United States General Accounting Office GAO Report to the Chairmen and Ranking Minority Members of the Senate Committee on Armed Services and the House Committee on National Security June 1997 GULF WAR ILLNESSES Improved Monitoring of Clinical Progress and Reexamination of Research Emphasis Are Needed GAO/NSIAD-97-163 United States GAO General Accounting Office Washington, D.C. 20548 National Security and International Affairs Division B-276835 June 23, 1997 The Honorable Strom Thurmond Chairman The Honorable Carl Levin Ranking Minority Member Committee on Armed Services United States Senate The Honorable Floyd Spence Chairman The Honorable Ronald Dellums Ranking Minority Member Committee on National Security House of Representatives Many of the approximately 700,000 veterans of the Persian Gulf War have complained of illnesses since the war’s end in 1991. Some fear they are suffering from chronic disabling conditions because of wartime exposures to one or more agents with known or suspected health effects. This report responds to the mandate of the fiscal year 1997 defense authorization act that we analyze the effectiveness of the government’s clinical care and medical research programs relating to illnesses that members of the armed forces might have contracted as a result of their service in the Gulf War.1 Specifically, we evaluated (1) the Department of Defense’s (DOD) and the Department of Veterans Affairs’ (VA) efforts to assess the quality of treatment and diagnostic services provided to Gulf War veterans and their provisions for follow-up of initial examinations, (2) the government’s research strategy to study the veterans’ illnesses and the methodological problems posed in its studies, and (3) the consistency of key official conclusions with available data on the causes of veterans’ illnesses. During their deployment associated with the Persian Gulf War, U.S. troops Background might have been exposed to a variety of potentially hazardous substances. These substances include compounds used to decontaminate equipment and protect it against chemical agents, fuel used as a sand suppressant in and around encampments, fuel oil used to burn human waste, fuel in 1 Our response to the referenced legislation resulted in two additional studies: Defense Health Care: Medical Surveillance Has Improved Since the Gulf War, but Results in Bosnia Are Mixed (GAO/NSIAD-97-136, May 13, 1997) and a classified report issued earlier this year on biological agent defense. Page 1 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 shower water, leaded vehicle exhaust used to dry sleeping bags, depleted uranium, parasites, pesticides, drugs to protect against chemical warfare agents (such as pyridostigmine bromide), and smoke from oil-well fires. DOD acknowledged in June 1996 that some veterans may have been exposed to the nerve agent sarin following the postwar demolition of Iraqi ammunition facilities. Shortly after the war, some veterans began reporting health problems that they believed might be due to exposure to chemicals, pesticides, and other agents used during the war. Accordingly, both DOD and VA established programs through which Gulf War veterans could receive medical examinations and diagnostic services. From 1992 to 1994, VA participants received a regular physical examination with basic laboratory tests, and in 1994, VA established a standardized examination to obtain information about exposures and symptoms related to diseases endemic to the Gulf region and to incorporate specific tests to detect the “biochemical fingerprints” of certain diseases. If a diagnosis was not apparent, the standard examination protocols provided for up to 22 additional tests and provided for additional specialty consultations. If the illness defied diagnosis, VA registrants might have been sent to one of four VA Persian Gulf referral centers. DOD initiated its Comprehensive Clinical Evaluation Program in June 1994. It was primarily intended to provide diagnostic services similar to the VA program and employed a similar clinical protocol. However, the VA program was among the first extensive efforts to gather data from veterans regarding the nature of their problems and the types of hazardous agents to which they might have been exposed. (See app. I for details.) Our review found that (1) although efforts have been made to diagnose Results in Brief veterans’ problems and care has been provided to many eligible veterans, neither DOD nor VA has systematically attempted to determine whether ill Gulf War veterans are any better or worse today than when they were first examined; (2) while the ongoing epidemiological research will provide descriptive data on veterans’ illnesses, formidable methodological problems are likely to prevent researchers from providing precise, accurate, and conclusive answers regarding the causes of veterans’ illnesses; and (3) support for some official conclusions regarding stress, leishmaniasis, and exposure to chemical agents was weak or subject to alternative interpretations. Page 2 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 Over 100,000 of the approximately 700,000 Gulf War veterans have participated in DOD and VA health examination programs established after the war. Based on the examinations and reports provided by DOD and VA, nearly 90 percent of the examined veterans are symptomatic, reporting a wide array of health complaints and disabling conditions. While VA and DOD health examination programs have sought to evaluate these veterans’ problems and refer eligible veterans for further care, neither DOD nor VA currently has mechanisms in place to determine whether these ill veterans are any better or worse today than when they were first examined. Both agencies have tried to measure or ensure the quality of veterans’ initial examinations. While some measures of quality are in place for military or VA health care in general, neither agency can now determine the appropriateness or effectiveness of the treatment received by ill Gulf War veterans.2 Federal research on Gulf War veterans’ illnesses has not been pursued proactively. Although these veterans’ health problems began surfacing in the early 1990s, the vast majority of research was not initiated until 1994 or later. And, much of this research was associated with legislation or external reviewers’ recommendations. Thus, although at least 91 studies have received federal financial support, about four-fifths of the funded studies are not complete, and certain studies will not be available until after 2000. Some hypotheses (for example, that veterans’ current symptoms are due to stress) were pursued earlier and more aggressively than others (for example, that symptoms are due to low-level exposure to chemical warfare agents), and some hypotheses that were initially unfunded by the federal government (for example, that symptoms are due to the delayed chronic effects of exposure to organophosphates,3 which were in pesticides used in the Gulf) were pursued with private sector funding. In recent years, VA and DOD have significantly broadened their research programs, to include efforts to seek external advice. Without accurate exposure information, the investment of millions of dollars in further epidemiological research on the risk factors (or potential causes) for veterans’ illnesses may result in little return. The government’s research has primarily involved epidemiological studies, most of which focus on the nature and prevalence of the veterans’ symptoms and illnesses or the identification of causes for the illnesses. While mortality 2 We are conducting further work addressing medical care provided to Gulf War veterans. See VA Health Care: Observations on Medical Care Provided to Persian Gulf Veterans (GAO/T-HEHS-97-158, June 19, 1997). 