The Health of Gulf War Veterans

admin — Wed, 19 Aug 2009 15:21:07 +0000

FINDINGS IN BRIEF

Gulf War illness, the multisymptom condition resulting from service in the 1990-1991 Gulf War, is the most prominent health issue affecting Gulf War veterans, but not the only one. The Congressionally mandated Research Advisory Committee on Gulf War Veterans’ Illnesses has reviewed the extensive evidence now available, including important findings from scientific research and government investigations not considered by earlier panels, to determine what is known about the health consequences of military service in the Gulf War. This evidence identifies the foremost causes of Gulf War illness, describes biological characteristics of this condition, and provides direction for future research urgently needed to improve the health of Gulf War veterans.

Gulf War illness is a serious condition that affects at least one fourth of the 697,000 U.S. veterans who served in the 1990-1991 Gulf War. This complex of multiple concurrent symptoms typically includes persistent memory and concentration problems, chronic headaches, widespread pain, gastrointestinal problems, and other chronic abnormalities not explained by well-established diagnoses. No effective treatments have been identified for Gulf War illness and studies indicate that few veterans have recovered over time.

Gulf War illness fundamentally differs from trauma and stress-related syndromes described after other wars. Studies consistently indicate that Gulf War illness is not the result of combat or other stressors and that Gulf War veterans have lower rates of posttraumatic stress disorder than veterans of other wars. No similar widespread, unexplained symptomatic illness has been identified in veterans who have served in war zones since the Gulf War, including current Middle East deployments. Evidence strongly and consistently indicates that two Gulf War neurotoxic exposures are causally associated with Gulf War illness: 1) use of pyridostigmine bromide (PB) pills, given to protect troops from effects of nerve agents, and 2) pesticide use during deployment. Evidence includes the consistent association of Gulf War illness with PB and pesticides across studies of Gulf War veterans, identified dose-response effects, and research findings in other populations and in animal models.

For several Gulf War exposures, an association with Gulf War illness cannot be ruled out. These include low-level exposure to nerve agents, close proximity to oil well fires, receipt of multiple vaccines, and effects of combinations of Gulf War exposures. There is some evidence supporting a possible association between these exposures and Gulf War illness, but that evidence is inconsistent or limited in important ways.

Other wartime exposures are not likely to have caused Gulf War illness for the majority of ill veterans. For remaining exposures, there is little evidence supporting an association with Gulf War illness or a major role is unlikely based on what is known about exposure patterns during the Gulf War and more recent deployments. These include depleted uranium, anthrax vaccine, fuels, solvents, sand and particulates, infectious diseases, and chemical agent resistant coating (CARC).

Gulf War illness is associated with diverse biological alterations that most prominently affect the brain and nervous system. Research findings in veterans with Gulf War illness include significant differences in brain structure and function, autonomic nervous system function, neuroendocrine and immune measures, and measures associated with vulnerability to neurotoxic chemicals. There is little evidence of peripheral neuropathies in Gulf War veterans.

Gulf War illness has both similarities and differences with multisymptom conditions in the general population. Symptom-defined conditions like chronic fatigue syndrome, fibromyalgia, and multiple chemical sensitivity occur at elevated rates in Gulf War veterans, but account for only a small proportion of veterans with Gulf War illness.

Studies indicate that Gulf War veterans have significantly higher rates of amyotrophic lateral sclerosis (ALS) than other veterans, and that Gulf War veterans potentially exposed to nerve agents have died from brain cancer at elevated rates. Although these conditions have affected relatively few veterans, they are cause for concern and require continued monitoring.

Important questions remain about other Gulf War health issues. These include questions about rates of other neurological diseases, cancers, and diagnosed conditions in Gulf War veterans, current information on overall and disease-specific mortality rates in Gulf War veterans, and unanswered questions concerning the health of veterans’ children.

Federal Gulf War research programs have not been effective, historically, in addressing priority issues related to Gulf War illness and the health of Gulf War veterans. Substantial federal Gulf War research funding has been used for studies that have little or no relevance to the health of Gulf War veterans, and for research on stress and psychiatric illness. Recent Congressional actions have brought about promising new program developments at the Departments of Defense and Veterans Affairs, but overall federal funding for Gulf War research has declined dramatically since 2001.

A renewed federal research commitment is needed to identify effective treatments for Gulf War illness and address other priority Gulf War health issues. Adequate funding is required to achieve the critical objectives of improving the health of Gulf War veterans and preventing similar problems in future deployments. This is a national obligation, made especially urgent by the many years that Gulf War veterans have waited for answers and assistance.

Back to Top of Page

EXECUTIVE SUMMARY:

More than seventeen years have passed since the United States and its international allies liberated Kuwait from the grip of Saddam Hussein’s Iraqi military forces in the 1990-1991 Gulf War. Despite the swift and decisive victory achieved in Operation Desert Storm, at least one fourth of the nearly 700,000 U.S. military personnel who served in the war have experienced a complex of difficult and persistent health problems since their return home. Illness profiles typically include some combination of chronic headaches, cognitive difficulties, widespread pain, unexplained fatigue, chronic diarrhea, skin rashes, respiratory problems, and other abnormalities. This symptom complex, now commonly referred to as Gulf War illness, is not explained by routine medical evaluations or by psychiatric diagnoses, and has persisted, for many veterans, for 17 years. While specific symptoms can vary between individuals, a remarkably consistent illness profile has emerged from hundreds of reports and studies of different Gulf War veteran populations from different regions of the U.S., and from allied countries.

For many years, diverse views about the cause or causes of Gulf war illness have been put forward and vigorously debated. Hundreds of burning oil well fires that turned the Kuwaiti sky black with smoke, dramatic reports of uranium-tipped munitions, sandstorms, secret vaccines, and frequent chemical alarms, along with the government’s acknowledgment of nerve agent releases in theater, led many to believe that veterans were suffering from effects of hazardous exposures that occurred during their deployment. Government officials and special committee reports maintained that there was little evidence that this was the case, and noted that veterans returning from other wars have often experienced chronic health problems related to the stressful circumstances of serving in a war zone. All sides called for research to better understand the problem. Multiple official investigations were launched and hundreds of research studies funded.

In 1998, the U.S. Congress mandated the appointment of a public advisory panel of independent scientists and veterans to advise on federal research studies and programs to address the health consequences of the Gulf War. The Research Advisory Committee on Gulf War Veterans’ Illnesses was appointed by the Secretary of Veterans Affairs in 2002 and directed to evaluate the effectiveness of government research in addressing central questions on the nature, causes, and treatments of Gulf War-related illnesses. According to its charter, the guiding principle for the Committee’s work is the premise that the fundamental goal of all Gulf War-related government research is to improve the health of Gulf War veterans, and the choice and success of federal Gulf War research should be judged accordingly. The Committee has convened public meetings on a regular basis to consider the broad spectrum of scientific research, investigative reports, and government research activities related to the health of Gulf War veterans. In addition to annual reports on Committee meetings and activities, it has periodically issued formal scientific recommendations and reports. The Committee’s last extended report, Scientific Progress in Understanding Gulf War Veterans’ Illnesses, issued in 2004, provided findings and recommendations on topics the Committee had considered up to that time. The present report provides a comprehensive review of information and evidence on topics reviewed by the Committee since that time, as well as additional information on topics considered in the 2004 report.

The central focus of this report is Gulf War illness, the multisymptom condition that affects veterans of the 1990-1991 Gulf War at significantly elevated rates. Despite considerable government, scientific, and media attention, little was clearly understood about Gulf War illness for many years. Now, 17 years after the war, the extensive body of scientific research and government investigations that is currently available provides the basis for an evidence-based assessment of the nature and causes of Gulf War illness. As described throughout the report, scientific evidence leaves no question that Gulf War illness is a real condition with real causes and serious consequences for affected veterans. Research has also shown that this pattern of illness does not occur after every war and cannot be attributed to psychological stressors during the Gulf War.

Although Gulf War illness is the most prominent and widespread issue related to the health of Gulf War veterans, it is not the only one. Additional issues of importance include diagnosed medical and psychiatric conditions affecting Gulf War veterans, and questions related to the health of veterans’ family members. Section 1 of this report provides an overview of information related to the prevalence and characteristics of Gulf War illness, and other health issues, from the large body of Gulf War epidemiologic research. Section 2 addresses evidence related to the causes of Gulf War illness, including what has been learned about effects of psychological stressors, oil well fires, depleted uranium, and other exposures of possible concern, and compares the weight of evidence related to each exposure as a cause or contributor to Gulf War illness. Section 3 addresses the nature of Gulf War illness, reviewing research on biological findings associated with Gulf War illness and its relationship with multisymptom conditions found in the general population. Section 4 reviews research programs sponsored by federal agencies to address Gulf War-related health issues. Research recommendations provided in relation to topics considered in each section are summarized and prioritized in Section 5 of the report.

Gulf War research has posed a complex scientific challenge for researchers. Most obviously, Gulf War illness does not fit neatly into well-established categories of disease. The underlying pathophysiology of Gulf War illness is not apparent from routine clinical tests, and the illness appears not to be the result of a single cause producing a well-known effect. There are relatively few sources of objectively measured data for studying Gulf War illness or its association with events and exposures in the Gulf War. Some observers have suggested that these complexities pose too difficult a challenge, and that it is unlikely that the nature and causes of Gulf War illness can ever be known. On the contrary, the Committee has found that the extensive scientific research and other diverse sources of information related to the health of Gulf War veterans paint a cohesive picture that yields important answers to basic questions about both the nature and causes of Gulf War illness. These, in turn, provide direction for future research that is urgently needed to improve the health of Gulf War veterans.

Epidemiologic Research: What is Gulf War Illness and How Many Veterans Are Affected?

Gulf War illness refers to the complex of symptoms that affects veterans of the 1990-1991 Gulf War at significantly excess rates. It is characterized by multiple diverse symptoms not explained by established medical diagnoses or standard laboratory tests, symptoms that typically include a combination of memory and concentration problems, persistent headache, unexplained fatigue, and widespread pain, and can also include chronic digestive difficulties, respiratory symptoms, and skin rashes. A similar profile of excess symptoms has been described in every study of U.S. Gulf War veterans from different regions and units, and in Gulf War veterans from the United Kingdom and other allied countries. Gulf War illness is not the only health condition related to Gulf War service, but it is by far the most common. Gulf War illness prevalence estimates vary with the specific case definition used. Studies consistently indicate, however, that an excess of 25 to 32 percent of veterans who served in the 1990-1991 Gulf War are affected by a complex of multiple symptoms, variously defined, over and above rates in contemporary military personnel who did not deploy to the Gulf War. That means that between 175,000 and 210,000 of the nearly 700,000 U.S. veterans who served in the 1990-1991 Gulf War suffer from this persistent pattern of symptoms as a result of their wartime service.

Research has not supported early speculation that Gulf War illness is a stress-related condition. Large population-based studies of Gulf War veterans consistently indicate that Gulf War illness is not the result of combat or other deployment stressors, and that rates of posttraumatic stress disorder (PTSD) and other psychiatric conditions are relatively low in Gulf War veterans. Gulf War illness differs fundamentally from trauma and stress-related syndromes that have been described after other wars. No Gulf War illness-type problem, that is, no widespread symptomatic illness not explained by medical or psychiatric diagnoses, has been reported in veterans who served in Bosnia in the 1990s or in current conflicts in Iraq and Afghanistan.

Epidemiologic studies indicate that rates of Gulf War illness vary in different subgroups of Gulf War veterans. Gulf War illness affects veterans who served in the Army and Marines at higher rates than those in the Navy and Air Force, and enlisted personnel more than officers. Studies also indicate that Gulf War illness rates differ according to where veterans were located during deployment, with highest rates among troops who served in forward areas. More specifically, studies consistently show that the rate of Gulf War illness is associated with particular exposures that veterans encountered during deployment.

Identified links between veteran-reported exposures and Gulf War illness have raised a great deal of interest, but have also been the source of considerable confusion. The use of self-reported exposure information raises a number of concerns, most obviously in relation to recall bias. These concerns emphasize the importance of assessing findings across a broad spectrum of studies, rather than relying on results from individual studies, and of evaluating the impact of recall and other information bias on study results where possible.

The Committee identified an additional problem that has had a profound effect on epidemiologic study results and their interpretation. Exposures assessed in Gulf War studies are highly correlated, that is, veterans who had one type of exposure also usually had many others. In analyzing the effects of any single exposure during the war, it is essential that effects of other exposures be considered and adjusted for, to avoid the well-known problem of “confounding,” or confusing the effects of multiple exposures with one another. Many Gulf War epidemiologic studies failed to control for confounding effects, yielding illogical results that made it appear as if all, or nearly all, wartime exposures caused Gulf War illness. In contrast, adjusted results—that is, those that controlled for effects of other exposures in theater—consistently identified a very limited number of significant risk factors for Gulf War illness.

The Urgent Need for Effective Treatments for Gulf War Illness

Gulf War illness has persisted for a very long time for most ill veterans—over seventeen years for many. Studies indicate that few veterans with Gulf War illness have recovered over time and only a small minority have substantially improved. The federal Gulf War research effort has yet to provide tangible results in achieving its ultimate objective, that is, to improve the health of Gulf War veterans. Few treatments have been studied and none have been shown to provide significant benefit for a substantial number of ill veterans.

Treatments that are effective in improving the health of veterans with Gulf War illness are urgently needed. In recent years, Congressional actions have led to promising initiatives in this effort at both the Department of Defense (DOD) and the Department of Veterans Affairs (VA). At DOD, the Office of Congressionally Directed Medical Research Programs has developed an innovative program aimed at identifying treatments and diagnostic tests for Gulf War illness. The program funded a limited number of new treatment studies in 2007 and has invited proposals for additional studies to be funded in 2009. In addition, VA has sponsored a center of excellence for Gulf War research at the University of Texas Southwestern, focused on identifying specific biological abnormalities that underlie Gulf War illness that can be targeted for treatment. Research to identify effective treatments for Gulf War illness has been given highest priority by the Committee and requires expanded federal support.

Other Health Issues Affecting Gulf War Veterans

Although Gulf War illness has been the most prominent health issue associated with military service in the 1990-1991 Gulf War, a number of other health issues are extremely important. Studies have indicated that veterans of the 1990-1991 Gulf War have developed amyotrophic lateral sclerosis (ALS) at twice the rate of nondeployed veterans of the same era. Gulf War veterans who were downwind from nerve agent releases resulting from weapons demolitions at Khamisiyah, Iraq, in March of 1991, have also been found to have twice the rate of death due to brain cancer as other veterans in theater. Recent studies have suggested that excess cases of ALS have declined in recent years, but the seriousness of both ALS and brain cancer are clear causes for concern and require continued monitoring for the foreseeable future. These findings also highlight the need for information on rates of other diagnosed diseases, particularly neurological diseases and cancers, which have only minimally been assessed in Gulf War veterans. Multiple studies have reported that rates of PTSD and other psychiatric disorders are higher in Gulf War veterans than in nondeployed era veterans but are, overall, substantially lower than in veterans of other wars.

Hospitalization and mortality studies have identified only limited differences between Gulf War and nondeployed era veterans. Early U.S. mortality studies indicated that Gulf War veterans had higher death rates due to accidents, and somewhat lower disease-related mortality rates. Although identified differences appeared to diminish in the years after the war, the most recent year for which comprehensive mortality information has been reported for U.S. Gulf War veterans is 1997. Given concerns about diseases of longer latency, it is extremely important that current disease-specific mortality rates for U.S. Gulf War veterans be made publicly available, and reported on a regular basis.

For many years, concerns have been raised about rates of birth defects in Gulf War veterans’ children and anomalous health problems in their family members. Large population-based studies in the U.S. and the U.K. have provided some evidence of excess rates of several types of birth defects among children born to Gulf War veterans, in comparison to nondeployed era veterans. The specific types of birth defects identified have differed in different studies, however, and rates, overall, have been in the normal range expected in the general population. Phase III of VA’s large U.S. National Survey of Gulf War Era Veterans and their Families included clinical evaluations of veterans’ spouses and children. On clinical evaluation, no notable differences were identified between spouses of Gulf War and nondeployed veterans. Findings from clinical evaluations of veterans’ children have not been reported from this study, however. Further, no studies have provided comprehensive information on the health of Gulf War veterans’ children, including rates of diagnosed conditions, symptomatic illness, and learning and behavioral disorders.

What Caused Gulf War Illness? Review of Evidence Relating Gulf War Illness to Experiences and Exposures During Deployment

In addition to the many physical and psychological challenges common to other wartime deployments, military personnel who served in the 1990-1991 Gulf War were exposed to a long list of potentially hazardous substances. Many possible “causes” of Gulf War illness have been suggested and even promoted in different quarters since the war. Understanding the causes of Gulf War illness has been particularly challenging because of the lack of hard data on individual exposures in theater. Efforts by early government and scientific panels to address this issue were also limited by the sparsity of scientific research information on the health of Gulf War veterans for the first 10 years after the war.

This is no longer the case today, as a result of the extensive number of government investigations and scientific studies conducted to better understand events of the Gulf War and their association with Gulf War illness. Government reports have provided important insights into the types and patterns of exposures encountered by Gulf War military personnel. The large number of epidemiologic and clinical studies of Gulf War veterans also allow assessment of associations between Gulf War experiences and chronic health problems across a broad spectrum of veteran groups and research designs. In addition, toxicological studies conducted in recent years have provided extensive information on biological effects of Gulf War-related exposures that were previously unknown.

The Committee found that epidemiologic research on Gulf War veterans, assessed across diverse study designs and populations, provided clearer and more consistent findings than had previously been assumed. When combined with what has been learned about patterns of exposures in theater and findings from toxicological research, a coherent picture emerges about the most likely causes of Gulf War illness. The Committee used a standardized approach for evaluating available evidence related to psychological stressors in theater and each of the other deployment-related hazards of possible concern. Three major categories of evidence were considered. First, the Committee reviewed what is known about the extent and patterns of veterans’ exposure to each potential hazard. Second, the Committee reviewed the broad spectrum of available scientific research to determine what is known, in general, about health effects of each exposure. This included consideration of epidemiologic and clinical studies of human populations, and laboratory studies conducted in animal models. Third, the Committee reviewed, in detail, results from the many studies of Gulf War veterans that assessed associations between symptom complexes and the exposure in question.

Individually, single studies or types of information might suggest that a specific exposure could have caused Gulf War illness. But it is important to consider evidence of all types and studies from all sources to determine what the evidence most clearly indicates did cause Gulf War illness. Of the many experiences and exposures associated with Gulf War service, studies of Gulf War veterans consistently implicate only two wartime exposures as significant risk factors for Gulf War illness: use of pyridostigmine bromide (PB) pills as a nerve agent protective measure, and use of pesticides during deployment. This is consistent with what is known about the extent and patterns of these exposures in theater, and with general information from other human and animal studies. Studies of Gulf War veterans have also consistently indicated that psychological stressors during deployment are not significantly associated with Gulf War illness. For several other deployment exposures an association with Gulf War illness cannot be ruled out, due to inconsistencies or limitations of available information. Remaining exposures appear unlikely, from available evidence, to have caused Gulf War illness for the majority of affected veterans.

Psychological stress. Studies of Gulf War veterans consistently indicate that serving in combat and other psychological stressors during the war are not significantly associated with Gulf War illness, after adjusting for effects of other wartime exposures. Time-limited biological effects of psychological stressors have long been described in human studies, and more extreme psychological stressors and trauma can lead to chronic psychiatric disorders such as PTSD. Combat and extreme psychological stressors were less widespread and less sustained in the Gulf War than in other wars, including current Middle East deployments, and PTSD rates are lower in Gulf War veterans than in veterans of other wars. Population-based studies generally indicate that between three and six percent of Gulf War veterans are diagnosed with PTSD and that the large majority of veterans with Gulf War illness have no psychiatric disorders. Serving in combat and other wartime stressors are associated with higher rates of PTSD in Gulf War veterans, but not with higher rates of Gulf War illness.

