology, Free University of Amsterdam, for their collaboration in the first phase of the study. The Dutch Ministry of Defence has funded this research.
|
Notes |
|---|
Address correspondence to M. de Vries, University Hospital Nijmegen, Department of Medical Psychology, PO Box 9101, 6500 HB Nijmegen, The Netherlands. e-mail: M.deVries{at}cksmps.azn.nl
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Soetekouw PM, Vries M de, Bergen LF van, Galama JM, Keijser A, Bleijenberg G, Meer JW van der. Somatic hypotheses of war syndromes. Eur J Clin Invest2000; in press.
2. Vries M de, Soetekouw PM, Bergen LF van, Meer JW van der, Bleijenberg G. Somatic and psychological symptoms after acts of war and peace operations. (Dutch) Ned Tijdschr Geneeskd1999; 143:2557–62.[Medline]
3. Unwin C, Blatchley N, Coker W, Ferry S, Hotopf M, Hull L, Ismail K, Palmer I, David A, Wessely S. Health of UK servicemen who served in Persian Gulf War. Lancet1999; 353:169–78.[Web of Science][Medline]
4. Ismail K, Everitt B, Blatchley N, Hull L, Unwin C, David A, et al. Is there a Gulf War syndrome? Lancet1999; 353:179–82.[Web of Science][Medline]
5.
Fukuda K, Nisenbaum R, Stewart G, Thompson WW, Robin L, Washko RM, Noah DL, Barrett DH, Randall B, Herwaldt BL, Mawle AC, Reeves WC. Chronic multisymptom illness affecting Air Force veterans of the Gulf War. JAMA1998; 280:981–8.
6. Vercoulen JH, Swanink CM, Fennis JF, Galama JM, Meer JW van der, Bleijenberg G. Dimensional assessment of chronic fatigue syndrome. J Psychosom Res1994; 38:383–92.[Web of Science][Medline]
7. Vercoulen JH, Alberts M, Bleijenberg G. The Checklist Individual Strength. (Dutch) Gedragstherapie1999; 32:131–6.
8. Vries M de, Soetekouw PMMB, Bleijenberg G, Meer JW van der. Post Cambodia Complaints Study, phase I: an inventory of nature, extent and origin. Policy report for the Ministry of Defence. University Hospital Nijmegen, Nijmegen, The Netherlands, 1998.
9.
Vercoulen JH, Hommes OR, Swanink CM, Jongen PJ, Fennis JF, Galama JM, Meer JW van der, Bleijenberg G. The measurement of fatigue in patients with multiple sclerosis. A multidimensional comparison with patients with chronic fatigue syndrome and healthy subjects. Arch Neurol1996; 53:642–9.
10. Ishoy T, Suadicani P, Guldager B, Appleyard M, Hein HO, Gyntelberg F. State of health after deployment in the Persian Gulf. The Danish Gulf War Study. Dan Med Bull1999; 46:416–19.[Web of Science][Medline]
11. Gray GC, Kaiser KS, Hawksworth AW, Hall FW, Barrett CE. Increased postwar symptoms and psychological morbidity among U.S. Navy Gulf War veterans. Am J Trop Med Hyg1999; 60:758–66.[Abstract]
12. Murphy FM, Kang H, Dalager NA, Lee KY, Allen RE, Mather SH, Kizer KW. The health status of Gulf War veterans: lessons learned from the Department of Veterans Affairs Health Registry. Mil Med1999; 164:327–31.[Web of Science][Medline]
13. Wolfe J, Proctor SP, Davis JD, Borgos MS, Friedman MJ. Health symptoms reported by Persian Gulf War veterans two years after return. Am J Ind Med1998; 33:104–13.[Web of Science][Medline]
14. Kroenke K, Koslowe P, Roy M. Symptoms in 18,495 Persian Gulf War veterans. Latency of onset and lack of association with self-reported exposures. J Occup Environ Med1998; 40:520–8.[Web of Science][Medline]
15.
Hyams KC, Wignall FS, Roswell R. War syndromes and their evaluation: from the U.S. Civil War to the Persian Gulf War. Ann Intern Med1996; 125:398–405.
16. Hodgson MJ, Kipen HM. Gulf War illnesses: causation and treatment. J Occup Environ Med1999; 41:443–52.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J S Kalkman, M L Schillings, S P van der Werf, G W Padberg, M J Zwarts, B G M van Engelen, and G Bleijenberg Experienced fatigue in facioscapulohumeral dystrophy, myotonic dystrophy, and HMSN-I J. Neurol. Neurosurg. Psychiatry, October 1, 2005; 76(10): 1406 - 1409. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. De Vries, P.M.M.B. Soetekouw, J.W.M. Van Der Meer, H. Folgering, and G. Bleijenberg Physical activity and exercise performance in symptomatic Cambodia veterans QJM, February 1, 2002; 95(2): 99 - 105. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||