3 Organophosphates are a class of chemicals found in some pesticides and chemical warfare agents. Page 3 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 information and data on the prevalence of various problems may be valuable, because of formidable methodological problems facing investigators, epidemiological research on Gulf War veterans’ illnesses will not be able to provide precise, accurate, and conclusive answers regarding the causes of veterans’ illnesses. Specifically, studies generally are hampered by the lack of (1) accurate, person-based, dose-specific exposure data; (2) known biological markers (such as detectable antibodies to specific agents or diseases); and (3) specific case definition (definition of particular syndromes or clusters of symptoms to study). While some prevalence data may be useful, we agree with the Institute of Medicine and the Presidential Advisory Committee on Gulf War Veterans’ Illnesses.4 that population-based comparisons that group together veterans with varied exposures may mask higher rates of illness among veterans with specific exposure histories. The plans for toxicological research on the health effects of low-level exposures to various agents will be useful in efforts to determine whether veterans’ current unexplained symptoms or conditions are consistent with such exposure. To date, the research program has not included an assessment of the clinical progress of ill Gulf War veterans, which is critical to identifying the appropriateness and effectiveness of their treatment and could be useful to provide direction to the research agenda. Six years after the war, little is conclusively known about the causes of Gulf War veterans’ illnesses. Not only were few strong, conclusive statements made in the executive branch reports we reviewed, but support was weak or subject to alternative interpretation for three conclusions made by the Presidential Advisory Committee and endorsed by DOD. In addition, two questions remain unresolved. First, the Committee concluded that stress is likely to be an important contributing factor to the broad range of illnesses currently being reported by Gulf War veterans and that studies have found higher rates of posttraumatic stress disorder (PTSD) in Gulf War veterans. However, the link between stress and these veterans’ physical symptoms is not well established in the evidence the Committee cited, and the reported prevalence of PTSD among Gulf War veterans may be overestimated because of problems in the methods used in studies to identify it (for example, there were frequent failures to exclude physical causes for 4 See Presidential Advisory Committee on Gulf War Veterans’ Illnesses, Final Report (Washington, D.C.: GPO), December 1996. Page 4 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 veterans’ symptoms or to conduct structured clinical interviews, which are necessary components of PTSD diagnosis). Second, based on a small number of diagnosed cases, VA and DOD concluded that the likelihood of leishmania tropica (a parasite) as an important risk factor for widely reported illness has diminished and the Committee found it unlikely to be “responsible for long term health effects in Gulf War veterans.” However, the extent of asymptomatic leishmania infection is unknown, and the possibility of prolonged latency and apparent clinical dormancy (up to 20 years) of an infection that may reemerge in the presence of immune deficiency underscores the need to retain leishmania among the potential risk factors. Third, the Committee concluded that it was unlikely that the health effects reported by Gulf War veterans are the results of exposure to organophosphate or mustard chemical warfare agents, even though there is substantial evidence that organophosphate compounds might be associated with delayed or long-term health effects similar to those experienced by the Gulf War veterans. Unresolved questions concern the extent to which veterans may have been exposed to (1) chemical agents as a result of fallout from the destruction of suspected chemical weapons storage sites and (2) the biological agent aflatoxin, the health effects of which may not be known for months, or even years, after exposure. DOD and VA officials have testified that whatever uncertainties may exist DOD and VA Have No about the cause of Gulf War veterans’ illnesses, the veterans are receiving Systematic Approach appropriate and effective symptomatic treatment. However, DOD and VA to Monitoring Gulf have no mechanism to monitor the quality, appropriateness, or effectiveness of care provided to Gulf War veterans after their initial War Veterans’ Health examination. Furthermore, DOD and VA officials said they had no plans to After Initial establish a mechanism to monitor these veterans’ progress. This monitoring and follow-up is important not only to ensure that diagnosed Examination conditions are properly treated but also because (1) undiagnosed signs and symptoms are not uncommon among ill veterans, (2) treatment for veterans with undiagnosed conditions is based on their symptoms, and (3) veterans with undiagnosed conditions or multiple diagnoses may see multiple providers. These agencies have relied on such mechanisms as training and standards for physician qualification, which may not be sufficient to ensure a given level of effectiveness for the care provided or Page 5 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 do not permit identification of the most effective treatments.4 In contrast, some steps have been taken to monitor quality and patient satisfaction with veterans’ initial registry examinations. (See app. II for details.) The bulk of ongoing federal research currently focuses on the Federal Research epidemiological study of veterans’ illnesses. While this approach may yield Strategy Lacks a descriptive data on veterans’ mortality and general health profiles, Coherent Approach methodological problems facing government epidemiological research on Gulf War veterans’ illnesses will severely limit its ability to identify the potential causes of the illnesses. Initially, the government was not proactive in acknowledging and collecting data on the factors that might have caused Gulf War veterans’ health problems, and the research agenda was not articulated until several years after the war ended. Delays and Focus of Our review of research projects and interviews with agency officials Federal Research Are showed that the vast majority of federal research was not initiated until Hindering Outcomes 1994. This 3-year delay has complicated the task facing researchers. In addition, it has limited the amount of completed research currently available. Of the 91 federally sponsored studies, 72 were ongoing when we reviewed them in early 1997, and some of the studies will not be complete until 2000 or later. The focus of federal research has primarily been the epidemiological study of the prevalence and cause of Gulf War illnesses rather than the diagnosis, treatment, and prevention of them. With respect to determining the causes, researchers will likely continue to find it difficult