Kuwaiti oil well fires. Widespread exposure to smoke from the Kuwaiti oil well fires was unique to military service in the 1991 Gulf War, and most prominently affected ground troops in forward locations. Epidemiologic findings relating oil well fire smoke exposure to Gulf War illness have been mixed, although a dose-response effect has been identified by several studies. There is little information from human or animal research to indicate whether intense exposure to petroleum smoke or vapors can lead to persistent multisymptom illness. Although studies of Gulf War veterans do not provide consistent evidence that exposure to oil fire smoke is a risk factor for Gulf War illness for most veterans, questions remain about effects for personnel located in close proximity to the burning wells for an extended period. Limited findings from epidemiologic studies indicate that higher-level exposures to smoke from the Kuwaiti oil well fires may be associated with increased rates of asthma in Gulf War veterans, and that an association with Gulf War illness cannot be ruled out.

Depleted uranium (DU). Low-level exposure to spent DU munitions and dust is thought to have been widespread during the Gulf War and was most prominent among ground troops in forward locations. Recent animal studies have demonstrated acute effects of soluble forms of DU on the brain and behavior, but persistent effects of short term, low-dose exposures like those encountered by the majority of Gulf War veterans have only minimally been assessed. There is little information from Gulf War or other human studies concerning chronic symptomatic illness in relation to DU or uranium exposure. Exposure to DU in post-Gulf War deployments, including current conflicts in the Middle East, has not been associated with widespread multisymptom illness. This suggests that exposure to DU munitions is not likely a primary cause of Gulf War illness. Questions remain about long-term health effects of higherdose exposures to DU, however, particularly in relation to other health outcomes.

Vaccines. Receipt of multiple vaccines over a brief time period is a common feature of overseas military deployments. About 150,000 Gulf War veterans are believed to have received one or two anthrax shots, most commonly troops who were in fixed support locations during the war. Although recent studies have demonstrated that the anthrax vaccine is highly reactogenic, there is no clear evidence from Gulf War studies that links the anthrax vaccine to Gulf War illness. Taken together, limited findings from Gulf War epidemiologic studies, the preferred administration to troops in support locations, and the lack of widespread multisymptom illness resulting from current deployments, combine to indicate that the anthrax vaccine is not a likely cause of Gulf War illness for most ill veterans. However, limited evidence from both animal research and Gulf War epidemiologic studies indicates that an association between Gulf War illness and receipt of a large number of vaccines cannot be ruled out.

Pyridostigmine bromide (PB). Widespread use of PB as a protective measure in the event of nerve gas exposure was unique to the 1990-1991 Gulf War. Pyridostigmine bromide is one of only two exposures consistently identified by Gulf War epidemiologic studies to be significantly associated with Gulf War illness. About half of Gulf War personnel are believed to have taken PB tablets during deployment, with greatest use among ground troops and those in forward locations. Several studies have identified dose-response effects, indicating that veterans who took PB for longer periods of time have higher illness rates than veterans who took less PB. In addition, clinical studies have identified significant associations between PB use during the Gulf War and neurocognitive and neuroendocrine alterations identified many years after the war. Taken together, these diverse types and sources of evidence provide a consistent and persuasive case that use of PB during the Gulf War is causally associated with Gulf War illness.

Pesticides. The widespread use of multiple types of pesticides and insect repellants in the Gulf War theater is credited with keeping rates of pest-borne diseases low. Pesticide use, assessed in different ways, is one of only two exposures consistently identified by Gulf War epidemiologic studies to be significantly associated with Gulf War illness. Multisymptom illness profiles similar to Gulf War illness have also been associated with low-level pesticide exposures in other human populations. In addition, Gulf War studies have identified dose-response effects, indicating that greater pesticide use is more strongly associated with Gulf War illness than more limited use. Pesticide use during the Gulf War has also been associated with neurocognitive deficits and neuroendocrine alterations in Gulf War veterans in clinical studies conducted many years after the war. Taken together, all available sources of evidence combine to support a consistent and compelling case that pesticide use during the Gulf War is causally associated with Gulf War illness.

Nerve agents. There have been no reports that U.S. forces encountered large-scale, high-dose exposures to chemical weapons during the Gulf War, but concerns have emerged related to possible longterm effects of low-dose nerve agent exposures. Recent animal studies have identified brain, autonomic, behavioral, neuroendocrine, and immune effects of low-level sarin exposure that were previously unknown. Studies of individuals exposed to symptomatic but sublethal doses of sarin in Japanese terrorist incidents in the 1990s have identified central nervous system effects that have persisted for many years. The extent of low-level exposure to nerve agents during the Gulf War, however, is unclear. Monitoring equipment used by U.S. forces had little capacity to detect nerve agents at levels that did not cause immediate symptoms. The Department of Defense estimates that about 100,000 U.S. troops may have been exposed to low levels of nerve agents following weapons demolitions in March of 1991 at Khamisiyah, Iraq, but questions have been raised about the models used to determine who was exposed, and at what levels. It is also unclear whether additional low-level exposures may have occurred in other locations. Veterans’ self-reported experiences concerning low-level nerve agent exposure in the Gulf War are particularly uncertain, and findings from epidemiologic studies linking chemical agents with Gulf War illness are inconsistent. Studies of Gulf War veterans have identified increased rates of brain cancer and measurable differences in brain structure and function that relate, in a dose-response manner, to modeled nerve agent exposure levels resulting from the Khamisiyah demolitions. Findings from Gulf War clinical studies, and from other human and animal research, suggest that an association between Gulf War illness and low-level nerve agent exposure cannot be ruled out, for whatever subgroups of veterans were exposed.

Infectious disease. A substantial proportion of Gulf War military personnel contracted acute gastrointestinal and respiratory infections during deployment, but there is little information concerning patterns of infection in theater and no evidence of widespread chronic illness resulting from those infections. Atypical leishmania infections were identified in a limited number of veterans who served in the 1990-1991 Gulf War, and a much larger number of leishmaniasis cases have been reported in personnel serving in the current Iraq War. Several studies have identified DNA indicators of mycoplasma infection in about 40 percent of symptomatic Gulf War veterans, but questions about testing methods have not been adequately addressed. Taken together, there is little clear evidence implicating infectious diseases as prominent causes of Gulf War illness. Questions remain, however, concerning the possibility that some individuals with Gulf War illness have undetected chronic leishmania and mycoplasma infections.

Other exposures in theater. A number of other potentially hazardous exposures in theater have been suggested as causing or contributing to Gulf War illness. These include fine sand and airborne particulates, exhaust from tent heaters, other fuel exposures, solvents, and freshly-applied CARC (chemical agent resistant coating) paint. For most, there is limited evidence of the types considered for other exposures. Available information, however, suggests that these exposures are not likely to have caused Gulf War illness for most affected veterans. Epidemiologic studies have provided little clear information linking any of these exposures to Gulf War illness and most were not most prevalent among ground troops who were forward deployed. Some, like sand, solvents, and fuel exposures, have also been widely encountered by personnel in current Middle East deployments. Information from human and animal studies indicates that fuel and solvent exposures can have neurological effects compatible with symptoms of Gulf War illness, but neither has been associated with Gulf War illness in studies of Gulf War veterans.

Combinations of exposures. Compared to the diverse types of evidence available related to effects of individual exposures, research on effects of combinations of Gulf War-related exposures is limited. Gulf War studies consistently indicate that exposures in theater were highly correlated—that is, that personnel most often experienced individual exposures in connection with multiple other exposures. This includes correlations between use of PB and pesticides and among different types of pesticides. Animal studies have identified significant effects of exposure to combinations of PB, pesticides and insect repellants, sarin, and stress, at dosage levels comparable to those experienced by veterans during the Gulf War. Diverse findings have been reported in relation to chemical absorption, metabolism, and biological effects of mixtures of neurotoxicants, which differ from those of individual exposures. There is little information from human studies, however, including the many epidemiologic studies of Gulf War veterans, concerning combined effects of Gulf War exposures.

A persuasive theoretical case can be made that exposure to mixtures of neurotoxic compounds in theater are likely contributors to Gulf War illness. Such a case would draw on the consistency of evidence from all sources indicating that both PB and pesticides are significantly associated with Gulf War illness, the high correlation between troops’ use of PB and pesticides during deployment, and synergistic effects between these exposures demonstrated by animal studies. Many of the pesticides used in the Gulf War, as well as PB and nerve agents, exert toxic effects on the brain and nervous system by altering levels of acetylcholine, an important nerve signaling chemical. Although such a case is compelling, little evidence is available from studies of Gulf War veterans to indicate whether or not Gulf War illness is associated with combinations of these exposures. This important possibility can and should be fully evaluated in Gulf War studies. Pending such assessments, it is not possible to definitively determine the extent to which mixtures of cholinergic and other neurotoxicant exposures during deployment contributed to Gulf War illness. Based on evidence from toxicological research in animals and what is known about patterns of exposures during the Gulf War, an association between Gulf War illness and combined effects of neurotoxicant exposures cannot be ruled out.

There is almost no research to indicate if other wartime exposures interact synergistically with these neurotoxic compounds or with one another. That is, the biological effects of different combinations of PB, multiple pesticides, low-level nerve agents, oil and dense smoke from burning wells, DU dust, fuel vapors, exhaust from tent heaters, CARC paint, airborne particulates, infectious agents, and receipt of multiple vaccines, experienced concurrently or over a brief time period, are unknown. Many have suggested that unknown and difficult-to-characterize effects may have been precipitated by an “exposure cocktail” or “toxic soup” effect during Gulf War deployment. While such a theory is intriguing, there is currently little evidence to indicate whether or not such effects actually occurred, and the extent to which they may have contributed to Gulf War illness.

What the Weight of Evidence Tells Us About the Causes of Gulf War Illness

Seventeen years after the Gulf War, answers to the question of what caused Gulf War illness remain vitally important. An extensive amount of available information now permits an evidence-based assessment of the relationship of Gulf War illness to the many experiences and exposures encountered by military personnel during the Gulf War. The strongest and most consistent evidence from Gulf War epidemiologic studies indicates that use of pyridostigmine bromide (PB) pills and pesticides are significant risk factors for Gulf War illness. The consistency of epidemiologic evidence linking these exposures to Gulf War illness, identified dose-response effects, findings from Gulf War clinical studies, additional research supporting biological plausibility, and the compatibility of these findings with known patterns of exposure during deployment, combine to provide a persuasive case that use of PB pills and pesticides during the 1990-1991 Gulf War are causally associated with Gulf War illness. Gulf War studies also consistently indicate that psychological stressors during deployment are not significantly associated with Gulf War illness.

Evidence related to other deployment-related exposures is not as abundant or consistent as evidence related to PB, pesticides, and psychological stressors. For several wartime exposures, there is some evidence supporting a possible association with Gulf War illness, but that evidence is inconsistent or limited in important ways. Clinical studies of Gulf War veterans, studies of other populations exposed to sarin, and findings from animal studies all suggest that low-level nerve agent exposure can produce persistent neurological effects that may be compatible with symptoms of Gulf War illness. Therefore, an association between Gulf War illness and low-level nerve agents cannot be ruled out for those veterans who were exposed. However, inconsistencies in epidemiologic studies and unreliable exposure information preclude a clear evaluation of the extent to which such exposures occurred and may have contributed to Gulf War illness. Limited evidence from several sources also suggests that an association with Gulf War illness cannot be ruled out in relation to combined effects of neurotoxicant exposures, receipt of multiple vaccines, and exposure to the Kuwaiti oil fires, particularly for personnel in close proximity to the burning wells for an extended period.

There is little reliable information from Gulf War studies concerning an association of DU or anthrax vaccine to Gulf War illness. The prominence of both exposures in more recent deployments, in the absence of widespread unexplained illness, suggests these exposures are unlikely to have been major causes of Gulf War illness for the majority of affected veterans. Fine blowing sand, solvents, and fuel exposures were also widely encountered in both the 1990-1991 Gulf War and in the current Iraq War and
from studies of Gulf War veterans have not supported an association between these exposures and Gulf War illness. All of the exposures described can be hazardous in some circumstances, however, and some veterans may have experienced adverse effects on a more limited basis.

The Nature of Gulf War Illness: Biological and Clinical Findings in Gulf War Veterans

Although veterans’ symptoms are the most obvious and consistent indicators of Gulf War illness, dozens of research studies conducted by multiple investigators have identified objective measures that significantly distinguish veterans with Gulf War illness from healthy controls. Identified differences to structure and function of the brain, function of the autonomic nervous system, neuroendocrine and immune alterations, and variability in enzymes that protect the body from neurotoxic chemicals. These findings provide indicators of diverse biological differences associated with Gulf War illness, but have not, as yet, provided measures that can be used as diagnostic tests. While scientific progress has been made in understanding the biological nature of Gulf War illness, important work remains in characterizing the specific pathophysiological processes that underlie veterans’ symptoms. The Committee reviewed the broad spectrum of studies that have evaluated biological and clinical parameters in Gulf War veterans, focusing most specifically on Gulf War illness.

Identified effects on the brain and central nervous system. Multiple lines of research have supported early indications that service in the Gulf War, for some veterans, resulted in long term effects on the central nervous system. Population-based studies of Gulf War veterans have consistently identified significantly excess rates of symptom complexes suggestive of central nervous system abnormalities. Studies have also indicated that Gulf War veterans developed amyotrophic lateral sclerosis (ALS) at twice the rate of nondeployed era veterans, and that veterans downwind from the Khamisiyah munitions demolitions have died from brain cancer at twice the rate of other Gulf War veterans. Earlier reports suggesting that Gulf War illness is not associated with neurological abnormalities generally referred to the lack of significant findings identified with standard clinical evaluations and peripheral nerve function testing. It is important to distinguish the lack of findings in these areas from the diverse central nervous system effects identified using specialized brain imaging scans, neuropsychological testing, and measures of balance and audiovestibular function.

Neuroimaging studies. Three research teams have identified significant differences between veterans with Gulf War illness and controls using proton magnetic resonance spectroscopy (MRS) scans of the brain. Findings indicate that symptomatic veterans have significantly reduced functioning brain cell mass in the brainstem, basal ganglia, and hippocampus. Reduced neuronal function in the left basal ganglia was correlated with increased central dopamine activity in one study. Symptomatic Gulf War veterans have also been reported to exhibit alterations in overall and regional cerebral blood flow, using specialized SPECT scan analyses. In addition, a significant correlation has been reported between reduced white matter volume in Gulf War veterans and levels of nerve agent exposures resulting from the Khamisiyah weapons demolitions. Preliminary results from three unpublished federal Gulf War research projects are also of great interest, and will be reviewed in final form as they become available. These include early results from a larger MRS study that appear not to support earlier findings of reduced neuronal function in the brainstem and basal ganglia of symptomatic Gulf War veterans. Preliminary findings from an additional SPECT study suggest that symptomatic Gulf War veterans differ from healthy controls in cerebral blood flow responses to cholinergic challenge. Early results from a third study indicate that symptomatic Gulf War veterans have significantly reduced total white matter volume compared to healthy controls. In contrast to the diverse findings reported from studies using specialized brain imaging methods, few abnormalities have been identified in symptomatic veterans using electroencephalograms (EEG), computed tomography (CT) scans, or standard magnetic resonance imaging (MRI) of the brain.

Overall, of the seven identified Gulf War research projects that evaluated brain structure and function using proton MRS, specialized SPECT scans, and specialized MRI assessments, six have identified significant differences between veterans with Gulf War illness and healthy controls, and one identified no case/control differences. An additional study has identified significant brain volume differences in Gulf War veterans in relation to modeled nerve agent exposures during the Gulf War. These findings have been important in documenting brain alterations in Gulf War veterans, but have often come from relatively small studies that assessed different types of abnormalities in different areas. Additional research is needed to determine if these findings can be replicated and/or further extended in larger samples.

Neuropsychological studies. Neuropsychological studies provide objective measures of brain function and have been used for many years to quantify neurocognitive deficits resulting from chemical exposures. They constitute the largest body of research on central nervous system function in Gulf War veterans. A wide variety of specialized tests are used to assess cognitive domains that include attention, executive system functioning, motor skills, visuospatial functioning, memory, and mood. Changes in affect and emotional functioning can be symptoms of brain injury, and so are important to measure in neuropsychological tests. But PTSD and other psychiatric conditions can themselves affect neurocognitive function, and so must be appropriately controlled for when analyzing test outcomes.

Research studies have consistently identified significant differences in neurocognitive function between symptomatic Gulf War veterans and healthy controls. These include differences on tests of attention and executive system functioning, memory, visuospatial skills, psychomotor skills, and mood and emotional functioning. Some studies indicate that symptomatic veterans display a slowing of response speed that affects their mental flexibility across multiple cognitive domains. Identified differences have generally been modest, but have consistently been significant and remained significant after adjustments for emotional functioning and psychiatric disorders. Studies also indicate that many symptomatic veterans who report cognitive difficulties do not have objectively measurable neurocognitive deficits. Two studies have identified subgroups of symptomatic Gulf War veterans with more marked neurocognitive impairment on measures of memory, attention, and response time, suggesting this subgroup should be the focus of additional study.

Studies have also evaluated veterans’ neurocognitive function in relation to exposures during the Gulf War. Significantly poorer performance on tests of memory, attention, and mood have been identified in relation to self-reported exposure to pesticides, PB, and chemical weapons. Neurocognitive effects have also been identified in relation to modeled nerve agent exposures resulting from the Khamisiyah weapons demolitions. Department of Defense-modeled nerve agent exposure levels were significantly correlated with slower performance on psychomotor and visuospatial tasks in a dose-response pattern—that is, greater exposure was associated with worse neurocognitive performance.

Autonomic nervous system dysfunction. The autonomic nervous system (ANS) is the part of the nervous system that regulates involuntary, or “automatic” physiological activities. Autonomic pathology can be associated with diverse symptoms such as dizziness, weakness, digestive abnormalities, and sexual dysfunction. Autonomic function is often assessed by determining effects of physiological challenges on ANS regulation of heart rate and blood pressure. The Committee reviewed results from seven published studies and two additional federal projects that assessed ANS function in symptomatic Gulf War veterans. Eight of nine projects identified significant ANS differences between veterans with Gulf War illness and healthy controls. Several studies demonstrated blunted autonomic responsivity to physiological challenges, for example, reduced cardiovascular compensation in response to orthostatic challenge on tilt table testing. Studies have also identified a general reduction in heart rate variability in the high frequency range among veterans with Gulf War illness, observed over a 24-hour period in one study and during nighttime hours in another. Although ANS differences have consistently been reported in veterans with Gulf War illness, specific ANS alterations identified by different studies have varied, as a result of differences in study characteristics and testing methods. Additional comprehensive research is needed to provide a clear characterization of Gulf War illness-related autonomic dysfunction.

Neuromuscular and sensory findings. Symptoms reported by Gulf War veterans frequently include muscle pain and weakness, or numbness and tingling sensations in the extremities. Such symptoms potentially indicate abnormalities in peripheral nerve function related to sensation and motor function. Nine studies have assessed peripheral sensory and neuromuscular function in Gulf War veterans. Overall, based on standard clinical examination, electromyography, and nerve conduction tests, these studies have provided little indication that veterans with Gulf War illness are affected by generalized polyneuropathies or abnormal neuromuscular transmission. Three of four studies that evaluated sensory threshold measures identified significantly higher (that is, less sensitive) thresholds in symptomatic compared to healthy veterans, however. Two identified higher cold sensory thresholds, and one reported a higher threshold for detecting light touch, suggesting that some Gulf War veterans may have subtle small sensory fiber neuropathies. Consistent findings that Gulf War veterans are not affected by more generalized polyneuropathies or neuromuscular abnormalities indicate that veterans’ neuromuscular symptoms are not attributable to overt muscle damage or peripheral nerve pathology.

Neuroendocrine alterations. A series of recent studies have provided detailed evaluation of hypothalamic-pituitary-adrenal (HPA) axis functioning in Gulf War veterans. Studies indicated that Gulf War veterans are similar to nondeployed veterans on baseline measures of cortisol and ACTH (adrenocorticotropic hormone), but had significantly greater suppression of both hormones in response to dexamethasone challenge. These responses were significantly associated with veterans’ symptoms, most prominently their musculoskeletal symptoms, but were unrelated to combat exposure or whether veterans had PTSD. Cortisol suppression was most pronounced in veterans who reported using PB during deployment. In addition, 24-hour ACTH levels were significantly reduced among Gulf War veterans who did not have PTSD, and were associated with veterans’ use of pesticides and PB. No HPA alterations were associated with combat stress, with other self-reported exposures during deployment, or with PTSD in Gulf War veterans. Overall, these studies suggest that Gulf War service and symptoms of Gulf War illness are associated with a unique profile of HPA alterations many years after the war, effects that differ from HPA findings associated with other conditions, including PTSD. Identified effects were independent of combat stress, but significantly associated with veterans’ use of PB and/or pesticides.