to detect effects of particular wartime exposure and to eliminate alternative explanations for Gulf War veterans’ illnesses because of the absence of valid and reliable data on exposures and the multiplicity of agents to which the veterans were exposed. Data on the prevalence of various health problems can be useful but requires careful interpretation in the absence of better information on the factors to which veterans were exposed. While multiple studies of the role of stress in the veterans’ illnesses have been supported with federal research dollars, basic toxicological questions regarding the substances to which they were exposed remain unanswered. Finally, there is an absence of efforts to measure Gulf War veterans’ clinical progress. This leaves the government unable to promptly determine the quality and effectiveness of treatments 4 See VA Health Care: Observations on Medical Care Provided to Persian Gulf Veterans (GAO/T-HEHS-97-158, June 19, 1997). Page 6 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 currently being provided to Gulf War veterans or to use this information when funding additional clinical research. Methodological Problems Federal researchers studying Gulf War illnesses have faced a number of Limit the Effectiveness of challenges and encountered significant problems in linking exposures or Ongoing Epidemiological potential causes to observed illnesses or symptoms. Research • Researchers have found it extremely difficult to gather information about unplanned exposures in the Gulf to such things as oil fire smoke and insects carrying infection, and DOD has acknowledged that records of the use of pyridostigmine bromide and vaccinations to protect against chemical/biological warfare exposures were inadequate. • Gulf War veterans were typically exposed to a wide array of agents, making it difficult to isolate and characterize the effects of individual agents or to study their combined effects. • Most epidemiological studies have relied only on self-reports for measuring most of the agents to which veterans may have been exposed during the Gulf War. • The passage of time following these exposures has made it increasingly difficult to have confidence in any information gathered about them through retrospective questioning of veterans. Reliance on self-reporting to assess exposures has two problems. Veterans’ recall after such a long time period may be inaccurate or biased. Moreover, there is often no straightforward way to test the validity of self-reported exposure information, making it impossible to separate bias in recalled information from actual differences in the frequency of exposures. As a result, findings from these studies may be spurious or equivocal. • Classifying the symptoms and identifying illnesses of Gulf War veterans have been difficult. From the outset, symptoms reported by veterans have been varied and difficult to classify into one or more distinct illnesses. Moreover, several different diagnoses might provide plausible explanations for some of the specific health complaints. It has thus been difficult to develop a case definition (that is, a reliable way to identify individuals with a specific disease), which is a criterion for doing effective epidemiological research. Appendix III provides more detailed information on the nature and extent of the federal government’s research efforts. Page 7 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 In the absence of official conclusions from DOD and VA, we examined Support for Key conclusions drawn in December 1996 by the Presidential Advisory Government Committee on Gulf War Veterans’ Illnesses, which was established by the Conclusions Is Weak President to review the activities of the executive branch regarding Gulf War veterans’ illnesses. In January 1997, DOD endorsed the Committee’s or Subject to conclusions about the likelihood that exposure to 10 commonly cited Alternative agents contributed to the explained and unexplained illnesses of these veterans. We found that the evidence to support several of these Interpretations conclusions is either weak or subject to alternative interpretations. The Committee concluded that “stress is likely to be an important contributing factor to the broad range of illnesses currently being reported by Gulf War veterans.” While stress can induce physical illness, the link between stress and these veterans’ physical symptoms has not been firmly established by the evidence the Committee cited. For example, a large-scale, federally funded study concluded that “for those veterans who deployed to the Gulf War and currently report physical symptoms, neither stress nor exposure to combat or its aftermath bear much relationship to their distress.” The Committee has stated that “epidemiological studies to assess the effects of stress invariably have found higher rates of posttraumatic stress disorder (PTSD) in Gulf War veterans than among individuals in nondeployed units or in the general U.S. population of the same age.” Our review indicated that the prevalence of PTSD among Gulf War veterans may be overestimated due to problems in the methods used to identify it. Specifically, the studies on PTSD to which the Committee refers have not excluded other conditions, such as neurological disorders that produce symptoms similar to PTSD and can also elevate scores on key measures of PTSD. We also believe that the use of broad and heterogenous groups of diagnoses (e.g., “psychological conditions” — ranging from tension headache to major depression) in reporting data from DOD’s clinical program may contribute to overestimation of the extent of serious psychological illnesses among Gulf War veterans. The Committee also concluded that “it is unlikely that infectious diseases endemic to the Gulf region are responsible for long term health effects in Gulf War veterans, except in a small known number of individuals.” Similarly, the Persian Gulf Veterans Coordinating Board (PGVCB)5 concluded that because of the small number of reported cases, “the likelihood of Leishmania tropica as an important risk factor for widely reported illness has diminished.” While this is the case for observed 5 The PGVCB, comprised of the Secretaries of Defense, Veterans Affairs, and Health and Human Services, is charged with coordinating the federal response to Gulf War veterans’ illnesses. Page 8 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 symptomatic infection with the parasite, the prevalence of asymptomatic infection is unknown, and such infection may reemerge in cases in which the patient’s immune system becomes deficient. However, as the Committee has noted, the infection may remain dormant up to 20 years. Because of this long latency and the lack of widely available screening methods, the infected population is hidden, and because even classic forms of Leishmaniasis are difficult to recognize, we believe that Leishmania should be retained as a potential risk factor for individuals who suffer from immune deficiency. The Committee also concluded that it is unlikely that the health effects reported by many Gulf War veterans were the result of (1) biological warfare agents, (2) chemical warfare agents, (3) depleted uranium, (4) infectious diseases endemic to the region, (5) oil-well fire smoke, (6) pesticides, (7) petroleum products, (8) pyridostigmine bromide, or (9) vaccines. However, our review of the conclusions made by the Committee indicated the following: • While the government found no evidence that biological weapons were deployed, during the Gulf War, the United States did not deploy a real-time biological warfare agent detection system during the war, and the effects of one agent, aflatoxin, would not be observed for many years. • Evidence from various sources indicates that chemical agents were present at Khamisiyah, Iraq, and elsewhere on the battlefield. The magnitude of the exposure to chemical agents has not been fully resolved. As we have previously noted, “available bomb damage assessments during the war concluded that 16 of 21 sites categorized by Gulf War planners as nuclear, biological, and chemical (NBC) facilities had been successfully destroyed. However, information compiled by the United Nations Special Commission (UNSCOM) since the end of Desert Storm reveals that the number of suspected NBC targets identified by U.S. planners, both prior to and during the campaign, did not fully encompass all the possible NBC targets identified by U.S. planners. UNSCOM has conducted investigations at a large number of the facilities suspected by the U.S. authorities as being NBC related. Regarding the few suspected chemical weapon sites that have not yet been inspected by UNSCOM, we have been able to determine that each was attacked by coalition aircraft during Desert Storm and that one site is located within the Kuwait theater of operation in close proximity to the border, where coalition ground forces were located. However, we have yet to learn why these facilities have not been investigated.”6 6 Operation Desert Storm: Evaluation of the Air Campaign (GAO/NSIAD-97-134, June 12, 1997), p. 2. Page 9 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 • Exposure to certain organophosphates can induce a delayed neurological condition without causing immediate symptoms. • Available research indicates that exposure to combinations of pyridostigmine bromide and other chemicals used during the Gulf War can cause damaging health effects greater than to these agents individually. (See app. IV for details.) Because of the numbers of Gulf War veterans who continue to experience Recommendations illnesses that may be related to their service during the Gulf War, we recommend that the Secretary of Defense, in conjunction with the Secretary of the Department of Veterans Affairs, (1) develop and implement a plan to monitor the clinical progress of Gulf War veterans in order to help promote appropriate and effective treatment and provide direction to the research agenda and (2) give greater priority to research on treatment for ill veterans and on low-level exposures to chemicals and their interactive effects and less priority to further epidemiological studies. We also recommend that the Secretaries of Defense and Veterans Affairs refine the current approaches of the clinical and research programs for diagnosing posttraumatic stress disorder consistent with suggestions recently made by the Institute of Medicine, which noted the need for improved documentation of screening procedures and patient histories (including occupational and environmental exposures) and the importance of ruling out alternative causes of impairment. We obtained comments on a draft of this report from DOD, VA, and the Agency Comments Presidential Advisory Committee on Gulf War Veterans’ Illnesses. DOD and Our Evaluation partially concurred with the report, indicating that the thrust of the recommendations had merit but did not fully take into account the complex set of health outcomes related to the war and did not recognize DOD’s accomplishments. DOD also noted that our findings differed from those of the Institute of Medicine and the Committee and commented that we had not carried out the same level of careful and thoughtful assessment as had those committees. VA commented that although some aspects of our report have merit, our recommendations reflected a lack of understanding of clinical research, epidemiology, and toxicology. VA indicated that (1) the creation of a new database was not likely to provide accurate and valid assessment of these Gulf War veterans’ health status; (2) DOD and VA are already giving greater Page 10 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 priority to research on low-level exposures to chemicals, but do not want to give less priority to epidemiological research; and (3) VA was already making efforts to improve current approaches to PTSD and other stress-related disorders. The Presidential Advisory Committee commented that our draft contained factual errors, did not provide references or citations to scientific literature, lacked substantiation and analytic rigor, and should not be issued in its current form. The Committee was particularly concerned with our finding that the support or evidence it had for some of its conclusions was weak. None of the comments we received provide evidence to challenge our principal findings and conclusions that (1) DOD and VA have no means to systematically determine whether symptomatic Gulf War veterans are better or worse today than when they were first examined and (2) ongoing epidemiological research will not provide precise, accurate, and conclusive answers regarding the causes of the Gulf War veterans’ illnesses. All of the comments we received seek to shift the onus of identifying and substantiating the causes of Gulf War illnesses to us, when in fact we merely reviewed the sufficiency and persuasiveness of the evidence behind the administration’s conclusions. In some instances, we found it to be weak or open to alternative interpretation. We believe the burden of proof is still on those who have made the assertions about the likely and unlikely causes of the illnesses. Nevertheless, in light of the comments we received, we have added more citations to the scientific support and documentation and modified the language in the text to clarify our position so that other readers will not misconstrue the meaning of our report. We also double-checked the information that was challenged in the comments we received and found that the data as originally presented was correct. Therefore, the thrust of our message remains unchanged. Our point-by-point evaluation of the detailed comments provided by DOD, VA, and the Committee are provided in appendixes V-VII. However, because the Committee’s comments were the most strident, our evaluation of its key points is summarized as follows. Regarding stress, the Committee states that we ignored its analytical approach, which was to compare the “known health effects of the risk factor to the symptoms reported by Gulf War veterans.” We found, Page 11 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 however, that the Committee offered little evidence that stress was an important contributor of the “broad range” of veterans’ symptoms. Given the nonspecific nature of the health effects associated with stress, almost any pattern of symptoms and illnesses would be compatible with it, making it difficult to scientifically test the hypothesis posed by the Committee’s approach (that the pattern of veterans’ illnesses is consistent with the known effects of stress). Although the Committee notes that scores on PTSD screening questionnaires are higher among Gulf veterans than among controls, confirmatory psychiatric interviews to eliminate alternative explanations for elevated PTSD screening scores were generally not done. Because we questioned