Vulnerability to neurotoxicants. A question often asked about Gulf War illness is why some Gulf War military personnel developed chronic symptoms during and after deployment, while others who served along side them remained well. It is well established that some people are more vulnerable to adverse effects of certain chemicals than others, due to variability in biological processes that neutralize those chemicals, and clear them from the body. The enzyme paraoxonase (PON1) circulates in the blood and hydrolyzes organophosphate compounds such as pesticides and nerve agents, converting them to relatively harmless chemicals that are then excreted. Individuals who produce different types and amounts of PON1 differ, sometimes dramatically, in their ability to neutralize different organophosphate compounds. The Committee reviewed results from four published studies and two additional federal projects that have assessed PON1 measures in Gulf War veterans. Five of the six projects identified significant PON1 differences that were associated with Gulf War illness or, more generally, with Gulf War service. Specific findings from these studies varied, however, reflecting different types of data that addressed different research questions. Additional research is needed to better characterize the precise nature of the PON1-Gulf War illness relationship. It is unknown if Gulf War illness is linked to biological variability in other enzymes that protect the body from neurotoxic exposures. Limited and preliminary information from three studies suggest a possible link between Gulf War illness and butyrylcholinesterase (BChE) that may involve the subset of veterans who have very low BChE activity and also experienced specific exposures during the war.

Immune parameters. There has been little indication that Gulf War service, overall, is associated with increased rates of diagnosable immune conditions, including autoimmune diseases and allergies, or with increased susceptibility to infectious disease. A well-known hypothesis, suggesting that Gulf War illness is related to a systemic shift favoring Th-2 type immunity, has not been supported by studies of Gulf War veterans. Veterans with Gulf War illness have been shown to differ from healthy controls on a number of immune parameters, however. A variety of specific differences have been identified by individual studies, and a number of consistent findings have emerged. Results from two studies, using different methods in different groups of symptomatic veterans, indicate that Gulf War illness is associated with a low-level, persistent immune activation, reflected in elevated levels of the cytokines IL-2, IFN-γ and IL-10. Several studies have also reported that NK cell numbers and/or cytotoxic activity are significantly reduced in veterans with Gulf War illness. A fuller understanding of immune function in ill Gulf War veterans is needed, particularly in veteran subgroups with different clinical characteristics and exposure histories.

Additional research and clinical findings in Gulf War veterans.

Additional information pertaining to biological and clinical characteristics of symptomatic Gulf War veterans is available from a variety of clinical reports and studies. Individual clinical studies have provided several findings of interest, such as increased sensitivity to pain and elevated rates of fibromyalgia in veterans with musculoskeletal symptoms, dyspepsia and persistent diarrhea similar to irritable bowel syndrome in veterans with gastrointestinal symptoms, abnormal pulmonary function in a subset of veterans with respiratory symptoms, and verification of rashes and other skin anomalies in veterans with dermatological symptoms. But overall, objective indicators of disease are often not identified in symptomatic Gulf War veterans who are referred for specialty evaluations. Clinical reports have also not provided explanations for identified problems, such as the causes of veterans’ persistent diarrhea or rashes. One study evaluated Gulf War veteran males and their sexual partners who experienced a painful burning reaction to the veterans’ seminal fluid, a problem reported by about seven percent of Gulf War veterans. Evaluations indicated that about 40 percent of the women had a hypersensitivity reaction to the veterans’ seminal fluid, but provided no explanation for the phenomenon, overall. In general, very limited information is available on health problems specific to women veterans. Single studies have reported that Gulf War veteran women report elevated rates of yeast and bladder infections and breast lumps or cysts, butresults are available from medical evaluations.

Single studies have identified additional significant differences between symptomatic veterans and controls on a number of specific laboratory tests. These include elevated rates of coagulation abnormalities in symptomatic veterans, an elevated proportion of symptomatic veterans with insertion/deletion polymorphisms in the gene encoding for angiotensin-converting enzyme, and identification of atypical circulating polyribonucleotides potentially indicative of chromosome alterations.

Future directions in identifying physiological mechanisms that underlie Gulf War illness.

To advance efforts to identify effective treatments and diagnostic tests for Gulf War illness, the Committee has recently expanded its work to review areas of research that may contribute to a better understanding of the specific pathophysiological mechanisms that underlie veterans’ symptoms. This has included preliminary discussions in several areas, including biological processes associated with neuroplasticity, disordered sensory processing and neuroendocrine dysregulation, and mitochondrial insufficiency. The Committee has also reviewed, in greater detail, diverse scientific findings that suggest a potential role for central nervous system inflammatory processes in the pathophysiology of Gulf War illness, and has identified this as a promising area for future research. The research considered indicates that neurotoxic Gulf War exposures may activate inflammatory processes in the brain and that increased brain levels of proinflammatory cytokines can produce a complex of multiple symptoms similar to Gulf War illness. Additional research suggests that these processes can become dysregulated by mechanisms that include repeated cycles of brain cell injury and glial activation, as well as autonomic and neuroendocrine disruption. Research in this area is especially warranted because of its possible clinical implications. Imaging methods are available that can potentially identify these processes in the brain and a variety of therapeutic agents are being studied for their effectiveness in treating dysregulated central inflammatory processes. Gulf War Illness in Relation to Other Multisymptom Conditions Parallels are commonly drawn between Gulf War illness and symptom-defined conditions such as chronic fatigue syndrome (CFS), fibromyalgia (FM), and multiple chemical sensitivity (MCS) found in the general population. The prevalence of CFS in Gulf War veterans is unique, and dramatically higher than CFS rates found in nondeployed veterans and in the general population. Rates of FM and MCS are also elevated in Gulf War veterans, but to a lesser degree. It is clear from multiple studies, however, that case definitions for CFS, FM, and MCS do not adequately describe the chronic symptom complex that affects Gulf War veterans at excess rates, and that only a fraction of veterans with Gulf War illness can be diagnosed with any of these conditions. Overall, research studies have identified both similarities and differences between Gulf War illness and other multisymptom conditions. General similarities are reflected in indicators of autonomic dysregulation and neurocognitive impairment in Gulf War illness, FM, and CFS, and by indications that Gulf War illness and MCS are linked to PON1 variability. In contrast, the epidemiologic profile of Gulf War illness significantly differs from multisymptom conditions in the general population. Studies have also identified immune parameters and a number of other measures that differ in veterans with Gulf War illness, compared to patients with CFS or FM. Many objective measures associated with these conditions have not been evaluated in veterans with Gulf War illness, however. Additional research in these areas can potentially provide useful insights into biological mechanisms that underlie Gulf War illness and contribute to identifying beneficial treatments.

Federal Gulf War Research Programs

In addition to scientific studies and government reports, the Committee is charged with reviewing federal research programs established to address health consequences of the 1991 Gulf War. Since 1994, th U.S. government has reported expenditures of $340 million, over $440 million if indirect costs are considered, for hundreds of studies identified as Gulf War research in interagency reports to Congress. This research has been funded primarily by the Department of Defense (DOD) and the Department of Veterans Affairs (VA). Many federally-funded studies have provided valuable insights regarding the health of Gulf War veterans, as detailed throughout this report. But much of the federally funded research has not advanced understanding of Gulf War illness or other Gulf War-related health problems. Consequently, federal Gulf War research programs have not, as yet, succeeded in achieving the primary objective of Gulf War research, that is, to improve the health of Gulf War veterans. The Committee identified major problems related to the historical use of research funds identified as “Gulf War research” expenditures by federal agencies. Historically, the large majority of Gulf War research funding was provided by DOD. In recent years, DOD has dramatically cut funding for projects identified as Gulf War research from nearly $30 million annually in 2001 to under $5 million in 2006. More troubling, many studies identified as “Gulf War research” at DOD over that period had little or no relevance to Gulf War illness or the health of Gulf War veterans. The DOD “Gulf War” portfolio consisted largely of costly projects that addressed broad questions related to current deployments and other health issues unrelated to the Gulf War. By 2006, less than 10 percent of the $4.7 million identified as DOD funding for “Gulf War research” supported studies that related to Gulf War illness or other health problems associated with Gulf War service.

The Department of Veterans Affairs had historically funded a smaller proportion of federal Gulf War research, but increased funding in recent years from a low of $4 million annually in 2002 to nearly $13 million in 2006. VA also historically identified a large number of studies as “Gulf War research” that had little relevance to Gulf War health issues. Until 2004, this included substantial funding for research on stress and psychiatric illness. By 2006, a larger number of studies had been funded that were related to Gulf War illness and effects of Gulf War exposures. Still, the largest amount of funding in VA’s Gulf War research portfolio, nearly 40 percent of the $13 million in 2006, supported projects focused on amyotrophic lateral sclerosis (ALS), few of which included Gulf War veterans or research issues related to the development of ALS in Gulf War veterans.

A number of important changes have taken place in federal Gulf War research programs in recent years. Beginning in 2006, Congressional actions brought about major changes in Gulf War research at both VA and DOD. Congress allocated an additional $15 million annually for Gulf War research at VA, and directed that it be used to support a center of excellence for Gulf War research at the University of Texas Southwestern (UTSW) in Dallas. The VA/UTSW program is focused on identifying biological abnormalities associated with Gulf War illness that can be targeted to develop diagnostic tests and treatments. Congress also appropriated $5 million in 2006 and $10 million in 2008 to support an innovative Gulf War research program managed by DOD’s Office of Congressionally Directed Medical Research Programs. The new DOD Gulf War research program is focused on identifying treatments for Gulf War illness and objective measures that distinguish ill from healthy veterans. Early indications suggest that developments at both VA and DOD represent promising new directions in the federal Gulf War research effort. The overall federal funding commitment for Gulf War research, however, remains substantially below historical funding levels and far below that warranted by the scope of the problem.

Research Priorities and Recommendations

The Committee is charged with determining what has been learned about the nature, causes, and treatments for Gulf War illness and advising on federal research, with the primary goal of improving the health of Gulf War veterans. In reviewing information on the broad variety of topics related to the health of Gulf War veterans, the Committee identified many scientific issues for which additional research was needed. Specific research recommendations have been provided in relation to each topic considered, and are compiled and prioritized in the final section of the report.

The Committee recommends that highest priority be given to research directed at identifying beneficial treatments for Gulf War illness. This includes clinical studies that systematically evaluate the effectiveness of currently available treatments, as well as research to identify specific pathophysiological mechanisms associated with Gulf War illness that can be targeted for treatment. The Committee also gives high priority to research aimed at identifying objective biological markers associated with Gulf War illness, especially those that advance efforts to improve diagnostic testing. Recommended research includes studies that expand on existing biological findings in Gulf War veterans—comprehensiveresearch on brain structure and function, autonomic function, neuroendocrine and immune alterations, and processes associated with biological vulnerability to neurotoxicants—as well as studies that investigate neuroinflammatory processes and utilize genomic and related technologies to identify biological characteristics of Gulf War illness. Additional research priority areas include studies that characterize effects of neurotoxic exposures associated with Gulf War illness, and epidemiologic studies to assess rates of neurological diseases in Gulf War veterans.

The Committee identified additional areas of research needed to address other important Gulf War health issues. These include epidemiologic studies to identify mortality and cancer rates in Gulf War veterans, evaluation of health problems in veterans’ children, and improved characterization of Gulf War-related health problems in relation to exposures in theater. Recommendations are also provided for improving clinical and epidemiologic research on Gulf War veterans, and emphasize the importance of evaluating outcomes in subgroups of Gulf War veterans identified by illness characteristics and exposures in theater.

The Committee recognizes the vital importance of Congressional support, agency commitment and leadership, and adequate federal funding for achieving critical scientific objectives related to the health of Gulf War veterans and preventing similar problems in future deployments. It therefore recommends that the Administration request and that Congress allocate not less than $60 million annually in the federal budget for Gulf War research, an amount commensurate with the scope of the problem, and compatible with funding levels between 1999 and 2001. The Committee also recommends that this funding be specifically directed to research most capable of improving the health of Gulf War veterans, as outlined in this report.

Conclusions

Veterans of the 1990-1991 Gulf War had the distinction of serving their country in a military operation that was a tremendous success, achieved in short order. But many had the misfortune of developing lasting health consequences that were poorly understood and, for too long, denied or trivialized. The extensive body of scientific research now available consistently indicates that Gulf War illness is real, that it is the result of neurotoxic exposures during Gulf War deployment, and that few veterans have recovered or substantially improved with time. Addressing the serious and persistent health problems affecting 175,000 Gulf War veterans remains the obligation of the federal government and all who are indebted to the military men and women who risked their lives in Iraq, Kuwait, and Saudi Arabia 17 years ago. This obligation is made more urgent by the length of time Gulf War veterans have waited for answers and assistance.

Back to Top of Page

INTRODUCTION:

More than 17 years have passed since the Gulf War. The events and successes of Operation Desert Storm are becoming a distant memory for some, with international attention now focused on current military operations in Iraq and Afghanistan. But for too many who served in the Persian Gulf theater in 1990 and 1991, the Gulf War has had lasting consequences—health consequences beyond the well-recognized effects of bullets and bombs and the psychological impact of war. This report describes what has been learned in the last 17 years about the health effects of military service in the Gulf War and identifies priority research issues that remain to be addressed. Although the report covers the broad spectrum of health concerns related to Gulf War service, its primary focus is the multisymptom condition that has come to be known as Gulf War illness. Over the years, this condition has been the foremost Gulf Warrelated health issue and the focus of intense political and scientific interest. It is also the condition for which the largest numbers of Gulf War veterans are still seeking clear answers and effective treatments.

The Gulf War was unlike any war fought before or since. In the days following Iraq’s invasion of Kuwait in August 1990, an international effort was swiftly mounted to stand up to the aggression of Saddam Hussein’s forces. During Operation Desert Shield, hundreds of thousands of American troops, along with military forces from the United Kingdom and dozens of other allied countries, established a strong foothold in the region over the course of a few months. By mid-January, 1991, Operation Desert Storm began with a massive air campaign. Six weeks later, on February 24, 1991, the ground offensive was launched as U.S. and allied troops moved into Southern Iraq and Kuwait. In just three days the allies achieved their primary objective, retaking Kuwait City as the Iraqis fled. The next day, just 100 hours after the ground war had begun, a cease fire was established. Kuwait was free and the U.S. and its allies had achieved a great victory in the desert.

The 1990-1991 Gulf War was an overwhelmingly successful campaign. Following the six-week air war and four-day ground war, victorious troops were welcomed home and hailed as heroes in parades and ceremonies across the nation. Just under 150 combat-related deaths occurred among the 700,000 Americans who deployed to the region, far fewer than had been anticipated before the war. The military medical system established for the war had also performed impressively. Even with the quick mobilization and harsh, unfamiliar desert environment, a record low number of troops required medical attention during deployment. The Gulf War was unquestionably a unique war—for its brevity, for the success with which it was executed, and for the decisive nature of the victory.

Yet, despite the successful staging and outcome of the Gulf War, military personnel who served in theater reported persistent, baffling symptoms during deployment and in the months and years that followed their return home. Reports indicated that Gulf War veterans who had served in different units, from all parts of the U.S. and from allied countries were affected by similar types of symptoms. Illness profiles typically included a complex of multiple symptoms not explained by conventional medical or psychiatric diagnoses—cognitive difficulties, persistent and widespread pain, fatigue, headaches, chronic diarrhea and other digestive abnormalities, and skin rashes. Just what these problems were and what had caused them was unknown.

Over the years, Gulf War illness has posed diverse and difficult challenges for veterans who are ill and for healthcare providers and research scientists working to address this condition. From the earliest time veterans’ symptoms became known, they have been surrounded by controversy and conjecture. And for most of the decade that followed the Gulf War, relatively little was understood about the nature and causes of Gulf War illness. Since the middle 1990s, Gulf War-related health problems have been the subject of numerous expert panel reports, U.S. and international government investigations, and hundreds of scientific studies. As a result, an enormous amount of information is now available on events and circumstances of the Gulf War and the health of Gulf War veterans. These resources, when considered in aggregate, provide long-needed answers to questions concerning Gulf War illness and provide a focus for the scientific research needed to effectively address it.

The Research Advisory Committee on Gulf War Veterans’ Illnesses. In 1998, with many questions remaining about veterans’ unexplained health problems, Congress mandated the appointment of an independent panel of scientists and veterans to review all federal research programs and available evidence relating to the health of Gulf War veterans. In response to Section 104 of Public Law 105- 368,1243 the Research Advisory Committee on Gulf War Veterans’ Illnesses was appointed in 2002 by then Secretary of Veterans Affairs Anthony J. Principi. The Committee was charged with assessing the effectiveness of the federal research effort in answering “central questions on the nature, causes, and treatments for Gulf War-associated illnesses” (Appendix B) and with providing scientific recommendations on federal research programs and studies.

Over the past six years, the Committee has had the privilege and responsibility of reviewing diverse types of information on the many topics pertinent to the health of Gulf War veterans. A challenge common to earlier government and expert panels was the sparsity of scientific information on which to base findings and recommendations. In contrast, the Committee found that the quantity of currently available information created a different kind of challenge, requiring a comprehensive review and cohesive synthesis of a voluminous number of reports and studies. A complete picture of what is currently known about the Gulf War and the health of Gulf War veterans was needed in order to make meaningful research recommendations on the best way forward.

Since its inception, the Committee has conducted its work in public meetings, convened three times per year. Due to the breadth of information to be considered, a systematic approach has been used in reviewing each area of interest. For each topic considered, relevant materials have been reviewed by the Committee, and scientists and government representatives with diverse expertise and perspectives have been invited to present results of their investigations. The information presented typically addressed what has been learned about particular exposures and events in theater and/or results of scientific studies concerning health effects of those exposures. Committee meetings have functioned in large part as symposia, providing opportunities for Committee members, visiting scientists, and government officials to review and discuss available information on each topic, as well as opportunities for comments and questions from members of the public.

In addition to annual reports on its activities and ongoing discourse with federal research officials, the Committee has periodically issued recommendations and formal reports concerning topics it has considered. An early “Interim Report” was issued in June, 2002, that provided the Committee’s preliminary impressions and recommendations based on an initial overview of available research information. The Committee’s first extensive report was issued in the fall of 2004, providing detailed information on topics considered to that time.1268 These included the scope of Gulf War illness and the need for treatment research, evidence concerning effects of Gulf War-related neurotoxic exposures, studies of birth defects in veterans’ children, and programmatic and funding issues related to federal research on the health of Gulf War veterans. In January, 2006, the Committee provided updated recommendations that outlined priority research objectives and topics to VA’s Office of Research and Development. Additional recommendations were provided to the Secretary in February, 2007, concerning the need for updating Gulf War illness-related research and educational materials for VA clinicians.1271 In 2008, the Committee reported its findings and recommendations concerning initial plans and research activities at the VA-funded Gulf War Illness and Chemical Exposure Research Program at the University of Texas Southwestern. Committee members have also testified before Congress on issues related to Gulf War illness research, including the need to identify effective treatments.

The present report summarizes information reviewed by the Committee since its last major report in 2004 and synthesizes all information considered by the Committee thus far. This synthesis forms the basis for the scientific recommendations made in each area, and for identification of research priorities. In the current report, as in all its activities, the Committee has been mindful of the guiding principle designated for its work, as described in the Committee’s charter. It states that “the fundamental goal of Gulf Warrelated government research, either basic or applied, is to ultimately improve the health of ill Gulf War veterans, and that the choice and success of research efforts shall be judged accordingly”(Appendix B).