the Committee’s support for its conclusion that the likelihood of Leishmania infection has diminished as an explanation for widespread illness, the Committee sought to transfer the burden of proof to us by asking that we justify any continued concern about asymptomatic infection. We found that the Committee’s justification for dismissal of Leishmania as a risk factor rests heavily on two ill-supported assumptions: (1) that diagnostic programs have been highly likely to detect the disease, even in the absence of any widely available screening or diagnostic tests and in the presence of nonspecific symptoms, and (2) that the course of various forms of leishmaniasis is well understood by scientists and by the doctors examining the veterans. Insofar as the prevalence of this infection is still unknown and it is impossible to predict which veterans’ immune systems will be weakened, and given the inability to identify this hidden population in the absence of a valid screening test, we believe it is premature to discount leishmania as a risk factor. Finally, regarding our evaluation of the Committee’s conclusion that low-level exposures to chemicals such as pesticides are unlikely to be associated with veterans’ health conditions, the Committee appears not to contest the fact that laboratory data document specific health effects in animals exposed to one or more organophosphate agents that are not detectable in the usual clinical tests. We find it difficult to reconcile the Committee’s dismissal of such exposure as an “unlikely” cause of veterans’ health problems with its acknowledgement of an absence of data on an important exposure scenario. Moreover, where the Committee apparently found no data to suggest a problem with low-level exposures, we found some data that do pose concerns. While the Committee argues that these studies were done on animals, they are consistent with standard toxicological practice employed by the Environmental Protection Agency and others. The Committee’s insistence that such effects be demonstrated Page 12 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 in humans appears unreasonable, as exposing humans to toxic substances for experimental research cannot be done for obvious ethical reasons. Also, while the study of occupational or accidental exposures, such as the sarin exposures that occurred in Japanese terrorist incidents, may provide some degree of information, the value of such information is generally limited by the poor description of the actual levels of the exposure in the case of accidents, the limited range of exposures (in the case of occupational use), or the lack of comparability with the circumstance in question. The hypotheses derived from such study would thus require confirmation and testing in controlled experimental settings. DOD provided two sets of comments, which we have reprinted in appendix V. We responded to DOD’s second set of comments, received on June 17, 1997, by incorporating appropriate changes in our report. VA and the Presidential Advisory Committee’s comments are reprinted in appendixes VI and VII, respectively. To address our first objective—the extent of DOD’s and VA’s clinical Scope and follow-up and monitoring of treatment and diagnostic services—we Methodology reviewed literature and agency documents and conducted structured interviews with DOD and VA officials managing the respective agencies’ registries for Persian Gulf War veterans that requested postwar evaluations. We asked questions designed to identify and contrast their methods for monitoring the quality and outcomes of their treatment and diagnostic programs and the health of the registered veterans. The second objective concerns the government’s research strategy to study the veterans’ illnesses and the methodological problems posed in its studies. To answer this question we conducted a systematic review of pertinent literature and agency documents and reports, including reports issued by the Presidential Advisory Committee and the Institute of Medicine. We also interviewed representatives of PGVCB’s Research Working Group and officials of VA and DOD. We surveyed primary investigators of ongoing epidemiological studies. We also collected data on project expenditures by fiscal year but did not attempt to independently verify these figures. Because different methodological standards apply to various types of research and because the overwhelming majority of federally sponsored research is categorized as epidemiological, we limited our survey of investigators to those responsible for ongoing epidemiological studies. Page 13 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 With an expert epidemiological consultant, we devised a questionnaire to assess critical elements of these studies (including the quality of exposure measurement, specificity of case definition, and steps to ensure adequate sample size) and to identify specific problems that the primary investigators may have encountered in implementing their studies. Of the 43 ongoing epidemiological studies identified by PGVCB in the November 1996 plan, we interviewed primary investigators for 31 (72 percent). We also reviewed and categorized descriptions of all 91 projects identified by April 1997, based on their apparent focus and primary objective. Finally, to review the progress of major ongoing research efforts, we visited the Walter Reed Army Institute of Research, the Naval Health Research Center, and two of VA’s Environmental Hazards Research Centers. To address the third objective, we reviewed major conclusions of the PGVCB and the Presidential Advisory Committee to determine the strength of evidence supporting these conclusions. The purpose of this review was not to critique PGVCB’s or the Presidential Advisory Committee’s efforts, per se, but rather to describe the amount of knowledge about Gulf War illnesses that has been generated by research 6 years after the war. We reviewed these conclusions because they were the strongest statements that we had come across on these matters by any official body. The Presidential Advisory Committee’s report was significant because the panel included a number of recognized experts in the scientific questions at issue who were assisted by a large staff of scientists and attorneys. In addition, the Committee conducted an extensive review of the research. Thus, evaluating these conclusions provided important evidence about how fruitful the federal research had been thus far. To address this objective, we reviewed scientific literature cited by the Presidential Advisory Committee as well as others. We also consulted experts in the fields of epidemiology, toxicology, and medicine and interviewed officials of DOD, VA, and the Central Intelligence Agency. We checked our own interpretation of key study findings with the authors and had independent experts review our draft report. Because of the scientific and multidisciplinary nature of this issue, we ensured that staff conducting the work had appropriate backgrounds in the fields of epidemiology, statistics, psychology, environmental health, toxicology, engineering, weapon design, and program evaluation and methodology. In addition, we used in-house expertise in chemical and biological warfare and in military and veterans health care systems. Also, experts who reviewed our work had backgrounds in medicine, public health, and research methods. Moreover, we held extensive discussions Page 14 