The current report differs from earlier Gulf War panel and committee reports in several important respects. First, the central focus of this report is Gulf War illness. The Committee reviewed available information on all health issues associated with Gulf War service, but prioritized information relating to the nature and causes of the undiagnosed, multisymptom illness affecting Gulf War veterans. Despite the prominence of this condition in the lives of ill veterans and the amount of government and media attention given to this problem, Gulf War illness has received surprisingly little in-depth consideration by previous scientific panels. The present report is also distinct from earlier Gulf War reports because it comes at a time when an unprecedented amount of information is available to inform the work and conclusions of the Committee, information that was not available to earlier review panels and scientific committees.

Lastly, this report is unique because of the specific charge and scope of activities assigned to the Committee. This has enabled a single panel to consider the extensive range of topics related to the health of Gulf War veterans. It has also required full consideration of the many types of scientific studies and government reports relevant to veterans’ health and effects of veterans’ experiences and exposures during the war. As a result, the Committee has had the opportunity to engage these complex issues in a more comprehensive manner than previously has been possible. Most importantly, it has permitted the Committee to synthesize diverse information from diverse sources in order to identify patterns and inconsistencies across a broad spectrum. In effect, the Committee was given the opportunity to assemble and evaluate all available pieces from a complex puzzle, and to determine what, collectively, they tell us about the nature and causes of Gulf War illness.

It is regrettable that, 17 years after the war, so little clear information has emerged from scientific committees that specifically addresses the nature and causes of Gulf War illness. It is perhaps understandable, in light of the many complexities related to research in this area, as will be described throughout this report. Most obviously, Gulf War illness does not fit neatly into our current concepts of disease. The underlying pathobiology of Gulf War illness is not apparent from routine clinical tests, and the illness appears not to be the result of a single cause producing a well-known effect. Researchers and clinicians are generally not familiar with methods required to evaluate and address health problems identified entirely by veterans’ symptoms. This might explain why Gulf War researchers and committees have often focused their attention on problems that are more routinely assessed and measured. It has become clear over the years, however, that the important questions surrounding Gulf War illness do not have simple answers. Addressing these questions requires that complex issues be engaged in a complex and comprehensive manner. Overly simplistic and compartmentalized approaches have provided little progress.

The present report is divided into several sections that reflect different aspects of available information on Gulf War-related health issues. The first section provides an overview of what has been learned from population studies, the large body of epidemiologic research on Gulf War veterans. The second section addresses the cause of Gulf War illness, reviewing what has been learned about the many Gulf Warrelated experiences and exposures that potentially contributed to veterans’ ill health—from the psychological stress of war to the effects of oil well fires, nerve agents, vaccines, and depleted uranium.

The third section addresses the nature of Gulf War illness, reviewing research on biological abnormalities associated with veterans’ symptoms, the relationship of Gulf War illness with multisymptom conditions in civilian populations, and topics the Committee has considered in exploring physiological mechanisms that may underlie veterans’ symptoms. The fourth section summarizes the current status of federal research programs related to the health of Gulf War veterans. Each of the first four sections includes research recommendations related to the specific topics considered. The fifth section summarizes and prioritizes these recommendations.

As is described throughout the report, there is no question that Gulf War illness is a real condition with real causes and serious consequences for affected veterans. Study after study has consistently documented this multisymptom condition in large numbers of Gulf War veterans. Research has also shown that this pattern of illness does not occur after every war and cannot be attributed to psychological stressors during the Gulf War. Because research studies have so compellingly demonstrated that Gulf War illness cannot be explained simply as the expected result of wartime stress, it remains the responsibility of the federal government to fully elaborate the source and nature of this condition, to care for affected veterans, and to prevent similar problems from happening in the future.

Some have suggested that the many scientific and political challenges that have impeded understanding of Gulf War illness are too complex, that the events of the war are too remote, and that answers to the many questions surrounding Gulf War illness might never be known.870,1765 On the contrary, the Committee has found that the diverse sources of information and research data associated with Gulf War service paint a cohesive picture that yields important answers to basic questions about both the nature and causes of Gulf War illness. These, in turn, provide direction for future research that is most capable of improving the health of Gulf War veterans. Completing this mission, that is, finding answers and treatments for ill Gulf War veterans, requires continued dedicated effort and cooperation between government officials, scientists, clinicians, and veterans. As will be evident from the information and recommendations that follow, the Committee believes that this is a challenge that can be met. It is also, unquestionably, an obligation that must be met.

Back to Top of Page

CHAPTER 1| Gulf War Illness and the Health of Gulf War Veterans
Characteristics and impact of Gulf War illness: Epidemiologic research
Gulf War illness prognosis and the need for treatments
Is there a unique Gulf War Syndrome?
Other Gulf War health issues
Special committee and government reports on the health of Gulf War veterans
Recommendations

CHAPTER 2| What Caused Gulf War Illness? Effects of Gulf War Experiences and Exposures
Psychological stressors and the health of Gulf War veterans
Kuwaiti oil well fires and the health of Gulf War veterans
Depleted uranium and the health of Gulf War veterans
Vaccines and Gulf War illness
Cholinergic and related neurotoxicants: Pyridostigmine bromide, pesticides, and nerve agents
Infectious diseases in Gulf War veterans
Other exposures in theater
Synthesis: What the weight of evidence tells us about the causes of Gulf War illness

CHAPTER 3| The Nature of Gulf War Illness
Biological and clinical characteristics of Gulf War illness
Gulf War illness in relation to multisymptom conditions in the general population

CHAPTER 4| Federal Research on Gulf War Illness and the Health of Gulf War Veterans
Historical funding and management of federal Gulf War research
Gulf War research at the Department of Veterans Affairs
Gulf War research at the Department of Defense
Gulf War research at the Centers for Disease Control and Prevention
Additional VA programs relevant to Gulf War research
Recommendations

CHAPTER 5| Research Priorities and Recommendations
Highest priority Gulf War research
Other research areas of importance for addressing Gulf War health issues
Guidelines for clinical and epidemiologic research on Gulf War veterans

Acknowledgements

References

Appendices
Appendix A| Association of Gulf War Experiences and Exposures with Chronic Symptoms and Multisymptom Illness: Results from Studies of Gulf War Veterans
Appendix B| Charter of the Research Advisory Committee on Gulf War Veterans’ Illnesses…… 437
Appendix C| Members of the Research Advisory Committee on Gulf War Veterans’ Illnesses

Abbreviations and Acronyms

Back to Top of Page

CHAPTER 1: Gulf War Illness and the Health of Gulf War Veterans

I arrived in Theater on January 6, 1991 … During official visits to strategic military cities there were frequent SCUD attacks during which I heard chemical alarms sound. When I asked if these alarms meant chemicals had been detected, I was told that the chemical alarms had malfunctioned. I became ill and was treated for nausea, headaches, vomiting, diarrhea, and high temperature. Rashes I had over my body I thought were normal and expected since I spent most days in the sand, wind, and sun with all the attendant fleas, flies, and desert parasites. Headaches I attributed to fatigue and lack of sleep. The symptoms…continued after I returned home and got progressively worse. –COL GR, Gulf War veteran1684

Unexplained illness in the wake of Desert Storm. In the years immediately following Desert Storm, widespread reports indicated that Gulf War veterans were suffering from a complex of symptoms that included memory problems, profound fatigue, chronic pain, persistent diarrhea, and unusual skin lesions. Similar symptom complexes were widely reported by veterans from different units in different parts of the U.S., and also by veterans from allied countries. Medical evaluations provided limited insights, since veterans’ symptoms were typically not associated with abnormalities on laboratory tests or other diagnostic measures. No clear explanation was apparent for this unexplained symptom complex, labeled “Gulf War Syndrome” by the media.

Veterans and other observers soon raised questions about whether hazardous exposures encountered during the Gulf War had made troops sick. Suggested causes included the billowing clouds of thick black smoke produced by the Kuwaiti oil wells that were set afire by retreating Iraqi soldiers in the closing days of the war. There were also widespread reports that alarms designed to detect chemical agents repeatedly sounded in some areas of theater after Coalition air bombing began in January of 1991. Additional concerns were raised about the use of measures that had never before been fielded by the military on a widespread basis. These included use of munitions and armoring containing depleted uranium, use of an anti-nerve agent prophylaxis regimen that included regular doses of the drug pyridostigmine bromide, and administration of the anthrax and botulinum toxoid vaccines.

Since the mid-1990s, the federal government has funded hundreds of research studies to investigate the health problems affecting Gulf War veterans.340 Multiple large epidemiologic studies and impressive data collections have been conducted in diverse populations of Gulf War veterans. These studies have provided extensive documentation of the symptoms and symptom complexes associated with Gulf War service, rates of psychiatric conditions in Gulf War veterans, and limited data on the extent to which Gulf War veterans have been affected by diagnosed medical diseases. Without exception, studies of Gulf War veterans have found that the most prevalent health problem associated with Gulf War service is the complex of multiple symptoms not explained by familiar medical or psychiatric diagnoses.

The condition once labeled “Gulf War Syndrome” by the media is now commonly referred to as Gulf War veterans’ illnesses or Gulf War illness. The Committee has adopted the term Gulf War illness for simplicity’s sake. It is used as an umbrella term to represent varying definitions and descriptions of the complex of multiple symptoms found at significantly excess rates in Gulf War veterans. As with other conditions, the specific symptoms affecting veterans with Gulf War illness can vary somewhat from person to person. The overall consistency of the types of health problems described in Gulf War veterans, however, indicates that it is most useful to consider this excess symptomatology as a cohesive “entity” to be studied as a “multisymptom illness” as opposed to considering symptoms individually. This is the approach adopted by most epidemiologic studies of Gulf War veterans. Use of the specific term, Gulf War illness, also allows this multisymptom condition to be clearly distinguished from other, more familiar, diagnosed conditions that affect individual veterans.

Although Gulf War illness is the most prominent condition affecting Gulf War veterans, it is just one health issue to be addressed in the larger context of the health of Gulf War veterans. Other Gulf Warrelated health issues of importance include rates of diagnosable medical conditions and post-war mortality among Gulf War veterans, and questions related to the risk of birth defects and other health problems in veterans’ family members.

Characteristics and Impact of Gulf War Illness: Epidemiologic Research

My symptoms began in the Gulf with severe abdominal cramping and severe diarrhea. I also had terrible headaches and bouts of dizziness and tingling. Once I returned to the base in Germany, the headaches persisted, and I experienced the cramps and diarrhea on a cyclic basis. I also went through periods of night sweats. And there were periods when I would sleep a lot because I was so fatigued. My joints were stiff, and my knees would swell after I ran. It was harder for me to do things without feeling short of breath. These symptoms became worse as time passed… Ever since my return from the Gulf, I’ve been plagued by multiple rashes and lesions on my face, neck, arms, and back. They come and go. –SSgt BJ, Army Gulf War veteran716

Significant scientific progress has been made in characterizing the health of Gulf War veterans, as described in the Committee’s 2004 report.1268 This progress has relied, in large part, on the many Gulf War epidemiologic studies conducted in the past decade. Epidemiologic research uses established methods to study patterns of disease and related factors in populations. Among the strengths of this research approach is its capacity for providing “big picture” information about the health of populations and statistical assessment of the relationship of health problems with demographic characteristics, biological and chemical exposures, and other factors that can affect health. It is, in fact, the only scientific approach capable of evaluating health problems in relation to the actual complex conditions of the Gulf War. Consequently, epidemiologic research has been a particularly important resource for understanding Gulf War illness and the health of Gulf War veterans.

The extensive body of Gulf War epidemiologic research has provided a consistent picture of the general characteristics of Gulf War illness and the patterns in which it affects diverse groups of Gulf War veterans. This research is not without limitations, however. It is important that findings from individual population studies of Gulf War veterans be evaluated in the context of identified limitations, and also considered in the context of the larger body of studies addressing similar questions and issues.

How many Gulf War veterans have Gulf War illness? The prevalence of Gulf War illness reported by different studies has varied with how Gulf War illness or “chronic multisymptom illness” (CMI) cases are defined. Because no specific Gulf War illness case definition has been widely accepted, the Committee reviewed prevalence estimates from all studies reporting rates of multisymptom illness, by any case definition, in both Gulf War veterans and nondeployed Gulf War-era veterans. The burden of multisymptom illness attributable to service in the Gulf War was determined by comparing rates found in Gulf War veterans to those in nondeployed era veterans. The excess rate in Gulf War veterans, that is, the rate over and above that in veterans who did not serve in the Gulf War, reflects the proportion of veterans whose multisymptom condition can be attributed to participation in the Gulf War.

As shown in Table 1, nearly all epidemiologic studies have reported that, regardless of the case definition used, an excess of 25 - 32 percent of Gulf War veterans have multisymptom illness related to service in the Gulf War. The only exception comes from results reported from Phase III of the U.S. National Survey of Gulf War Era Veterans and Their Families, which found only a 13 percent excess rate of multisymptom illness in Gulf War veterans.142 It is not clear why the rate of excess illness from this study was lower, by about half, than all other studies, including a later follow-up of veterans from the same U.S. A similar degree of excess ill health related to Gulf War service is suggested by studies that have assessed veterans’ health using more general indicators. For example, half of Iowa Gulf War veterans, but only 14 percent of nondeployed era veterans, indicated they had health problems that they attributed to their military service in 1990-1991, an excess of 36 percent in Gulf War veterans.350 Similarly, an excess of 35 percent of Kansas Gulf War veterans reported having health problems attributable to military service in 1990-1991.1476 A 2002 British study determined that 53 percent of Gulf War veterans fell into one of four clusters defined by patterns of elevated symptom scores, compared to 28 percent of nondeployed era veterans, an excess of 25 percent in Gulf War veterans.421 A more recent British study reported that 61 percent of Gulf War veterans, and 37 percent of nondeployed era veterans reported new health problems since the Gulf War, an excess of 24 percent in Gulf War veterans.1411 Overall, these studies provide a consistent indication that excess subjective ill health attributable to service in the Gulf War affects between 24 and 36 percent of those who served.

Due to the consistency of estimates of the excess prevalence of multisymptom illness from diverse studies, the Committee concludes that approximately 25 to 30 percent of veterans who served in the Gulf War have been affected by Gulf War illness. That is, studies indicate that between 175,000 and 210,000 of the 700,000 American veterans of the 1990-1991 Gulf War are affected by a complex of multiple symptoms attributable to their service in the war.

Characteristics of Gulf War Illness Gulf War illness has been widely described in government testimony, media reports, and scientific studies. The condition is typically characterized as a combination of diverse symptoms such as memory problems, chronic headaches, widespread pain, unexplained fatigue, mood changes, persistent diarrhea, respiratory problems and skin rashes. One of the major challenges of identifying, treating, and understanding Gulf War illness is that ill veterans often have no abnormal findings on clinical diagnostic tests. As a result, Gulf War illness is characterized on the basis of veterans’ symptoms that are, by definition, self-reported. While this presents a number of difficulties in clinical practice, it is not an impediment to assessing Gulf War illness in epidemiologic studies of large groups of veterans, where general patterns of symptoms can be assessed and compared.

In research studies, Gulf War illness is routinely defined by the presence of multiple symptoms affecting different systems. The majority of these symptoms fall into general categories, or domains, that have often been characterized statistically in large studies. Symptom domains identified in broadly representative populations of Gulf War veterans are summarized in Table 2. Despite the diverse methods used to characterize symptoms, the categories of symptoms that affect Gulf War veterans at excess rates are remarkably consistent across studies. The two symptom groups most commonly identified include those indicative of neurological/cognitive problems (e.g., chronic headache, cognitive difficulties, mood disturbances, vision and balance abnormalities) and symptoms of persistent, widespread pain in joints and muscles. Symptoms related to persistent fatigue (e.g. extreme tiredness, sleep abnormalities) are reported just as frequently, classified in different studies either as a specific symptom domain, or as part of the neurological domain.

Two additional symptom groups are also consistently found at excess rates in Gulf War veterans, but are typically reported by fewer veterans than neurological, pain, and fatigue symptoms. These include respiratory symptoms (e.g. wheezing, coughing) and gastrointestinal problems (e.g. chronic diarrhea, abdominal cramping). Skin symptoms (unexplained rashes and lesions) are also routinely reported, but have usually not been assessed by multiple variables, as required for identifying symptom “groups.” Factor analysis of symptoms. A number of Gulf War studies have defined symptom domains that affect Gulf War veterans using factor analysis. This statistical technique is generally used as a data reduction method in developing psychometric instruments or defining patient subgroups in studies of identified medical or psychiatric conditions. Factor analysis identifies “latent” constructs, or factors, that may underlie sets of highly correlated variables. When applied to general health symptoms in diverse populations, the factor constructs typically reflect the correlation between symptoms resulting from problems affecting particular organs or biological processes. These correlations tend to be independent of the specific diseases causing those symptoms. For example, symptoms of coughing, wheezing, and shortness of breath are highly correlated in any population, regardless of whether different individuals in that population have pneumonia, emphysema, or colds.

With limited exception,752 the types of symptom domains identified in Gulf War studies by factor analysis also occur in nondeployed veterans, and in diverse, nonveteran populations.1127,1341,1786,1830 This would generally be expected, since factor-identified symptom “groupings,” in studies that assess general health symptoms in heterogeneous populations, simply reflect the high correlation between symptoms resulting from distress in a particular organ or biological process, regardless of the underlying disease.1082 What is unique to Gulf War veterans is that persistent symptoms occur concurrently in multiple domains at excess rates, and with greater severity, than in nondeployed veterans.240,464,698,752,1395,1476 As described in the Committee’s 2004 report, individual Gulf War veterans experience chronic symptoms in multiple domains at the same time. If a unique pattern of Gulf War symptoms were to be identified using factor analysis, it would likely require consideration of higher-order factors, that is, second or third-level factors that reflect “groupings” of the symptom factors identified in both Gulf War and nondeployed veterans.1082 Defining Gulf War illness. Several research teams have developed case definitions for use in their investigations of the prevalence of and risk factors for Gulf War-related multisymptom illness. Generally, case definitions that are nonspecific, that is, those based on a small number of common symptoms, identify a relatively large number of “cases” among both Gulf War and nondeployed era veterans. More restrictive case definitions, in terms of the types and severity of symptoms required, identify fewer cases. Case definitions used to describe Gulf War illness and the methods for arriving at them have varied between studies, as summarized below.

Haley syndromes. In 1997, Dr. Robert Haley and colleagues at the University of Texas Southwestern Medical School defined three syndromes based on factor analysis of an extensive, detailed battery of dichotomous and scaled symptoms in 249 members of the 24th Reserve Naval Mobile Construction Battalion.565 Syndrome 1, labeled “impaired cognition,” included problems with attention, memory, depression, and sleep abnormalities. Syndrome 2, labeled “confusion-ataxia,” was characterized by problems with thinking and balance and was the most severe of the three syndromes. Syndrome 3, labeled “arthro-myo-neuropathy” was associated with joint and muscle pain. Twenty percent of the veterans in the study had one of more of the three defined syndromes.

Chronic multisymptom illness (CMI). In 1998, Dr. Keiji Fukuda and investigators from the U.S. Centers for Disease Control and Prevention (CDC) defined a complex of symptoms, termed “chronic multisymptom illness” in a population of 3,723 deployed and nondeployed Air Force Gulf War era veterans.464 The symptom complex was defined using two parallel methods: one assessed symptoms that affected more than 25 percent of Gulf War veterans, and the other used factor analyses of dichotomous symptoms reported by both Gulf War and nondeployed veterans assessed together. The resulting CMI definition required cases to report one or more symptoms lasting six months or longer in at least two of three categories: fatigue, mood-cognition (feeling depressed, difficulty remembering or concentrating, feeling moody, feeling anxious, trouble finding words, difficulty sleeping) and musculoskeletal pain (joint pain, joint stiffness, muscle pain). Severe CMI cases rated each defining symptom as severe, other cases were labeled “mild-to-moderate” CMI. As defined, the CMI symptom complex affected 45 percent of Gulf War veterans in the Air Force sample, and 15 percent of nondeployed era veterans.

Oregon-defined Gulf War unexplained illness. In 1998, Dr. Peter Spencer and colleagues from the Oregon Health Sciences University defined cases of Gulf War-related unexplained illness (GWUI) for inclusion in a case control study.1465 GWUI cases were veterans who had at least one defining symptom from any of three categories, but no diagnostic explanation for that symptom. The three categories included unexplained symptoms associated with fatigue, cognitive/psychological problems, and musculoskeletal complaints.

Kansas-defined Gulf War illness. In 2000, Dr. Lea Steele reported a case definition for Gulf War illness, identified empirically as the pattern of symptoms that significantly distinguished 1,548 Gulf War veterans from 482 nondeployed era veterans in the Kansas Gulf Veterans Health Study.1476 Kansasdefined Gulf War illness criteria excluded veterans diagnosed with specified medical or psychiatric conditions that might explain their symptoms. Symptom criteria required that veterans report multiple or moderately severe symptoms in at least three of six defined categories: fatigue/sleep problems, pain symptoms, neurological/cognitive/mood symptoms, gastrointestinal symptoms, respiratory symptoms, and skin symptoms. Gulf War illness, as defined in the Kansas study, affected 34 percent of Gulf War veterans, and eight percent of nondeployed veterans.

U.S. National Survey-defined Gulf War syndrome. VA investigators identified a complex of four neurological symptoms that constituted a unique factor in Gulf War veterans but not in nondeployed era veterans in a large U.S. national sample.752 These symptoms included blurred vision, loss of balance, tremors/shaking, and speech difficulty. Investigators reported that 277 (2%) of the over 10,000 Gulf War veterans in the study were “cases” who endorsed all four symptoms. Cases were also significantly more likely to report a number of other symptoms and diagnosed conditions including migraines, seizures, and diarrhea.

Other studies have distinguished ill from healthy veterans in ways that did not require veterans to report specific symptoms. The 2002 Navy Seabee Health Study defined Gulf War illness “cases” as veterans who reported being diagnosed with at least one of four conditions (chronic fatigue syndrome, posttraumatic stress disorder, multiple chemical sensitivity, irritable bowel syndrome) and/or veterans who reported having at least 12 health problems.527 Researchers evaluating a cohort of Army veterans who returned from theater through Fort Devens, Massachusetts, have used several methods to classify symptomatic veterans. In some studies of this cohort, veterans who reported having five or more frequent symptoms from the Health System Checklist were identified as “high symptom” cases, and compared to veterans with fewer symptoms.1802 An alternate approach relied on comparison of scores in nine defined “body system” groups.1239 A separate study of veterans enrolled in VA’s Gulf War Registry also distinguished “high symptom” and “low symptom” veteran subgroups, with categories defined statistically using results of factor and cluster analyses.570 Studies have also used the chronic fatigue syndrome case definition465 to distinguish symptomatic Gulf War veterans from controls.321,435,1090

Seventeen years after the Gulf War, no case definition has been widely accepted as the preferred standard for defining the complex of multiple symptoms affecting Gulf War veterans, nor have there been published efforts to optimize or validate a Gulf War illness case definition. The Fukuda CMI case definition has been modified for use in several surveys and for two Gulf War illness clinical trials. That definition is generally considered overly broad, that is, nonspecific for the health problems affecting Gulf War veterans. The only case definition developed by characterizing a pattern of multiple symptom types that differed between Gulf War and nondeployed era veterans is the Kansas case definition. In a random sample of over 2,000 Kansas Gulf War era veterans, the Kansas definition more specifically distinguished symptomatic Gulf War from nondeployed era veterans than the CMI case definition.1476

The severity and functional impact of Gulf War illness. Although not well characterized by any research studies, anecdotal reports indicate that the severity of Gulf War illness is highly variable. Some veterans are mildly or moderately affected by their symptoms, but still able to maintain many of their usual activities. Others veterans have more severe, even disabling, illness. Different studies have shown that between 13 and 50 percent who meet CMI criteria for Gulf War illness can be classified as “severe” cases.142,464,1804 Illness severity is also said to vary for individuals, with symptoms waxing and waning over time.

Several studies have evaluated the degree to which Gulf War illness has affected veterans’ functional status, including their ability to work. The Medical Outcomes Study Short Form Survey (SF36),1753 and a special SF36 developed for veteran populations,779 have been widely used in Gulf War studies. The SF36 evaluates functional status in eight defined areas (e.g., physical functioning, social functioning, general health), providing a quantitative measure of health-related quality of life. Studies consistently report that veterans meeting any case definition of Gulf War illness have significantly lower scores on all SF36 indicators than population norms and than healthy veterans.87,142,160,449,464,567,1542,1726 Veterans seen at VA’s specialty referral clinics for multisymptom illness, the War Related Injury and Illness Study Centers, exhibit considerable functional impairment. Those veterans’ mean score of 30 on the SF36 physical component scale (PCS) is substantially below the national average score of 50.907 In Gulf War studies, SF36 scores vary with the case definition used and the domains assessed. The highly symptomatic Gulf War veterans who meet defining criteria for the Haley syndromes had lower SF36 domain scores, indicating worse functional status, than scores for conditions such as congestive heart failure and chronic obstructive pulmonary disease.567 In contrast, veterans meeting the CMI case definition in the U.S. national survey were considerably less functionally impaired, with a mean SF36 PCS score of 43.142 In general, studies indicate that most veterans with Gulf War illness continue to work, although this varies with illness severity. A relatively high rate of unemployment (29%) was reported in Gulf War veterans seeking treatment at a VA Gulf War illness clinic in Seattle.1281 More representative figures come from a population-based study in the Pacific Northwest, where 21 percent of Gulf War veterans with two or more unexplained health problems were unemployed, compared to 13 percent of veterans without symptoms. Employed-but-symptomatic veterans were also more likely to miss at least seven days of work due to illness over a one-year period (29%) than healthy veterans (4%).160

Which Veterans are Most Affected by Gulf War Illness? Epidemiologic studies traditionally describe patterns of disease in populations. Insights about the causes of a condition can often be drawn from identifying subgroups that are affected at higher and lower rates. Studies of Gulf War illness have reported patterns of this type, identifying different rates of illness in relation to the characteristics of veterans’ military service and deployment to the Gulf War theater.

Differences related to branch of service and military rank. Epidemiologic studies have consistently indicated that Gulf War veterans who served in the Army and Marines have higher rates of multisymptom illness than those in the Navy and Air Force.753,1466,1476 Similarly, Army veterans are disproportionately represented in VA and DOD Gulf War Registry programs. That is, Army personnel constituted just 50 percent of the deployed force, but account for 77 percent of Gulf War veterans enrolled in registries. Conversely, Air Force and Navy veterans are significantly underrepresented in U.S. Gulf War registries.1651

Studies also consistently report that enlisted personnel have higher rates of Gulf War illness than officers.241,511,697,753,1124,1466,1476 Comparisons between reservists and active duty personnel have produced mixed results, with some studies finding similar Gulf War illness rates in the two groups, and others reporting somewhat higher rates in either active duty or reserve veterans.692,697,753,1466,1476,1804 Demographic characteristics. Rates of Gulf War illness have generally not differed markedly with veterans’ demographic characteristics such as gender, age, and race. Gulf War illness affects women at about the same, or slightly higher rates than men,142,160,240,464,753,1476,1699,1804 and younger veterans at about the same rates as older veterans.511,753,1124,1476,1804 Whites and nonwhites are also affected at similar rates.142,160,240,464,753,1466,1476,1699,1804

Location in theater. Several studies have reported that Gulf War illness rates differ with the locations where veterans served during the war. That is, veterans who served in some areas of theater have higher rates of Gulf War illness than veterans who were in other locations. The study of Kansas veterans indicated that veterans who entered Iraq or Kuwait, countries where all battles took place, had significantly higher rates of Gulf War illness (42%) than veterans who served exclusively in support areas on land (32%) or on board ship during deployment (21%).1476 Similarly, U.S. and Canadian ground troops had higher rates of multisymptom illness than those who served on board ship511,753 and Iowa veterans who had been in Iraq, Kuwait, or Saudi Arabia had more health conditions than those located elsewhere in theater.692

Two studies have reported increased illness risk in more narrowly defined locations. Navy Seabees located in a specific sector of northeastern Saudi Arabia on the third day of the air war had over four times the rate of Gulf War illness as veterans in other areas, suggesting a link with a particular event or exposure in that location.564 A more recent report, using troop location data and geographical information system (GIS) methods, identified several localized spatial clusters where veterans with severe Gulf War illness were more likely to have been located at certain time periods.1236 Taken together, these studies indicate that Gulf War illness did not randomly affect all Gulf War veterans who deployed to the region, but occurred as a result of events, experiences, or exposures that differed by location.

In fact, epidemiologic studies have consistently found that Gulf War illness rates do vary significantly according to veteran-reported experiences and exposures during the war. Observed associations between Gulf War illness and veterans’ exposures have raised a great deal of interest, but have also been the source of considerable confusion. Research related to illness-exposure associations will be considered throughout this report, and analyzed in detail in Section 2. It is important that it be considered in the larger context of limitations inherent in the use of self-reported data, as well as other methodological issues affecting studies of Gulf War veterans.

Evaluating Causal Factors in Gulf War Illness Limitations and shortcomings of Gulf War epidemiologic research. A great deal has been learned from the many epidemiologic studies conducted in different populations of Gulf War veterans. But like all areas of scientific investigation, epidemiologic research has limitations, some of which are especially problematic in Gulf War studies. In addition to issues that are specific to Gulf War research, broader issues such as shortcomings in how research questions have been posed or how studies have been designed and executed have also greatly affected the degree to which Gulf War epidemiologic studies have been useful and informative. Therefore, it is essential that research limitations be identified and carefully considered when interpreting results of Gulf War epidemiologic studies.

Research on the health of Gulf War veterans is unusually complex and challenging for a number of reasons. Relatively little objectively measured information is available on either the primary health outcome of interest—Gulf War illness—or on potential causal factors assessed in epidemiologic studies. Gulf War illness is generally identified on the basis of veterans’ symptoms which are, by definition, selfreported. In addition, wartime events and exposures have most often been assessed using veterans’ own reports of what they experienced during deployment. Although not generally an optimal data resource, it is the only option available for many Gulf War exposures of interest.

To add to these complexities, both the primary health outcome of interest—Gulf War illness—and the etiologic factors being investigated are multifactorial. That is, Gulf War illness encompasses multiple symptoms that co-occur in different ways. Likewise, the Gulf War experience included a wide array of potentially hazardous and stressful exposures. Even under the best circumstances, understanding relationships between multifaceted exposures and multifaceted health outcomes can be a complex challenge.

Such issues, particular to studies of Gulf War illness, must also be considered in the context of limitations and problems more generally associated with epidemiologic research. Typical issues relate to biases that can result from the size and characteristics of the study sample, the response rate, identification of suitable comparison groups, the content and wording of questions, methods used to assess outcomes, and statistical problems stemming from multiple comparisons. These issues are well recognized and have been discussed at length in reviews and committee reports on the health of Gulf War veterans.104,646,686 An additional concern that has received less attention, but one that can have serious consequences, relates to methods used in analyzing and reporting the data collected in Gulf War studies. The present discussion focuses on sources of error in epidemiologic studies that have had the greatest impact on research findings, and interpretation of findings, in studies of Gulf War veterans.

Some reports have suggested that, given the limitations associated with studies of Gulf War veterans and the lack of data on measured exposures in theater, little useful information can be obtained from epidemiologic studies for understanding Gulf War illness and its relationship to exposures.79,667,686 But after reviewing results from the many Gulf War epidemiologic studies and carefully considering the impact of identifiable limitations, the Committee has concluded that data from these studies are interpretable and informative. In its analysis of Gulf War epidemiologic research, the Committee has emphasized patterns of illness and associations that are consistent across multiple studies. It has also given more credence to findings from studies that have used preferred methods in sampling, data collection, and data analysis.

A key methodological issue raised in the Committee’s 2004 report is the importance of evaluating health outcomes in identifiable subgroups of Gulf War veterans, as opposed to assessing all deployed veterans as a single group. This requires assessing the health status of veterans who served in particular locations, those in particular units, or those known to have had specific exposures.79,210,593,1477 Combining all Gulf War veterans into a single group may obscure or completely mask health effects due to events or exposures that did not affect all deployed personnel. For example, studies have reported selected health outcomes in veteran subgroups identified by modeled exposure to oil well fire smoke as well as nerve agents in relation to the Khamisiyah demolitions. Several have provided stark examples of links between exposures and disease or biological abnormalities that were not apparent when all deployed veterans were evaluated as a single group.190,192,599,1237

Confounding and risk factors for Gulf War illness. A major challenge in understanding results from Gulf War epidemiologic studies relates to the complex exposure scenario present in the Gulf War theater. Studies have typically evaluated effects of 20 or more experiences and exposures in theater— everything from combat experiences and other sources of stress, to oil well fires, vaccines, pesticides, and chemical alarms. Studies have consistently found that Gulf War exposures are highly correlated.161,241,458,1466 That is, veterans who reported some specific exposures during deployment were significantly more likely to also report other specific types of exposures. Those familiar with epidemiologic methods will quickly recognize the serious potential this raises for confounding, that is, confusing the effects of one deployment-related exposure with effects of multiple other exposures. 811 In less complex settings, confounding can be a major source of error that gives rise to incorrect—even nonsensical—findings, misleading both investigators and those who read their studies.734 In a uniquely complex exposure scenario such as the Gulf War, the impact of confounding can be profound. Fortunately, familiar analytic methods are available that can both identify which exposures are related to which other ones, and “tease out” effects of individual exposures. This allows epidemiologic studies to identify “independent” associations between illness and each exposure in a complex setting. In scenarios like the Gulf War, where many veterans encountered multiple varied exposures, use of such methods is essential to determine which Gulf War experiences are truly linked to ill health and which only appear to be, as a result of confounding.

Many Gulf War epidemiologic studies were careful to control for possible confounding by demographic factors such as age and gender, or military characteristics such as rank and branch of service. As detailed in Appendix A, adjustment for demographic factors typically had little effect on study results. In contrast, consistent, sometimes dramatic, confounding effects were demonstrated by studies that adjusted preliminary results for effects of multiple deployment-related exposures, as shown in Table 3. Invariably, unadjusted, or “crude” analyses suggested that most exposures and experiences in theater—from bagging sand to hearing chemical alarms—were significant risk factors for Gulf War illness. But relatively few significant risk factors were identified after adjustments were made for the effects of multiple exposures, as demonstrated in Table 3 and detailed in Appendix A.

The Committee was concerned and somewhat surprised to find that many Gulf War epidemiologic studies had not accounted for the high degree of confounding introduced by the complex Gulf War exposure scenario. As a result, some studies involving impressive population samples and data collections actually reported that nearly all of the exposures in the Gulf War appeared to be significant risk factors for chronic Seven Gulf War population-based studies systematically evaluated exposure/illness relationships using analyses that adjusted for effects of multiple exposures in theater. These included the CDC study of Air Force veterans,1124 two studies of Army veterans from the northeastern U.S. who returned from the war through Fort Devens, Massachusetts,1239,1804 large studies of British Gulf War veterans241 and U.S. Navy Seabees,527 and studies that assessed neurological and gastrointestinal symptoms in Danish veterans.695,1507

In addition, two studies evaluated a limited number of individual exposures while adjusting for effects of one or two other selected exposures, as opposed to controlling for confounding in a more comprehensive way. These included a large study of Gulf War veterans from the states of Washington and Oregon1466 and the study of Navy Seabees from the 24th Naval Mobile Construction Battalion.564 Studies that assessed illness-exposure relationships using statistical methods that accounted for effects of multiple exposures were generally considered the most informative by the Committee. Moreover, the limited number of risk factors for Gulf War illness identified by these studies were surprisingly consistent.

Information bias: Misclassification. An additional source of error that can occur in epidemiologic studies stems from inaccurate classification of the exposures and/or health outcomes being assessed. This is a particular concern in Gulf War studies, which have usually relied on self-reported information for both exposures and health status. As a result, recall bias—the tendency for individuals to recall or report information inaccurately—has the potential to be particularly problematic in Gulf War research.

No external, objective validation is possible for most veteran-reported exposures. However, several studies have assessed the reliability with which veterans report exposures using test-retest methods.988,1165,1767,1804 Overall, veterans have reported some exposures more reliably than others.

Generally, the most reliably reported exposures were those that veterans experienced first hand and were unique to the war, including encountering smoke from oil well fires, taking pyridostigmine bromide, and having a SCUD missile explode nearby. Lower, but fair reliability was associated with exposure to substances such as pesticides and fuels, and hearing chemical alarms. Exposures about which veterans might have had little first-hand knowledge at the time of exposure, such as exposures to depleted uranium and CARC paint, were reported least reliably. 988,1165,1767

Questions about the accuracy of veterans’ self-reported exposures require that identified risk factors for Gulf War illness be assessed and interpreted with caution. Errors resulting from misclassification can produce both overestimates and underestimates of the degree to which a particular exposure is actually associated with illness. It is useful to note that studies of both U.S. and U.K. Gulf War veterans have found that the reliability of self-reported exposures was unrelated to veterans’ health status, that is, symptomatic veterans report exposures with the same degree of reliability as healthy veterans.988,1767,1804

This indicates a potential for “nondifferential” misclassification of exposures, that could lead to underestimates of the degree of risk resulting from some Gulf War exposures, particularly those reported less reliably.

Unlike exposures, Gulf War studies have generally found that veterans report medical conditions with a high degree of reliability.692,751,789,989 For example, medical record reviews for a subset of veterans participating in the U.S. national survey of Gulf War era veterans indicated that self-reported conditions related to clinic visits and hospitalizations were reported accurately 93 percent of the time.751

Studies have assessed the impact of reporting biases on epidemiologic findings in Gulf War studies using different approaches.692,988,1088,1165 One recent study, for example, reported that veterans in VA’s national survey who had been notified that they were potentially exposed to nerve agents following weapons demolitions at Khamisiyah, Iraq, were no more likely to report symptoms, medical conditions, or healthcare visits than other veterans. Investigators concluded that, contrary to expectation, veterans who believed they may have been exposed to nerve gas showed no tendency to “over report” health problems.1165 A study of Gulf War veterans in the Pacific Northwest found that media coverage of both the Khamisiyah weapons demolitions and studies showing adverse effects of Gulf War exposures had very little impact on veterans’ reports of chemical agents and other exposures in theater.988 Iowa investigators reported that Gulf War veterans were no more likely to respond to health questions in a socially desirable way than nondeployed era veterans.692 And in a study of the Fort Devens cohort, inclusion of a measure of recall bias in multivariable analyses had no impact on identified associations between exposures and Gulf War illness.1804

Taken together, such studies suggest that despite obvious concerns related to the potential effect of recall bias on Gulf War studies, its actual impact does not appear to have been extensive enough to render study results uninformative. Still, the potential for error introduced by recall bias and other sources of information bias is an important reason for considering patterns of results provided by multiple studies, rather than relying on individual studies, especially when assessing relationships between experiences in theater and Gulf War illness. Gulf War Illness Prognosis and the Need for Treatments Are Veterans with Gulf War Illness Getting Better or Worse With Time? The question of whether veterans with Gulf War illness have generally recovered or become worse is an important one. Four studies have assessed the health of Gulf War veterans over time, all leading to the same conclusion. In 1998, investigators from the Boston VA Environmental Hazards Center reported that veterans in the Fort Devens cohort, evaluated at two time periods between 1992 and 1996, exhibited no significant differences in either the types or average number of symptoms reported.1239 When veterans from the same group were evaluated a third time two years later, 90 percent of those who had previously been identified as CMI cases continued to meet defining criteria for CMI.1804

Similarly, a study of over 1,000 British Gulf War veterans found that their symptomatic ill health remained relatively stable over time. In two evaluations, four years apart, British Gulf War veterans exhibited a slight worsening of functional status, but improved slightly on measures of fatigue and psychological distress.644 Declining health was most associated with veterans’ having more severe symptoms at baseline, believing they had “Gulf War Syndrome,” and having more psychological distress.643 Recently, New Jersey investigators also reported little change in the health of symptomatic Gulf War veterans over time. Among nearly 400 U.S. Gulf War veterans surveyed in both 1995 and 2000, no significant changes in the average number or severity of symptoms were found. Veterans who had been highly symptomatic in 1995 remained so in 2000, although as a group they experienced a slight reduction in symptoms.1163

Additional insights into the development and prognosis of Gulf War illness were provided by preliminary results from VA’s longitudinal study of nearly 6,000 Gulf War veterans, presented to the Committee by Dr. Han Kang.745,748 In this national sample, 35 percent of Gulf War veterans indicated they had developed multisymptom illness since the war, with most (67%) reporting that onset occurred between 1991 and 1993. Only two percent of those who had developed multisymptom illness said they had since recovered. Seven percent felt they were “much improved” but 15 percent indicated their condition had become “much worse” over time.