GAO/NSIAD-97-163 Gulf War Illnesses B-276835 with experts in academia in each of the substantive fields relevant to this issue. Our work was completed between October 1996 and April 1997 in accordance with generally accepted government auditing standards. We are sending copies of this report to other interested congressional committees; the Secretaries of Defense, Veterans Affairs, and Health and Human Services; the Chair of the Presidential Advisory Committee; and other interested parties. We will make copies available to others upon request. If you have any questions or would like additional information, please contact me at (202) 512-3092 or Sushil K. Sharma, Ph.D, Dr.P.H., Assistant Director, at (202) 512-3460. Major contributors to this report are listed in appendix VIII. Kwai-Cheung Chan Director, Special Studies and Evaluations Page 15 GAO/NSIAD-97-163 Gulf War Illnesses Contents Letter 1 Appendix I 20 U.S. Troops’ Exposure to Hazardous Substances 20 Government Health Health Examination Programs for Gulf War Veterans 21 and Research Government Research Program 23 Programs for Gulf War Veterans Appendix II 27 No Mechanisms Exist to Monitor Veterans’ Quality of Care 27 No Systematic Efforts to Ensure Quality and Measure Satisfaction With Initial 28 Approach for Examination Are Not Adequate Monitoring Veterans’ Treatment After Initial Examination Appendix III 30 Goals of the Epidemiological Research 30 Federal Strategy to Federal Research Was Delayed and Lacked a Proactive Approach 37 Research Gulf War Some Hypotheses Received Early Emphasis 42 Some Hypotheses Were Not Initially Pursued 43 Illnesses Lacks a Additional Hypotheses Were Pursued in the Private Sector 43 Coherent Approach Federal Research Emphasis 46 Formidable Methodological Problems 48 Appendix IV 54 Extent of Posttraumatic Stress Disorder May Be Overestimated 57 Support for Key Extent of Asymptomatic Leishmania Infection Is Unknown 60 Official Conclusions Evidence of Exposure to Biological and Chemical Weapons Has 62 Not Been Aggressively Pursued Is Weak or Subject to Impact of DOD Denials on Federal Research 64 Different Interpretations Page 16 GAO/NSIAD-97-163 Gulf War Illnesses Contents Appendix V 65 Comments From the Department of Defense Appendix VI 77 Comments From the Department of Veterans Affairs Appendix VII 100 Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses Appendix VIII 135 Major Contributors to This Report Glossary 136 Related GAO Products 139 Tables Table III.1: Research Objectives Identified by PGVCB 30 Table III.2: PGVCB Research Projects 32 Table III.3: Events Coincident With Changes in PGVCB Research 40 Agenda Table III.4: Number of Studies by Primary Research Focus and 46 Study Type Table III.5: Primary Emphasis of 91 Federally Sponsored 47 Research Projects Identified by PGVCB Page 17 GAO/NSIAD-97-163 Gulf War Illnesses Contents Table III.6: Ongoing Epidemiological Studies Using Measures 51 Other Than Self-Reports to Assess Key Exposures Table IV.1: PAC Conclusions on Health Effects of Different 54 Individual Exposure Agents Figures Figure I.1: Percent of 3,181 VA Registrants That Reported Having 23 Been Exposed to Various Agents During the Gulf War Figure I.2: Epidemiologic Study Cycles 26 Figure III.1: Cumulative Percentages of 91 Federally Funded 38 Studies to Be Completed as a Function of Time Abbreviations CARC Chemical agent resistant coating CCEP Comprehensive Clinical Evaluation Program CDC Centers for Disease Control CIA Central Intelligence Agency DEET N,N-diethyl-m-toluamide DOD Department of Defense DS2 Decontaminating solution 2 HHS Department of Health and Human Services IOM Institute of Medicine NBC Nuclear, Biological, and Chemical NHRC Naval Health Research Center OPIDN Organophosphate-induced delayed neuropathy PAC Presidential Advisory Committee on Gulf War Veterans’ Illnesses PGHREP Persian Gulf Health Registry Examination Program PGVCB Persian Gulf Veterans Coordinating Board PTSD Posttraumatic stress disorder UNSCOM United Nations Special Commission on Iraq USAMRIID U.S. Army Medical Research Institute of Infectious Diseases VA Department of Veterans Affairs WRAIR Walter Reed Army Institute of Research WRAMC Walter Reed Army Medical Center Page 18 GAO/NSIAD-97-163 Gulf War Illnesses Page 19 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans In the aftermath of the Persian Gulf War, which ended on February 28, 1991, many veterans have experienced illnesses that they believe they contracted while in the Gulf due to exposures to hazardous materials or chemical and biological warfare agents. The Department of Veterans Affairs (VA) and the Department of Defense (DOD) later initiated health programs offering physical examinations and diagnostic services to these veterans. As it became apparent that the symptoms and causes of these illnesses varied widely and the illnesses were difficult to diagnose, the government began to research the reasons for the veterans’ health problems. During their service in the Gulf, U.S. troops were reportedly exposed U.S. Troops’ Exposure before, during, and after the war to a variety of potentially hazardous to Hazardous substances. These include decontaminating and protective compounds Substances (particularly decontaminating solution 2, or DS2, and chemical agent resistant coating (CARC)), diesel fuel used as a sand suppressant in and around encampments, fuel oil used to burn human waste, fuel in shower water, and leaded vehicle exhaust used to dry sleeping bags. For example, as we reported to staff of the House Subcommittee on Oversight and Investigations in 1994, DS2 was to be widely distributed among Army units and equipment in Saudi Arabia, though the Army did not know how much or where the solution was distributed. The potential effects of DS2 on humans include mild or severe burns, corrosion to tissues of the skin or eye, liver damage, and adverse reproductive effects.7 Other potential hazards associated with Gulf service included infectious diseases (most prominently leishmaniasis, a parasitic infection), the use of pyridostigmine bromide and vaccines (to protect against chemical and biological weapons), depleted uranium (contained in certain ammunition and in the fragments of exploded rounds), pesticides and insect repellents, chemical and biological warfare agents, and compounds and particulate matter contained in the extensive smoke from the oil-well fires at the end of the war. Shortly after the war, some veterans began reporting health problems that they believed might be due to their participation in the war. As we noted in May 1995, the 123rd Army Reserve Unit in Indiana reported unexpected signs and symptoms that could not easily be explained.8 Veterans in other 7 Also see Hazardous Materials: DOD Should Eliminate DS2 From Its Inventory of Decontaminants (GAO/NSIAD-90-10, Apr. 25, 1990). 