Results from all longitudinal Gulf War studies clearly indicate that few veterans with Gulf War illness have recovered over time and only a small minority have substantially improved. Studies also indicate that the majority of symptomatic Gulf War veterans have not become progressively worse with time. However, a subgroup of veterans do appear to have become worse in the years since they first became ill.

The Urgent Need for Effective Treatments for Gulf War Illness Gulf War illness has persisted for a very long time for most ill veterans—seventeen years for many. Special panels and government committees assembled to address questions related to the health of Gulf War veterans have consistently emphasized the importance of providing adequate treatments for affected veterans. But effective treatments for Gulf War illness have not yet been identified. The federal government has sponsored just three completed clinical trials to study treatments for Gulf War illness, only two of which have published study results. In addition, many thousands of ill veterans have been seen for this condition in government and private healthcare settings in the 17 years since the war. But few systematic evaluations have reported on the degree to which the treatments veterans receive have been useful in improving their health. The Committee’s 2004 report indicated that the federal government had spent over 21 million dollars for treatment research up to that time, the majority ($15 million) for two large multi-center clinical trials. Additional funding was provided for an unpublished The only other completed Gulf War illness clinical trial was a study of a complex, high dose antibiotic regimen conducted by the Louisiana Medical Foundation, headed by the late Dr. Edward Hyman.670 The intervention was unconventional, and the theory on which it was based was controversial.331,671,672,1454,1455 Study results were never published, but were presented to the Committee by two of Dr. Hyman’s coinvestigators, Dr. Quentin Deming and Mr. William Weiss. Briefly, the study was a randomized, double blind, placebo controlled trial of intravenous, then oral antibiotics over a four month period. Specific regimens and dosages varied, according to the presence of excreted gram-positive cocci detected by microscopic evaluation of patients’ urine, and by patients’ symptoms.332,670

Although both the theory and intervention were unconventional, investigators used standard methods to evaluate the health status of veterans before and after treatment. Results shared with the Committee indicated that the treatment group improved significantly compared to the placebo group, with reductions in the mean number of headaches per month (from 12.5 to 2.5, p < 0.001), significantly improved scores on two fatigue scales, and improvement on the McGill Pain Inventory. The median SF36 PCS score was reported to improve 22 points for the treatment group, compared to seven points for the placebo group, and investigators indicated that no excess of side effects had been observed in the treatment group. No significant differences were seen on measures of sleep quality, neuropsychological impairment, or frequency of diarrhea.332 The Committee was intrigued by the apparent benefit provided by the treatment, but concerned that study results had not been scientifically peer reviewed and published. The biological rationale for the treatment approach was also puzzling. So although the empirical results appeared extremely promising they were overshadowed by questions surrounding the study, most prominently the role of excreted bacteria and the lack of scientific review and successful publication. Therefore, the Committee was unable to come to firm conclusions regarding the meaning and importance of the study findings and appropriate follow up.

There are few other sources of systematically-collected data on the effects of treatments used for Gulf War illness. Two investigators have published observational findings on treatment outcomes in case series of ill Gulf War veterans, as shown in Table 5. Dr. Garth Nicolson reported substantial benefit for a subset of Gulf War veterans treated with multiple courses of antibiotics,1118,1119 and Dr. Charles Engel reported slight functional improvement in veterans treated with a multidisciplinary intervention that included CBT.406

Gulf War veterans with multisymptom illness who participated in VA’s national longitudinal study were asked about their experience with treatments and lifestyle practices in relation to their symptoms. Preliminary findings were presented to the Committee by Dr. Han Kang.745 Symptomatic veterans reported using prescription and over-the-counter medications most frequently, followed by physical therapy and nutritional supplements. The most highly rated category was over-the-counter medication, which eight percent of ill veterans said had provided benefit for their symptoms, most prominently headache and joint pain. About the same proportion indicated that diet and nutritional supplements had helped, mostly for fatigue, joint pain, and gastrointestinal symptoms. Six percent reported physical therapy had helped with somatic pain and five percent indicated that antidepressants had been helpful for improving depression symptoms and sleep difficulties. Among unconventional therapies, about two percent of symptomatic veterans reported that relaxation therapy had been helpful for joint pain, fatigue, and headache. A similar number indicated that herbal medicines had provided benefit for memory loss, fatigue, and joint pain.

Veterans also reported whether different activities and lifestyle behaviors had affected their symptoms. Factors most often associated with improved symptoms were avoiding stressful situations (25%), maintaining a well-balanced diet (20%), and cutting back on work or social activities (18%). The factors Gulf War Illness Prognosis and the Need for Treatments  39 most often said to make symptoms worse were vigorous exercise (35%) and maintaining a busy schedule (23%). About the same number of veterans indicated that light exercise improved (16%) as worsened (18%) their symptoms. These findings provide an interesting first look at the general types of approaches veterans have used in addressing their illness. The Committee looks forward to reviewing additional results from this study to learn, in more detail, about veterans’ appraisals of specific treatments.

No other systematically-collected data are available on effects of treatments for Gulf War illness. Two physicians have provided public testimony on their clinical experience in treating a limited number of veterans. In 1993, Dr. Myra Shayevitz provided testimony to Congress describing improvements in 25 symptomatic Gulf War veterans treated in an environmental clinic piloted at the Northampton, Massachusetts VA Medical Center (VAMC).1399 The clinic intervention included reduced exposures to chemicals, improved nutrition, and patient education and support. Several of Dr. Shayevitz’s patients also provided written comments attesting to their improved health. Dr. David Root provided testimony to the Presidential Special Oversight Board in 1998 and to the CDC Gulf War Research Planning Conference in 1999 about dramatic improvements observed in several highly symptomatic Gulf War veterans he had treated with an intensive sauna/detoxification regimen used routinely for treatment of chemical injury and substance abuse.1307,1308

VA’s Gulf War research portfolio currently includes three clinical studies that provide treatments for symptomatic Gulf War veterans. A study conducted at the East Orange, New Jersey, VAMC is evaluating the effectiveness of CBT administered by telephone to veterans with Gulf War illness.226 A second study, conducted at the Northport, New York, VAMC, is evaluating continuous positive airway pressure (CPAP) treatment for Gulf War veterans with disordered sleep. The third study, conducted by investigators at the Salt Lake City VAMC, will treat small bowel bacterial overgrowth in veterans with persistent diarrhea.

In addition, VA and DOD collaborated in convening expert panels that developed clinical guidelines for evaluating veterans with post-deployment health concerns,1656 and for evaluation and management of veterans with medically unexplained fatigue and pain.1655 Treatment guidelines for medically unexplained symptoms were based on what was known about treatment of fibromyalgia and chronic fatigue syndrome at the time the guidelines were developed in 2001. No information is available that indicates whether government clinicians have used these guidelines in treating ill Gulf War veterans, or if recommended treatments have been effective. The treatment guidelines have also become outdated. Since 2001 a large amount of additional information has become available on medical treatments for these conditions, particularly fibromyalgia, as will be described in a later chapter.

Future prospects for federally-sponsored Gulf War illness treatment research. As described in the Committee’s 2004 report, there are two general approaches for identifying effective therapeutic interventions. The first, an empirical approach, is based on clinical observations that certain treatments provide improvements for certain conditions. Potentially beneficial treatments identified in this way can be systematically assessed using outcomes research and randomized clinical trials to scientifically determine their effectiveness. The second approach requires that specific biological mechanisms underlying a disease be identified, so that treatments to counteract those processes can be identified and tested for their effectiveness. For Gulf War illness, a complex condition for which specific pathophysiological mechanisms are not well understood, both approaches will likely be needed in order to identify the most effective treatments in the most timely way.

In response to recommendations in the Committee’s 2004 report, the Secretary of Veterans Affairs announced that VA would fund a Gulf War illness treatment research initiative, largely focused on identifying and evaluating treatments already available and being used to treat Gulf War illness and conditions with similar features. Although a draft funding announcement for a treatment research center was provided for Committee review in late 2005, no final announcement was released and a treatment research center has not been funded.

In 2006, two major changes occurred in federal funding for Gulf War illness research, as will be described in detail in a later section. These changes included a total of 15 million dollars allocated in FY2006 and FY2008 for a Gulf War illness research program managed by the Office of Congressionally Directed Medical Research Programs (CDMRP) at DOD,1596 and a 15 million dollar annual allocation for a comprehensive Gulf War illness research center at the University of Texas Southwestern (UTSW), funded by VA. The two recently-funded programs have been directed to coordinate their efforts and will, fundamentally, utilize the two approaches previously described for identifying effective treatments. The initial funding solicitation issued by the CDMRP Gulf War illness research program indicated that highest priority would be given to studies that identify and evaluate treatments for Gulf War illness. This included funding for smaller scale studies to provide data on treatments currently being used for Gulf War illness and similar conditions as well as treatments that address biological processes thought to underlie Gulf War illness. The UTSW program, on the other hand, is focused on determining specific biological mechanisms that underlie veterans’ symptoms, in order to identify treatments to address those processes. Both programs have only recently begun implementing studies, and the Committee looks forward to monitoring their progress. The CDMRP program announced, in 2007, that nine Gulf War illness studies were funded with the initial program allocation. These included pilot trials of treatments for veterans with Gulf War illness, and animal studies that will evaluate effects of treatments on biological processes identified in animal models for Gulf War illness.737 The Committee regards both programmatic initiatives to be positive steps forward in focusing Gulf War research on the highest priority objective, that is, to improve the health of ill veterans. Is There a Unique Gulf War Syndrome? The question of whether the multisymptom illness affecting Gulf War veterans should be considered a “unique Gulf War Syndrome” has been widely discussed and interpreted.134,252,324,556,667,668,686,918,1089,1757 What is meant by the question has often been unclear, as have attempts to answer it. For some observers, a “unique syndrome” has meant that there should be just one constellation of symptoms affecting Gulf War veterans—a single symptom complex constituting a single syndrome. For others, a “unique syndrome” has meant that a single, unique cause for the symptoms should be demonstrated. For still others, a “unique syndrome” has meant that similar symptoms would not be found in people who did not serve in the Gulf War. And for several researchers, the question has hinged on whether a particular statistical technique, factor analysis, identifies symptom correlations in Gulf War veterans that are not found in other groups.

However the question of a unique syndrome in interpreted, extensive descriptive and analytic research has clearly demonstrated that an illness, characterized by a complex of multiple symptoms, resulted from service in the Gulf War. The specific symptoms affecting individual veterans can differ from person to person, but the general types of symptoms are remarkably consistent across diverse Gulf War veteran populations. Whether this Gulf War-related symptom complex represents several syndromes, or one syndrome with several subtypes, is an issue of taxonomy that can only be definitively resolved as objective markers become more firmly established.

Gulf War illness, as a consistent complex of symptoms affecting a defined population, fits most definitions of what constitutes a syndrome. But this syndrome might not be considered unique, from different perspectives. That is, there could be more than one type of pathophysiological process affecting Gulf War veterans that leads to similar, overlapping symptom profiles. There could also be more than one cause for these symptoms. And, lastly, Gulf War illness has some similarities to multisymptom conditions found in other populations, as will be discussed in detail in a later section of this report.

The central issue of importance is that at least one fourth of veterans who deployed to the Gulf War as healthy men and women developed an identifiable pattern of persistent, difficult symptoms as a consequence of their military service. Whether this illness should be referred to as one or more syndromes—unique or otherwise—is of less consequence. There is overwhelming evidence demonstrating that Gulf War illness, however labeled, is a widespread problem in Gulf War veterans and no evidence to the contrary.

Is Gulf War illness the same thing that happens after every war? Several commentaries and reviews have described Gulf War illness as a condition that parallels syndromes historically described in soldiers after they return from war.669,720,721,1500 These have included “irritable heart” or “Da Costa’s syndrome” in Civil War veterans,302 shell shock and “effort syndrome” in World War I veterans, battle fatigue in World War II veterans, and posttraumatic stress disorder in Vietnam veterans. In all eras, soldiers serving in war have suffered from acute and chronic health problems that often affect more troops than the number injured and killed in battle. This has historically included the effects of infectious disease and extreme environmental conditions, but in more recent times has also included effects of radioactive fallout, chemical defoliants, and chemical weapons.191

Experiences common to all wars include combat and the hardships of deployment, both of which can have long-term physical and psychological consequences. Commentators who have characterized Gulf War illness in the context of other post war syndromes have suggested—explicitly or implicitly—that because the psychological impact of war can have long-term consequences, Gulf War illness is probably another post-war stress syndrome, the result of psychological factors. This idea was accepted by some at face value before data that specifically addressed these issues became available. Research studies have not supported the view that Gulf War illness is the same type of problem that occurs after every war, nor that it can accurately be considered a post-war stress syndrome. As early as 1994, a National Institutes of Health Technology Assessment panel observed that symptom profiles affecting Gulf War veterans differed from those of Vietnam veterans. Data from VA registries indicated that symptoms of fatigue, muscle pain, headache, joint pain, and shortness of breath were more common in Gulf War than Vietnam veterans.1121 British investigators have since systematically evaluated the health and symptoms of military personnel who served in the 1990-1991 Gulf War, in Bosnia during the 1990s, and in Iraq in the current conflict. No “Gulf War syndrome”-like effect, that is, no pattern of excess symptoms affecting a sizable number of veterans, was found in Bosnia or Iraq War veterans.631,642,1698 The effect was only observed in veterans who served in the 1991 Gulf War.1088

Clinical reports on U.S. veterans who served in Operations Iraqi Freedom and Enduring Freedom also have indicated that returning personnel have not been affected by high rates of symptomatic illness that is not explained by diagnosable medical or psychiatric conditions.615,653

In contrast to Vietnam veterans and personnel returning from current conflicts in the Middle East, population-based studies have consistently found that 1990-1991 Gulf War veterans have low rates of posttraumatic stress disorder and other psychiatric conditions, as detailed in the next section of the report. Further, studies that have comprehensively assessed risk factors associated with the Gulf War consistently indicate that Gulf War illness is not associated with serving in combat or other stressors during deployment. Available evidence therefore indicates that Gulf War illness is not the same thing that happens after every war and is not a post-war stress syndrome. Each war is unique, each has its own profile of risks and health consequences.291,1216,1723 All wars present some degree of trauma for troops in battle, but many wars also present other hazards. The effects of blister agents in World War I or the Iran-Iraq War, for example, should not be equated to the psychological consequences of soldiers fearing for their lives or seeing a buddy die on the battlefield. Neither should the effects of Agent Orange be confused with the effects of the traumatic experiences many soldiers encountered in the jungles of Vietnam. Service in the 1991 Gulf War resulted in a complex health problem not typical of other wars that cannot be understood simply as the expected result of deployment-related stress. Other Gulf War Health Issues Gulf War illness is the most prevalent health problem affecting Gulf War veterans, but not the only health issue related to Gulf War service. Additional important issues include rates of diagnosed medical and psychiatric conditions in Gulf War veterans, particularly neurological conditions, cancers, and respiratory diseases, as well as causes and rates of mortality. Although Gulf War epidemiologic studies have commonly reported hospitalization and mortality rates, relatively little information is available concerning diagnosed diseases not normally associated with hospitalization or premature death. In addition, important questions about health problems affecting veterans’ children and other family members have persisted since the Gulf War.

Diagnosed Diseases Affecting Gulf War Veterans Amyotrophic lateral sclerosis. The most serious condition reported to affect Gulf War veterans at a higher-than-expected rate is amyotrophic lateral sclerosis, also known as ALS or Lou Gehrig’s disease. This serious and progressive neurodegenerative disease most often strikes individuals between age 55 and 75, affects men more often than women, and is almost universally fatal. A 2003 VA study reported that Gulf War veterans were about twice as likely to have ALS as nondeployed era veterans based on 40 confirmed Gulf War-deployed ALS cases.636 The excess risk was particularly pronounced in Air Force Gulf War veterans, who had ALS at nearly three times the rate of their nondeployed peers.

The VA research team made a concerted effort to determine whether the excess ALS rate observed in Gulf War veterans could be an artifact of ascertainment error, that is, failure to detect some ALS cases among the nondeployed. After adjusting for this potential bias using three different methods, results still indicated a significant excess of ALS in Gulf War veterans.260,633 Research from the University of Texas Southwestern raised additional concern that Gulf War veterans may have developed ALS at a youngerthan- normal age, finding that a large number of cases occurred in veterans under age 45.557 In addition, military hospitalization data indicated that active duty personnel who had served in the Gulf War had a 1.7 times higher rate in ALS hospitalizations between 1991 and 1997, compared to nondeployed era veterans, an excess that was not statistically significant.1432

A later report from a 2005 study of over 400,000 men in an American Cancer Society cohort indicated that men who had served in the military, overall, were more likely to have died of ALS than men who were not in the military.1759 This raised questions about whether an excess risk of ALS is related to military service in general, rather than Gulf War service specifically.1206,1310 As a result, VA commissioned a special report from the Insitute of Medicine, which concluded that there was limited, but suggestive evidence that ALS is associated with military service in general.685

Results of the Cancer Society study are important in providing a preliminary indication that military service could be a risk factor for ALS. But it is unclear why researchers and government officials have suggested, based on findings from this study, that ALS may be linked to military service, but not specifically with Gulf War service. The VA Gulf War ALS study found that ALS affected Gulf War veterans at twice the rate of nondeployed Gulf War era military personnel. If military personnel are, overall, at increased risk for ALS, the observed excess of ALS in Gulf War veterans compared to other military personnel would be of particular concern.

The Cancer Society study provided information on ALS among military veterans serving from World War II through the Vietnam eras, but no insights on rates of ALS in deployed vs. nondeployed veterans or in Gulf War veterans compared to veterans of other eras.1759 Therefore, results of this study do not deployment. This excess could be of greater concern if military service in general is also a risk factor for ALS.

Recently, additional findings reported from the large VA ALS study indicated that most new ALS cases among Gulf War veterans identified in the 10 years after the war had their initial onset by 1996. The excess of ALS cases declined after that time—both in Gulf War veterans overall, and in those under age 45.634,635 Additional analyses also identified differences in ALS risk related to geographical areas where troops were located during deployment.1052,1053 These recent reports indicate that ALS in Gulf War veterans occurred in the pattern of a time-limited disease “outbreak,” resulting from events or exposures during Gulf War deployment. If the post-1996 pattern of new onset ALS cases continues, the number of excess ALS cases among Gulf War veterans will be less than had initially been suggested by early studies. But it is not known if the risk of ALS, which normally increases after age 55, will differ in Gulf War veterans as they age. The seriousness of this disorder requires that ALS rates in Gulf War veterans continue to be monitored for the foreseeable future.

In response to early reports that ALS was associated with Gulf War service, VA developed an ALS registry for Gulf War era veterans. That registry has since been expanded to include all veterans with ALS who served in the military during any period.756 In addition, VA has developed a brain tissue bank that will enroll and collect tissues from veterans with ALS identified in the registry.441,1225 Other neurological diseases. Very little information is available concerning rates of other diagnosed neurological diseases in Gulf War veterans. In light of the excess of ALS in Gulf War veterans, as well as consistent findings related to persistent neurological symptoms, it is important to determine if other neurological diseases have disproportionately affected Gulf War veterans. In its 2004 report, the Committee recommended that rates of multiple sclerosis (MS), Parkinson’s disease, brain cancers, and difficult-to-characterize neurological disorders be identified in Gulf War veterans and suitable comparison groups. Since the Committee’s report was issued, veterans’ organizations and members of Congress have called on the federal government to conduct research to determine the rate of MS in Gulf War veterans.1694,1721 In 2008, VA initiated a case/control study of veterans who were service-connected for MS disability by the Veterans Benefits Administration (VBA).1747 This study will not identify incidence or prevalence rates of MS in Gulf War veterans but may provide insights concerning characteristics of MS and risk factors for MS potentially related to Gulf War service.

Other than limited information from hospitalization studies, the only other studies that have assessed neurological disease in Gulf War veterans evaluated rates of mortality due to neurological disease. A 2005 study, conducted by investigators from the Washington, D.C., VAMC, identified an excess rate of brain cancer deaths among Gulf War veterans who, according to DOD models, were potentially exposed to low levels of nerve agents in relation to chemical weapons demolitions at Khamisiyah, Iraq, in 1991.192 Veterans in affected areas were twice as likely, overall, to have died from brain cancer between 1991 and 2000 as veterans in other locations. Excess rates were most apparent during the last few years of follow up (1997-2000). A dose-response effect was also noted, wherein higher brain cancer mortality occurred in veterans who were in affected areas for longer periods of time.190,192

Researchers are currently conducting an updated mortality study to evaluate causes of death in U.S. Gulf War veterans through 2004.105 Preliminary results, shared with the Committee in 2008, are similar to findings reported in 2005. Investigators continue to identify a significant excess of brain cancer deaths among Gulf War veterans potentially exposed to nerve agents related to the Khamisiyah demolitions. These mortality studies provide useful information on deaths due to brain cancer, and demonstrate the importance of evaluating diseases in subgroups of Gulf War veterans with specific exposure and/or location histories. However, other types of research are still needed to determine whether Gulf War service is associated with excess rates of diagnosed neurological diseases that have not been fatal. Cancer in Gulf War veterans. Government committees and special panels have long called for studies to determine if Gulf War veterans have developed cancer at higher-than-expected rates since Desert Storm.1227,1673,1682 Identifying cancer rates in Gulf War veterans is especially important now, 17 years after the war, since many cancers first become apparent 10 to 20 years after an initiating event. The most comprehensive study of cancer in Gulf War veterans comes from Great Britain. A 2003 study identified the incidence of multiple types of cancer between the years 1991 and 2002 in the entire cohort of U.K. Gulf War veterans and a matched comparison group, using data from the British National Health Service.943 No differences were found between Gulf War and era veterans for rates of all cancers combined, nor for any site-specific cancers.

In the absence of a similar cancer data resource in the United States, comprehensive information on cancer rates in U.S. Gulf War veterans has not been reported. As previously described, results from a national study found an excess of brain cancer deaths in relation to the Khamisiyah weapons demolitions. The 2005 study identified 25 brain cancer deaths in veterans potentially exposed to nerve agents, an excess of 14 brain cancer deaths per 100,000 exposed veterans.190 In contrast, in a population-based survey of about 1,800 Gulf War veterans in five U.S. states, no excess of physician-diagnosed cancer was reported by veterans who had been within 50 kilometers of Khamisiyah. Overall, however, three times as many Gulf War veterans as nondeployed era veterans in this sample reported being diagnosed with some type of cancer. The excess of reported cancer diagnoses—21 cases among Gulf War veterans, and three cases among nondeployed veterans—did not reach statistical significance.989

Only limited information is available concerning verified cancer diagnoses in U.S. Gulf War era veterans. An early hospitalization study reported that, in the months immediately following Desert Storm, active duty Gulf War veterans were twice as likely to be hospitalized for testicular cancer as nondeployed era veterans.523 This difference was no longer apparent after five months, leading investigators to conclude that the temporary rate spike had been due to Gulf veterans deferring care for this condition until they returned home from deployment.820

A later study, using 1991-1999 data from cancer registries in New Jersey and the District of Columbia (D.C.), reported a two-fold proportional excess of testicular cancer in Gulf War veterans, compared to nondeployed era veterans.894 Proportional excesses were also reported for non-Hodgkin’s lymphoma and brain cancer from D.C. registry data, but not the New Jersey registry. This team has continued to collect and analyze cancer data on Gulf War and era veterans from additional state cancer registries. Preliminary results from a total of eight registries were shared with the Committee by Dr. Paul Levine. Data from some states suggested slight excesses in the crude incidence of testicular and brain cancers in Gulf War veterans compared to nondeployed era veterans between 1991 and 1999. Proportional differences were not significant, however, after adjustments for age and race.892 This ongoing investigation currently includes data from 28 state cancer registries, which cover about 83 percent of U.S. Gulf War and era veterans.959 The Committee looks forward to reviewing additional results from this important research.

Although studies to date have raised only limited concerns about cancer in Gulf War veterans, a number of important questions have not yet been adequately addressed. Very limited cancer data have been reported for U.S. Gulf War veterans in general, and no published research on cases occurring after 1999. Because of the extended latency periods associated with most cancers, it is important that cancer information be brought up to date and that cancer rates be assessed in Gulf War veterans on an ongoing basis. In addition, cancer rates should be evaluated in relation to identifiable exposure and location subgroups, as was done in the 2003 British study and the U.S. mortality study related to Khamisiyah. Data from VA’s longitudinal study can also be used to provide an indication of whether veteran-reported cancers are associated with exposures in theater.

Other diagnosed conditions affecting Gulf War veterans. Limited information is available on rates of other diagnosed diseases in Gulf War veterans. In addition to symptoms, epidemiologic studies have asked veterans to report if they had been diagnosed with a variety of medical conditions. Several types of diagnoses are consistently reported at higher rates by Gulf War veterans than nondeployed era veterans. These include migraines,527,751,1476,1698 seizures,748,751,989,1476 digestive conditions,527,748,751,1411,1476 respiratory conditions,527,751,1411,1476,1698 and skin disorders.527,751,1411,1476,1698 Generally, fewer excess medical conditions have been reported by Australian and Danish Gulf War veterans than U.S. and U.K. Gulf War veterans.696,789

Rates of respiratory conditions have been evaluated in several Gulf War studies.285,866,1434 As will be described in more detail, one study reported that a higher proportion of Gulf War than nondeployed veterans had been hospitalized for respiratory conditions, including asthma.528 In addition, one wellconducted study found that the subset of Gulf War veterans with greatest exposure to pollutants from oil well fires had significantly elevated asthma rates.285

Multiple studies have evaluated rates of diagnosed psychiatric conditions in Gulf War veterans. Gulf War veterans generally have higher rates of posttraumatic stress disorder and other psychiatric diagnoses than nondeployed Gulf War era veterans,1488 but lower rates of psychiatric illness than combat veterans of other wars. Findings on psychological stressors and psychiatric conditions are described in detail in the next section of the report.

Most of the studies that have provided clinical examinations of Gulf War veterans have either included a relatively small number of veterans464,565 or were case/control studies160 and so could not provide reliable prevalence estimates for diagnosed conditions. Prevalence rates of selected diagnosed medical conditions were provided in 2005 from Phase III of the U.S. National Survey of Gulf War era veterans.393 This portion of the large U.S. national study provided clinical evaluations of 1,061 Gulf War veterans and 1,128 nondeployed era veterans 10 years after the war. Reported outcomes included SF36 PCS scores and 12 medical conditions: fibromyalgia, chronic fatigue syndrome, skin conditions, dyspepsia, hypertension, hepatitis, symptomatic arthralgias, obstructive lung disease, diabetes, peripheral neuropathy, and both hypo- and hyperthyroidism.

Results from the U.S. national study indicated that Gulf War veterans had a dramatically higher rate of chronic fatigue syndrome than nondeployed veterans (1.6% vs. 0.1%, OR = 40.6), and significantly higher rates of fibromyalgia (2.0% vs.1.2%, OR = 2.3), skin conditions (34.6% vs. 26.8%, OR = 1.4), and dyspepsia (9.1% vs. 6.0 %, OR = 1.9). None of the other 12 conditions were significantly more common in Gulf War veterans. On average, the general health status of Gulf War veterans, measured by the SF36 PCS, was only slightly worse in Gulf War than nondeployed veterans (49 vs.51). This difference was statistically significant, but of minor clinical significance. Abnormalities identified on clinical examination and mean values for all laboratory tests were also similar for Gulf War and nondeployed veterans.393

These long-awaited findings from the large VA clinical study provided useful information about the 12 conditions assessed but few additional insights concerning the health of Gulf War veterans. It is not clear why the 12 outcomes assessed were selected for evaluation, since many had not been shown in earlier phases of the study to be problematic for Gulf War veterans. For conditions like diabetes and hepatitis, clinical evaluations largely provided validation of what veterans had already reported. Regrettably, this large clinical study has not provided information on many of the conditions found to affect Gulf War veterans at excess rates in earlier phases of the study, conditions like recurrent headaches and migraines, diarrhea and colitis, seizures, and sinusitis. Neither was information provided on other medical conditions of interest such as cancers, autoimmune disorders, and heart disease. Mean values for deployed and nondeployed veterans were reported for all measures, but no information was provided on subgroups of potential interest, for example, subgroups of veterans with abnormal findings on laboratory tests, and subgroups of veterans who reported specific exposures. So, while additional insights have been provided from VA’s Phase III clinical study, important questions remain about the extent to which Gulf War veterans may be disproportionately affected by diagnosed medical conditions. It is important to determine if Gulf War veterans and, in particular, subgroups of Gulf War veterans with specific exposures during the war, have excess rates of diagnosable neurological conditions, cancer, respiratory diseases, or other chronic diseases.

Mortality Rates Among Gulf War Veterans The question of whether there is an abnormally high rate of death among Gulf War veterans, or if veterans have died at younger-than-expected ages, is of great interest and importance. In the seventeen years since Desert Storm, government reports and research studies from both the U.S. and the U.K. have consistently indicated that Gulf War veterans have not, overall, had higher rates of death due to diseases but have had higher rates of accident-related deaths than nondeployed era veterans. Overall mortality rates in both deployed and nondeployed era veterans are lower than in the general population, however.

Post-war mortality statistics are available from three published studies of U.S. Gulf War veterans and two studies of British veterans, along with regular mortality reports provided by the U.K. Ministry of Defence.

The most recent published information on mortality in U.S. Gulf War and nondeployed veterans reports on deaths through 1997, identified by VA’s Beneficiary Identification and Records Locator Subsystem (BIRLS) and the Social Security Administration, with causes of death identified by the U.S. National Death Index.749 That study reported that early post-war figures indicating lower disease-related mortality in Gulf War veterans, and higher accident-related mortality, had become more similar over time. By 1996-1997, rates of mortality resulting from both disease and accidents were nearly identical in deployed and nondeployed veterans. Later information has been published on mortality rates through the year 2000 for Gulf War veterans only, in relation to the Khamisiyah plume models, as previously described.192

Preliminary findings from an ongoing mortality study conducted by investigators at the Washington, D.C.,VAMC were shared with the Committee in 2008.105 That study is evaluating overall and causespecific deaths that occurred among U.S. Gulf War and nondeployed era veterans through 2004. Early results indicate that, overall, Gulf War veterans continue to have a lower mortality rate due to diseases, and a higher mortality rate due to accidents, than nondeployed era veterans. However, investigators reported that female Gulf War veterans have significantly greater mortality, overall, than nondeployed female era veterans, and excess deaths due to digestive system diseases and external causes, including motor vehicle accidents. Preliminary findings also continue to indicate that brain cancer mortality is elevated among Gulf War veterans in relation to modeled levels of exposure to nerve agents. These preliminary findings are provocative, and the Committee looks forward to further reviewing results of this important study as they are finalized.

Mortality rates among British Gulf War veterans through 2006 have shown trends similar to those observed in U.S. veterans. Over time, excess rates of accident-related deaths identified in the years just after the war have become more comparable to those of nondeployed veterans. In a recent report, however, the U.K. Ministry of Defence reported that between 1991 and 2007, veterans of the 1991 Gulf War had a higher rate of suicide, or possible suicide, than nondeployed veterans of the same era.1569 Overall rates of death due to diseases remained somewhat lower in Gulf War veterans, compared to nondeployed era veterans.1569 Additional details of interest are provided by a report on mortality in British veterans in relation to experiences/exposures during the war.944 Just two associations were identified, neither of which reached statistical significance. Veterans who reported handling pesticides during the war were twice as likely as unexposed veterans to die from accident-related causes, and veterans who reported depleted uranium exposure were twice as likely to die from disease-related causes. A number of theories have been put forward to explain why Gulf War veterans have experienced higher rates of fatal accidents, most prominently motor vehicle accidents.124 These have included indications that veterans have a greater propensity for risk taking behavior after hostile deployments,750 findings of poorer attention and response speed in cognitively impaired veterans,554 reports of greater use of alcohol by combat veterans,908 and the general similarities between characteristics of deployed military personnel and people with the highest rates of motor vehicle accidents in the general population.629,908 Mortality studies have provided little indication that Gulf War veterans, overall, have suffered excess rates of deaths due to diseases. However, the most recent comprehensive comparisons between U.S. Gulf War and nondeployed veterans that have been published only include deaths that occurred before 1998.

Deaths due to diseases with longer latency periods would likely only have become apparent in more recent years. Therefore, it is important that current figures for overall mortality, as well as diseasespecific mortality, for U.S. Gulf War era veterans be comprehensively evaluated and made publicly available. Information on disease-specific mortality rates during the past 10 years are of particular importance, and the Committee urges VA to make this information available at the earliest possible time. Additional information on mortality rates among subgroups of Gulf War veterans—defined, for example, by exposures and locations in theater and by branch of service—is also needed to determine if Gulf veteran subgroups have been affected by any causes of death not apparent when all veterans are assessed as a single group.

Hospitalization Rates Among Gulf War Veterans Between 1996 and 2006, 14 studies reported rates of hospitalization in Gulf War veterans and comparison groups.526,1428 Nearly all of these studies were limited to information on active duty military personnel who were admitted to military hospitals. They therefore do not include the vast majority of Gulf War veterans or hospital admissions. Recent VA figures indicate that over 90 percent of Gulf War veterans had left the military by 2007.1650

Few differences between Gulf War and nondeployed era veterans have been reported from Gulf War hospitalization studies. Both all-cause hospitalization rates and disease-specific hospitalizations have been similar, overall, in comparisons between active duty Gulf War and nondeployed military personnel from the same era. The few exceptions come from just three studies. The first reported that Gulf War veterans were hospitalized for fibromyalgia at significantly excess rates between 1991 and 1997, but not for lupus. Findings on ALS hospitalizations during this period were inconclusive due to small numbers, as previously described.1432 No more recent information concerning hospitalizations for these conditions has been reported. A second study found that a higher proportion of Gulf War Marine Corps veterans than Vietnam Marine veterans were hospitalized for musculoskeletal conditions.148 The third study included hospitalization information from nonmilitary hospitals. That study included 1991-1994 national data from DOD and VA hospitals, as well as civilian hospitals in the state of California. Results indicated a higher proportion of Gulf War than nondeployed era veterans had been hospitalized for injuries and for respiratory and digestive diseases.528 Excess hospitalizations due to cardiac dysrythmia were also reported among active-duty personnel who were, according to DOD models, potentially exposed to lowlevel nerve agents in relation to the Khamisiyah weapons demolitions.1433 Modeled exposure to pollutants from oil well fire smoke was not associated with increased hospitalization risk.1434

As discussed in the Committee’s 2004 report, the large majority of cases of Gulf War illness would not be identified using hospitalization data, since it is extremely uncommon for patients with undiagnosed, symptom-defined illness to be hospitalized. There is also little reason to expect that a number of other types of conditions reported to affect Gulf War veterans at excess rates would be identified by the hospitalization studies conducted to date. Nearly all studies report only on hospitalizations among active duty personnel in military hospitals. Veterans with serious conditions that might lead to hospitalization, but who were no longer in the military, would not have been included in the studies. In addition, medical conditions shown by some studies to have affected subgroups of Gulf War veterans affected by a particular exposure, such as asthma and brain cancer, would potentially go undetected in hospitalization studies that simply compare all deployed veterans to nondeployed veterans. Further, diseases with long latency periods, potentially detectable at their later stages using hospital admission data, would not likely be found in studies evaluating hospital admissions before 2000, the most recent year for which Gulf veteran hospitalization data have been reported.

An enormous amount of effort and care have been used to analyze and report military hospitalization rates in Gulf War veterans. Results of these studies have been reassuring, to some degree, by indicating that Gulf War veterans have not been admitted to military hospitals at exceedingly high rates in conjunction with the types of injuries and acute and chronic diseases that normally lead to hospitalization. Gulf War hospitalization studies have largely been used to report on disease rates that are easiest to quantify using data routinely collected for administrative purposes. Unfortunately, this “low hanging fruit” is not particularly informative with respect to the types of health problems known or expected to be of greatest concern for Gulf War veterans. Consequently, hospitalization studies have added little to our understanding of health issues related to Gulf War service. It is possible that hospitalization data may be more informative in future years, if diseases of long latency that require hospitalization emerge in sufficient numbers. It will be important, however, that any future studies of hospitalization rates in Gulf War veterans include nonmilitary hospitalizations, and determine disease-specific rates in relation to Gulf

War veteran subgroups of interest. Birth Defects and the Health of Gulf War Veterans’ Family Members In addition to issues related specifically to the health of veterans, concerns have persisted since the mid 1990s that veterans’ family members, particularly their children born after the war, have had health problems related to some aspect of veterans’ Gulf War service. These issues were reviewed and discussed in detail in the Committee’s 2004 report, including results of studies conducted to assess rates of birth defects in veterans’ children. Since that time, findings from a large VA study that evaluated spouses of Gulf War veterans have been published, providing a first look at whether veterans’ spouses have been affected by excess health problems in the wake of Desert Storm.

Birth defects in children of Gulf War veterans. In 1995, a cover story in Life magazine reported on several children, born to Gulf War veterans, who had serious birth defects including Goldenhar Syndrome, a congenital disorder characterized by abnormal development of facial structures.171 This, along with reports of birth anomalies in a National Guard unit that had served in Desert Storm,1194 raised public concern and stimulated research to determine whether children born to Gulf War veterans had abnormally high rates of birth defects. As discussed in the Committee’s 2004 report, early studies found little evidence of a problem284,285,1194 but had important limitations relating both to the samples and sources of data used. Later studies used larger and/or more representative samples of Gulf War veterans, and more comprehensive methods to identify health problems in children under one year. These studies did find that a limited number of adverse birth outcomes, though rare, occurred more commonly in Gulf War veterans than nondeployed veterans.

A study of over 75,0000 children born in military hospitals between 1991 and 1993 indicated that infants born to Gulf War veterans were about three times more likely to have Goldenhar syndrome-related diagnoses than infants born to nondeployed veterans.56 This excess was not statistically significant, however, because the total number of cases in both Gulf War and nondeployed veterans was extremely small. The first indication of a significant excess of birth defects related to Gulf War service came from a 2001 report from VA’s large national survey of Gulf War era veterans. Study results indicated that children born to male Gulf War veterans after the war had twice the rate of “likely” birth defects as children born to nondeployed era veterans. Children born to female Gulf War veterans had three times the rate of “likely” birth defects.747 Because these data relied on veterans’ self-reports, investigators conducted medical record reviews to evaluate diagnoses for veteran-reported birth defects where possible.

These reviews, conducted for two-thirds of reported birth defects, confirmed veterans’ reports in 88 percent of cases. Resulting adjusted estimates continued to indicate that children of Gulf War veterans had significantly more birth defects than children of era veterans.744

A large British survey of Gulf War veterans also reported a significant excess of veteran-reported birth defects among children conceived between 1991 and 1997 by male Gulf War veterans, compared to nondeployed veterans. Birth defects affecting the musculoskeletal and genitourinary systems were most prominent.361 For the subset of birth defects confirmed by medical records, excess rates were similar but less pronounced. Both the U.S. and U.K. national studies have therefore suggested that birth defect rates were higher in children of Gulf War veterans than children of nondeployed veterans, but fell within the normal range expected in the general population.

Results of an impressive data collection effort by the U.S. Naval Health Research Center also indicated an excess of birth defects in children of Gulf War veterans. This study linked Gulf War military service information to 1989-1993 data from six states with active birth defects surveillance programs.57 Results indicated that three types of birth defects were significantly more common in children born to Gulf War veterans, conceived after the war. Children of male veterans had higher rates of two types of heart valve defects—tricuspid valve insufficiency and aortic valve stenosis. Male children of female Gulf War veterans were more likely to be born with hypospadias, a defect in the urethral opening. In contrast, there were similar rates of birth defects in children of Gulf War and nondeployed veterans who had been conceived before the war.