8 Operation Desert Storm: Health Concerns of Selected Indiana Persian Gulf War Veterans (GAO/HEHS-95-102, May 1995). Page 20 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans units began to report similar symptoms that also could not be easily explained. Many veterans reported exposure to chemicals, pesticides, and other agents, such as vaccines and pyridostigmine bromide, as possible causes of their illnesses. Consistent with the Veteran’s Health Care Act of 1992, both VA and DOD Health Examination have established programs through which they provide medical Programs for Gulf War examination and diagnostic services, free of charge, to Gulf War veterans. Veterans The VA launched its Persian Gulf Health Registry Examination Program (PGHREP) in 1992, and DOD initiated the Comprehensive Clinical Evaluation Program (CCEP) in June 1994. PGHREP is currently available at most VA medical centers, and DOD’s CCEP examinations are available at 148 sites worldwide.9 Initially, PGHREP participants received a regular physical examination with basic laboratory tests. However, in 1994, VA established a standardized examination to (1) obtain information about symptoms and exposures; (2) call the clinician’s attention to diseases endemic to the Gulf region; and (3) direct baseline laboratory studies, including a chest X-ray (if one has not been done recently), blood count, urinalysis, and blood chemistry and enzyme analyses for detection of certain diseases. If a diagnosis is not apparent, facilities follow the clinical evaluation protocol originally developed for VA’s referral centers and now used in VA and military medical centers nationwide. The examination protocol suggests 22 additional baseline tests and additional specialty consultations, from which further diagnostic procedures may be considered, depending on the veteran’s symptoms. If the illness cannot be diagnosed, a VA registrant may be referred to one of four VA Persian Gulf Referral Centers located in Washington, D.C.; Houston, Texas; Los Angeles, California; and Birmingham, Alabama. Although these registry programs are primarily intended to provide diagnostic services, the VA’s registry program, in particular, was among the first extensive efforts to gather data from veterans regarding the nature of their problems and the types of factors to which they might have been exposed. However, during the first 2 years of the PGHREP’s operation, when exposure and symptom information was freshest in most respondents’ 9 A declining proportion of Gulf War veterans are covered by military medical health services. Thus, an increasing number would be eligible only for VA care, and some portion have access to neither VA nor DOD health care services apart from the PGHREP and CCEP examination programs. It is important to note that receipt of a VA registry examination does not entail eligibility for free treatment of any conditions detected. Page 21 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans memory, efforts to collect data on symptoms and self-reported exposure(s) were more limited in scope. In February 1997, the VA reported its analysis of the self-reported exposures to hazardous substances identified by the 3,181 veterans who registered after the VA revised its data collection forms. Although interpreting such information is difficult due to self-selection of registry participants and problems with retrospective reporting based on recall, many veterans reported exposure to multiple substances. VA reported that 89 percent of these registered veterans believed they had been exposed to diesel or other petrochemical fumes, 88 percent to passive smoking, 72 percent to smoke from oil fires, 72 percent to skin exposure to fuel, 70 percent to burning trash/feces, 64 percent to smoke from tent heaters, 64 percent to pesticides in cream or spray form, 60 percent to pyridostigmine bromide, 53 percent to paints or solvents, 34 percent to microwaves, 32 percent to CARC, 14 percent to depleted uranium, 12 percent to nerve gas, and 6 percent to mustard gas.10 (See fig. I.1.) As of June 1996, DOD acknowledged the potential exposure of some veterans to the nerve agent sarin following the postwar U.S. demolition of Iraqi ammunition facilities. 10 H. Kang et al., “A Review of the Department of Veterans Affairs Revised Persian Gulf Registry and In-Patient Treatment Files” (Washington, DC: VA Environmental Epidemiology Service, Feb. 1997), table 17. Page 22 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans Figure I.1: Percent of 3,181 VA Registrants That Reported Having Been Exposed to Various Agents During the Gulf War Diesel fumes 89 Passive smoking 88 Oil well fires 72 Fuels (skin) 72 Burning trash/feces 70 Tent heaters 64 Pesticides (skin) 64 Pyridostigmine bromide 60 Paints/solvents 53 Microwaves 34 CARC 32 Depleted uranium 14 Nerve gas 12 Mustard agent 6 0 20 40 60 80 100 Percent Source: Data provided in H. Kang et al., “A Review of the Department of Veterans Affairs Revised Persian Gulf Registry and Inpatient Treatment Files” (Washington, D.C.: VA Environmental Epidemiology Service, Feb. 1997), table 17. In addition to providing examination services, in the 6 years since the end Government Research of the Gulf War, the federal government, primarily through DOD and VA, has Program sponsored a variety of research on Gulf War veterans’ illnesses. DOD research is one component of a broader agenda coordinated under the aegis of the Persian Gulf Veterans’ Coordinating Board (PGVCB), which comprises the Secretaries of the Department of Health and Human Services (HHS), VA, and DOD. The details of this agenda are described in the PGVCB publication entitled A Working Plan for Research on Persian Gulf Veterans’ Illnesses, first published in 1995 and revised in November 1996.11 11 A Working Plan for Research on Persian Gulf Veterans’ Illnesses (First Revision),” Department of Veterans Affairs, November 1996. Page 23 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans This agenda was developed in response to an Institute of Medicine (IOM) conclusion that “the DOD and VA should determine the specific research questions that need to be answered. Epidemiologic studies should be designed with the objective of answering these questions given the input of experts in epidemiologic research methods and data analysis, along with the input of experts in the subject matter areas to be investigated.” Accordingly, most of the research sponsored under this agenda is characterized by PGVCB as epidemiological. The objectives of epidemiologic study are to determine the extent of disease in the population, the causes of disease and its modes of transmission, the natural history of disease, and the basis for developing preventive strategies or interventions.12 To conduct such studies, investigators must follow a few basic, generally accepted principles. First, they must specify diagnostic criteria to (1) reliably determine who has the disease or condition being studied and who does not and (2) select appropriate controls (people who do not have the disease or condition). Second, the investigators must have valid and reliable methods of collecting data on the past exposure(s) of those in the study to possible factors that may have caused the symptoms. The need for accurate, dose-specific exposure information is particularly critical when low-level or intermittent exposure to drugs, chemicals, or air pollutants is possible. It is important not only to assess the presence or absence of exposure but also to characterize the intensity and duration of exposure. To the extent that the actual exposure of individuals is misclassified, it is difficult to detect any effects of the exposure. Another means of linking environmental factors to disease is to determine whether or not there is evidence that as the exposure increases, the risk of disease also increases. However, this dose-response pattern can be detected only if the degree of exposure among different groups can be determined. Finally, in addition to specific case definition and dose-specific exposure information