Studies have also reported other adverse pregnancy outcomes in relation to Gulf War service. Military hospital data revealed a significant excess of ectopic pregnancies and spontaneous abortions among women Gulf War veterans whose pregnancies were conceived soon after their return from theater.55 In addition, male Gulf War veterans in both the large U.S. and U.K. Gulf War surveys reported higher rates of miscarriages, but not still births, in pregnancies they had fathered.361,747 British Gulf War veterans were also reported to have higher rates of infertility than nondeployed veterans.949

Few additional studies related to pregnancy outcomes have been reported in the years since the Committee’s 2004 report. A postal survey collected data on pregnancy outcomes between 1991 and 1995 reported by over 4,000 U.S. Gulf War and nondeployed era veterans. No significant excess of low birth weight, ectopic pregnancies, stillbirths, or miscarriages were reported for male or female Gulf War veterans, when analyzed separately.1761 Similar results were reported from a postal survey of Australian Gulf War veterans.791 Neither study provided information on birth defects, however. It is difficult to draw firm conclusions related to birth defects and pregnancy outcomes in Gulf War veterans, due to the diversity and limitations of study results reported to date. The three studies most representative of Gulf War era veterans in the U.S. and U.K. have all indicated significant, but modest, excess rates of birth defects in children of Gulf War veterans. Information on specific types of birth defects has been inconsistent, however,362 and overall rates are still within the normal range found in the general population.

Some of the remaining important questions concerning birth defects in children of Gulf War veterans might be answerable using existing data. For example, differences in specific types of birth defects reported in different studies might relate to effects of combining all deployed Gulf War veterans into a single group, rather than analyzing birth defects in relation to characteristics of the veteran parents’ deployment or health. Birth defect rates, if related to veterans’ service in the Gulf War, could be most pronounced in identifiable subgroups of veterans, for example, veterans with multisymptom illness, veterans who were in certain areas of theater, or those exposed to certain hazardous substances. Birth defects might also have been more problematic during certain periods after veterans returned, for example, in pregnancies conceived soon after Desert Storm, as opposed to more recent conceptions.

Identifying patterns and risk factors for birth defects in defined populations can be extremely challenging, particularly for birth defects that are uncommon.57 In addition to strategies aimed at obtaining additional information from existing data, other research approaches will be needed to determine if birth defects might be associated with Gulf War service generally, or with specific aspects of Gulf War service. This could include case-control studies to evaluate Gulf War service and specific parental exposures as risk factors for extremely rare types of birth defects.54 A study of this type recently reported that Gulf War service was not a significant risk factor for new cases of Goldenhar Syndrome between 1996 and 2002, although military service in the Army was a modest risk factor.1764

Another innovative approach for assembling and evaluating data on birth defects was presented to the Committee by Ms. Betty Mekdeci. Ms. Mekdeci directs Birth Defect Research for Children (BDRC), a private nonprofit organization that maintains special registries of children with birth defects, including children of Gulf War veterans. The analytic approach of the organization involves comparing proportional patterns of birth defects in different populations, in order to raise hypotheses about potential problems in a given group. BDRC has identified a number of problems that appear to disproportionately affect the over 3,000 children of Gulf War veterans in their birth defect registry. This includes 33 children with Goldenhar syndrome—substantially more cases than had been identified in the large military hospital study. BDRC data also indicates that the majority of identified children with Goldenhar Syndrome born to Gulf War veterans were born in 1992 and 1993, with fewer cases born after 1993.1020

Health problems in other family members. Media reports have also suggested that family members and close contacts of Gulf War veterans have experienced anomalous health problems since veterans returned from Desert Storm.130,1055,1250 Suggested causes have included transmissible infections or contamination by chemical substances brought home on veterans’ uniforms and gear. A 1994 report from the U.S. Senate Banking Committee indicated that many of the 1,200 ill veterans interviewed reported that family members had developed health problems similar to their own.1688 In response to these reports, VA provided free medical examinations to family members of Gulf War veterans who were enrolled in the Gulf War Registry. No information from VA’s Gulf War family registry program has ever been issued, however. Research studies have provided some information on the health of veterans’ family members, but have been limited to studies of birth defects among infants and the recent study on veterans’ spouses. Research on rates of diagnosed diseases, symptomatic illness, and learning and behavioral disorders among older children of Gulf War veterans is needed in order to determine whether they have been affected by excess health problems, as has been suggested by media and veterans’ reports and by the 1994 Senate investigation.

The large national U.S. study of Gulf War veterans included, in Phase III, clinical evaluations of a sample of 539 spouses of Gulf War veterans and 600 spouses of nondeployed Gulf War era veterans. Standardized medical, psychiatric, and neuropsychological examinations were performed ten years after the war at 16 VA medical centers throughout the U.S. Nearly ninety percent of spouses evaluated in the study were women. Health problems self-reported by Gulf veterans’ spouses were very similar to those of nondeployed veterans’ spouses, except that Gulf veterans’ spouses were significantly more likely to report having skin rashes (28%) and hepatitis (1%) than nondeployed spouses. There were no significant differences between the two groups on medical examination, however, except that Gulf veterans’ spouses had significantly fewer “group 1” or mild skin anomalies, such as moles, skin tags, and scars. There were no significant differences in rates of fibromyalgia or chronic fatigue syndrome in veterans’ spouses. Nor were there differences in diagnosed conditions such as diabetes, lung diseases, or hepatitis. Functional status, as measured by the SF36 PCS, was also nearly identical in the two groups.394

The long-anticipated results of this important study thus indicated that, overall, the health of spouses of Gulf War veterans was similar to that of spouses of veterans who did not serve in the Gulf War. These results are reassuring, in some measure. But additional information is needed before the question of Gulf War illness, or other health problems in family members, can be laid to rest. As with Gulf War veterans, the most prominent remaining questions about the health of veterans’ family members relate to undiagnosed symptoms and symptom complexes. Specifically, are symptoms or groups of symptoms more common in spouses of Gulf War veterans than nondeployed veterans? Are higher rates of symptoms or diagnosed conditions experienced by spouses of veterans with Gulf War illness? And are any health problems in veterans’ spouses associated with characteristics of veterans’ service in the Gulf War, such as veterans’ locations, experiences, or exposures in theater? The majority of these questions should be answerable using data already collected for the Phase III study.

Phase III of the U.S. national study also included clinical examinations of children of Gulf War and nondeployed era veterans. Results have not yet been reported, but are of great interest and importance. Reported information should include rates of symptoms and symptom complexes in veterans’ children, as well as comprehensive information on diagnosed medical and behavioral conditions. Comparisons should also be made between health outcomes in children of veteran subgroups of interest, as described previously. The Committee urges investigators to complete and publish results of the children’s evaluations, as well as additional results from the spouses’ evaluations, as soon as possible. Special Committee and Government Reports on the Health of Gulf War Veterans In the seventeen years since Desert Storm, numerous government committees and specially-appointed expert panels have been assembled to investigate the health problems affecting Gulf War veterans and/or the government’s response to these problems. Relatively few scientific studies were available to inform the conclusions of early panels. Their reports routinely called for more research, specifically epidemiologic studies, to better characterize the health of Gulf War veterans. As described throughout the present report, many studies of the types recommended by previous panels have now been completed, allowing a more comprehensive evaluation of Gulf War-related health issues.

In 1994, the Defense Science Board Task Force on Persian Gulf War Health Effects reported that “veterans in the hundreds have complained of a range of symptoms not yet explained by any clear-cut diagnosis” and indicated that research was needed to determine if these complaints were precipitated by service in Desert Storm.1595 In the same year, a panel convened at the National Institutes of Health recommended that comprehensive epidemiologic studies be undertaken to better characterize health problems affecting Gulf War veterans and their causes.1121 The Senate Banking Committee also issued reports in 1994 that detailed their investigations of chemical exposures in the Gulf War and unexplained health problems affecting veterans and their families. This report also called for in-depth epidemiologic investigations to determine the nature and causes of veterans’ conditions.1688

The Presidential Advisory Committee on Gulf War Veterans’ Illnesses, a panel of scientists and veterans appointed by President Clinton, issued reports in 1996 and 1997 that recommended additional research to characterize veterans’ health problems. The panel indicated that research was needed on effects of individual and combined chemical exposures, and physical responses to stress.1227 Similarly, reports issued by the Institute of Medicine (IOM) during this period called for additional research focused on priority questions about the health of Gulf War veterans and emphasized the importance of coordinating data collection efforts between the federal agencies involved in this effort.675,676

Perception of Gulf War veterans’ unexplained health problems and federal efforts to address them changed markedly when DOD announced, in 1996, that demolition of Iraqi munitions caches at Khamisiyah, Iraq, in March of 1991 had potentially exposed thousands of U.S. troops to low levels of the nerve agents sarin and cyclosarin. The Department of Defense established the Office of the Special Assistant for Gulf War Illnesses (OSAGWI), which initiated an extensive series of investigations, and commissioned the RAND Corporation to provide scientific reports on specific topics of concern. Special House and Senate committees undertook investigations and issued comprehensive reports detailing their findings.1684,1690 Federal research conferences were held to highlight emerging results from scientific studies on the health of Gulf War veterans. At the direction of Congress, the U.S. General Accounting Office (GAO) investigated diverse Gulf War health and programmatic issues, issuing multiple reports that evaluated the status of the federal response and gaps that had not been adequately addressed. A second committee, the Presidential Special Oversight Board (PSOB) for Department of Defense Investigations of Chemical and Biological Incidents, was appointed by President Clinton in 1998. The PSOB issued its final report in 2000, providing general support for DOD’s investigations of exposures during the Gulf War, but again calling for additional scientific research to better characterize the relationship of toxic exposures to Gulf War illness.1232

The Institute of Medicine’s Gulf War and Health reports. In 1998, with few conclusive answers to continuing questions about Gulf War illness and the federal response to this problem, Congress directed VA to contract with the National Academy of Sciences (NAS) to review available research in order to assist the Secretary of Veterans Affairs in making decisions about Gulf War-related disability compensation. Public Laws 105-277 and 105-3681242,1243 directed that this review identify conditions that affect Gulf War veterans at excess rates and assess the scientific evidence concerning associations between those conditions and a detailed list of Gulf War exposures. In response, VA commissioned the Institute of Medicine (IOM), of the National Academies, to conduct a series of reviews using a methodology previously established to evaluate diseases affecting Vietnam veterans in relation to Agent Orange.663

To date, the resulting Gulf War and Health series has included nine reports, including two updated reports, and provided hundreds of conclusions.177,679,682-689,1740 The Committee was concerned to find that the IOM reviews were not conducted in accordance with the laws that mandated them. As a result, the Gulf War and Health reports have provided little information that is directly relevant to health conditions that affect Gulf War veterans at excess rates, or their association with Gulf War exposures.

The 1998 legislation specifically directed that VA commission reviews that identify both diagnosed and undiagnosed illnesses that affect Gulf War veterans at excess rates and, based on a comprehensive consideration of available research, determine whether there is evidence that those illnesses are associated with Gulf War exposures or Gulf War service. However, the health conditions considered in the IOM Gulf War and Health reports have primarily included multiple types of cancer and a number of other diagnosed diseases—conditions for which there are no indications that Gulf War veterans have been affected at excess rates. In contrast, the IOM reports have provided almost no information on conditions that do occur at excess rates in Gulf War veterans. That is, the Gulf War and Health reports have not provided findings on possible associations between Gulf War illness or ALS and most Gulf War exposures. Nor do they provide findings on conditions like migraines and seizures, which preliminary information suggests may affect Gulf War veterans at excess rates, in relation to Gulf War exposures.

The legislation also directed that determinations be based on scientific evidence provided by both human and animal studies. Most studies that evaluate biological effects of hazardous exposures are done in animals, for ethical reasons. In recent years, a large number of animal studies have identified biological effects of Gulf War exposures and combinations of exposures that were previously unknown. Although animal research was sometimes described in the IOM reports, findings from animal studies were not considered in drawing conclusions about the evidence that Gulf War exposures were associated with health outcomes. Unlike IOM’s earlier Agent Orange reports, the standards used to determine levels of evidence for the Gulf War and Health reports were expressly limited to consideration of results from human studies.137,678,679 The omission of animal studies was especially striking in IOM’s updated report on sarin, which had been requested by the Secretary of Veterans Affairs in 2003 specifically because of new research in animals that demonstrated adverse effects of low-level sarin exposure.683,1641

A very limited number of exposure-disease associations have been identified in the IOM reviews. For example, in Volume 3 of the series, the IOM panel concluded that there is sufficient evidence to indicate that lung cancer is associated with petroleum combustion products.684 Findings of this type might potentially be relevant to the health of Gulf War veterans in future years. But there has been no indication that lung cancer, or the vast majority of conditions considered in the IOM Gulf War and Health reports, have affected Gulf War veterans at excess rates. The hundreds of findings provided in the IOM reports are largely inconclusive, indicating that there is insufficient evidence to determine if the diseases considered are associated with the exposures considered, based on the types of studies considered.

The specific information included in the Gulf War and Health reports is also problematic, in that it appears to reflect a process of reporting selected results from subgroups of studies, rather than integrating and analyzing results from all available research. This is a pervasive problem, but several examples are illustrative. A very prominent example relates to the limited or complete lack of consideration, in all Gulf War and Health reports, of results from the many epidemiologic studies that have assessed associations between Gulf War exposures and Gulf War multisymptom illness. Another straightforward example comes from Volume 4, which reported the rate of multisymptom illness in Gulf War veterans from just one study, as opposed to the seven studies identified by the present report. The one Gulf War illness prevalence estimate provided was atypical, and substantially lower than all other studies.686 An additional example relates to a highly publicized finding that, although Gulf War veterans have multiple excess symptoms, there is no unique Gulf War syndrome.232,686 This conclusion was based solely on several studies that had unsuccessfully attempted to identify a unique syndrome using factor analysis and a related statistical technique, as previously described. The finding did not consider basic questions about whether the statistical techniques were capable of identifying syndromes—unique or otherwise.

Unfortunately, this conclusion was widely misinterpreted in media reports to indicate that there was no widespread problem associated with multisymptom illness in Gulf War veterans.

In short, IOM’s Gulf War and Health series of reports have been skewed and limited by a restrictive approach to the scientific tasks mandated by Congress, an approach directed by VA in commissioning the reports. These limitations are most notably reflected in the selective types of information reviewed and the lack of in-depth analysis of the research literature and scientific questions associated with the health of Gulf War veterans. There is a fundamental disconnect between the Congressional directive to VA and VA’s charge to IOM for reviewing evidence on Gulf War exposures and their association with illnesses affecting Gulf War veterans. The reports have particularly fallen short in advancing understanding of associations between Gulf War exposures and Gulf War illness, the most prominent health issue affecting Gulf War veterans. Recommendations Despite the brief duration and successful execution of the 1990-1991 Gulf War, 25-32 percent of Gulf War veterans developed the chronic multisymptom condition known as Gulf War illness as a consequence of their Gulf War service. Longitudinal studies indicate that few veterans with Gulf War illness have recovered or significantly improved with time. The Committee gives highest priority to research focused on identifying effective treatments for Gulf War illness. This research should include:

Studies that identify and systematically evaluate the effectiveness of currently available treatments used for Gulf War illness or conditions with similarities to Gulf War illness. Preliminary research should include pilot trials and/or observational studies capable of identifying promising treatments suitable for evaluation in larger clinical trials.

Research to identify specific pathophysiological mechanisms underlying Gulf War illness that are potentially amenable to treatment interventions.

Research to evaluate novel therapies based on scientific findings as they emerge. The Committee considers the information provided by VA’s national longitudinal study of Gulf War veterans and continued monitoring of the health of Gulf War veterans over time to be extremely important and recommends that VA:

Make results from the national longitudinal study of Gulf War veterans publicly available at the earliest possible time, including comprehensive findings related to multisymptom illness, treatments and practices used by veterans to address their symptoms, and rates of medical diagnoses. Results should include outcomes assessed according to the guidelines for epidemiologic research provided below.

Continue to monitor health and disease outcomes among veterans assessed in the National Survey of Gulf War Era Veterans and Their Families, conducting longitudinal surveys and appropriate clinical follow-up studies at five year intervals.
Although it is the most prevalent health problem affecting Gulf War veterans, Gulf War illness is just one of a number of important Gulf War health issues. To provide needed information on other health issues of concern for Gulf War veterans, the Committee recommends the following research:

Epidemiologic research to identify rates of diagnosed neurological diseases (including multiple sclerosis, Parkinson’s disease, amyotrophic lateral sclerosis, and brain cancers), as well as central nervous system abnormalities that are difficult to precisely diagnose, in Gulf War veterans and appropriate comparison groups.

Completion of current research comparing cancer rates in Gulf War and nondeployed era veterans, and repeated assessment of cancer rates in Gulf War era veterans at regular intervals.

Provide current information on overall and cause-specific mortality rates in Gulf War veterans, and update this information, at minimum, at five year intervals. This should include information on mortality in subgroups of Gulf War veterans identified by deployment locations, branch of service, and exposures reported in the National Survey of Gulf War-era Veterans and Their Families.

Further evaluate indications of possible increased risk of specific types of birth defects, and other health problems in children of Gulf War veterans, using innovative study designs.

That VA make available comprehensive information on family members of Gulf War veterans from the national study of Gulf War era veterans and family members. This should include information on diagnosed conditions, multisymptom illness, behavioral problems, and birth defects. Health parameters should also be assessed in subgroups of interest, such as family members of veterans with/without Gulf War illness, and subgroups defined by Gulf War exposures and other characteristics of veterans’ wartime service.

Because of shortcomings and limitations in many epidemiologic studies of Gulf War veterans, the Committee recommends the following principles for collecting and analyzing data on Gulf War illness and the health of Gulf War veterans in ongoing and future studies and, where indicated, for reanalyzing data in studies already completed. Studies of Gulf War veterans should use well-constructed and clearly-described case definitions for Gulf War illness and illness subgroups. Pending more widespread acceptance of an established case definition, preferred case definitions are those that most clearly distinguish the pattern of symptoms in Gulf War veterans from those in nondeployed era veterans, such as the Kansas Gulf War illness case definition. In addition to general comparisons between Gulf War and nondeployed veterans, Gulf War research studies should analyze results in relation to Gulf War veteran subgroups of interest, including ill vs. well veterans and subgroups defined according to veterans’ locations in theater, exposures, and other military and deployment characteristics potentially relevant to the outcomes evaluated. Associations between deployment-related exposures and health outcomes in Gulf War veterans should be evaluated using analytic methods that appropriately control for the effects of confounding introduced by multiple exposures during deployment.

The Department of Veterans Affairs has not adhered to requirements set forth by Congress in commissioning the Gulf War and Health series of reports produced by the Institute of Medicine. As a result, these reports have not addressed fundamental questions regarding Gulf War-related health conditions and their relation to Gulf War exposures. The Committee therefore recommends: That VA, in commissioning reports mandated by Congress in PL 105-277 and 105-368, substantially change the approach designated for reviewing scientific information and preparing the reports. As directed by Congress, these reports should address both diagnosed and undiagnosed illnesses affecting Gulf War veterans. Conclusions should be based on findings from the full range of Gulf War epidemiologic studies, animal studies, and other research that provides information on effects of Gulf War-related exposures. That VA contract with the Institute of Medicine to redo previously completed Gulf War and Health reports to adhere to requirements set forth by Congress. That responsibility for contracting reports mandated by PL 105-277 and PL 105-368 be reassigned from VA’s Office of Public Health and Environmental Hazards to another office within VA, to be designated by the Secretary.

Veteran Services Transparency

VST’s Letter to the Veteran’s Administration

AGENT ORANGE A LETHAL LEGACY: For Vietnam War Veterans, Injustice Follows Injuries

Your Latest Blog Post here

The Health of Gulf War Veterans

FINDINGS IN BRIEF

EXECUTIVE SUMMARY:

INTRODUCTION:

TABLE OF CONTENTS:

CHAPTER 1: Gulf War Illness and the Health of Gulf War Veterans