with known accuracy, it is important that a sufficient number of persons be studied to have a reasonable likelihood of detecting any 12 A. M. Lilienfeld and D. E. Lilienfeld, Foundations of Epidemiology (New York: Oxford University Press, 1980); L. Goodis, Epidemiology (Philadelphia, PA: W. B. Saunders Company, 1996); and D. E. Lilienfeld and P. D. Stolley, Foundations of Epidemiology 3rd ed. (New York: Oxford University Press, 1994). Page 24 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans relationship between exposures and disease. To the extent that this relationship is subtle or obscured by loose case definition (that is, a case definition that is too broad and encompasses different types of illnesses) or problems in measuring exposure, larger samples would be required. For example, the IOM has noted that “very large groups must be studied in order to identify the small risks associated with low levels of exposure, whereas a relatively small study may be able to detect the effect of heavy or sustained exposure to a toxic substance. In this way, a study’s precision or statistical power is also linked to the extent of the exposure and the accuracy of its measurement. Inaccurate assessment of exposure can obscure the existence of such a trend and thus make it less likely that a true risk will be identified.”13 Research programs, such as the federal program for Gulf War illnesses, are designed to lead to information and treatments in a timely and efficient manner. In the conventional model of epidemiological research, a research program to investigate a disease outbreak follows a study cycle. (See fig. I.2 for an illustration of this cycle.)14 First, descriptive studies are conducted to gather basic information about patterns of illness, the natural environment, and exposures of interest (step one). Once enough information is gathered, researchers create hypotheses to explain the patterns that they see in these descriptive data (step two). Analytic studies are then conducted to test the hypotheses (step three). The results of these analytic studies are evaluated (step four). They may suggest a need to gather additional descriptive data (step one), or they may yield new or refined hypotheses (step two) to be tested in further analytic studies (step three). The cycle continues until the disease is adequately understood to permit the development of treatments. Applying this cycle is a useful way of organizing a research program when valid descriptive information can be acquired about exposures and dose/response relationships. However, when this information is not available, retaining this model, that is conducting descriptive studies that lead to hypotheses that are then tested, may not be timely or effective. 13 Veterans and Agent Orange: Update 1996 (Washington, D.C.: Institute of Medicine, 1996), pp. 99-100. 14 This research model is shared by other scientific fields that also develop hypotheses and then test them through field or laboratory research. Page 25 GAO/NSIAD-97-163 Gulf War Illnesses Appendix I Government Health and Research Programs for Gulf War Veterans Figure I.2: Epidemiologic Study Cycles 1. Descriptive studies- data aggregation and analysis 4. Analysis of results 2. Model-building and suggests further formulation of descriptive studies hypotheses and new hypotheses 3. Analytic studies to test hypotheses Source: Mausner and Bahn, Epidemiology—An Introductory Text, p. 155. Page 26 GAO/NSIAD-97-163 Gulf War Illnesses Appendix II No Systematic Approach for Monitoring Veterans’ Treatment After Initial Examination Most veterans with symptoms who were evaluated in the VA and DOD health programs received a diagnosis (78 percent and 80 percent, respectively), while at least 20 percent have symptoms or signs that elude diagnosis. However, an analysis of 222 VA registrants in Portland, Oregon, showed that only 19 percent had symptoms that were fully explained by the coded registry diagnoses. This suggests that undiagnosed signs and symptoms may be more common than is apparent from initial analyses of registry data. The most commonly reported symptoms in VA and DOD registries include fatigue, muscle and joint pain, gastrointestinal complaints, headache, skin rashes, depression, neurologic and neurocognitive impairments, memory loss, shortness of breath, and sleep disturbance. It is noteworthy that veterans participate in the registry programs even though (1) participation in the programs is voluntary and some members of the active duty service may perceive it as career-limiting; (2) the health registry programs provide only diagnostic services and treatment incidental to diagnosis (for example, removal of malignancies found during a diagnostic biopsy); and (3) the examination can be lengthy. Officials of both DOD and VA have testified that whatever uncertainties may No Mechanisms Exist exist about the cause of veterans’ illnesses, the veterans are at least to Monitor Veterans’ receiving appropriate and effective symptomatic treatment.15 In the case of Quality of Care veterans with no clear diagnosis, treatment is based on symptoms, and veterans with multiple diagnoses may see multiple types of providers. However, these agencies had no mechanisms for monitoring the quality of these veterans’ care or their clinical progress after their initial examination, nor did they describe plans to establish such mechanisms.16 VA delegates monitoring responsibilities to local veterans hospitals, which may monitor the quality of a subsample of services.17 15 Testimony before the House Government Reform and Oversight Committee’s Subcommittee on Human Resources and Intergovernmental Relations by VA’s Chief Public Health and Environmental Hazards Officer, December 11, 1996, and a written statement submitted by Dr. Stephen C. Joseph, Assistant Secretary of Defense for Health Affairs, to a September 25, 1996, joint hearing of the Senate Select Intelligence and Veterans’ Affairs Committees. 16 Since November 1996, DOD has been working with independent contractors to determine appropriate health outcomes and other metrics that would characterize the current health status of those participating in the CCEP. In addition, a 5-year follow-up policy has been developed for the small number of soldiers suspected of being exposed to depleted uranium particles. 17 The term “quality assurance” is used to describe prospective processes or requirements—such as licensure, inspections, or training—generally intended to promote a certain level of performance based on criteria that might be the subject of a quality measurement program (that is, indicators of the achievement of program goals or the capacity to achieve such goals). Page 27 GAO/NSIAD-97-163 Gulf War Illnesses Appendix II No Systematic Approach for Monitoring Veterans’ Treatment After Initial Examination VA officials involved in administering PGHREP told us that they regarded monitoring the clinical progress of registry participants as a separate research project, and DOD’s CCEP manager made similar comments. Instead, the two agencies have relied on such quality assurance mechanisms as standards for physician qualification and process measurements, although these do not necessarily ensure a given level of effectiveness for the care provided.18 Although VA’s Central Office samples a subset of all veterans having contact with VA hospitals to determine their satisfaction with VA health care, VA officials told us that this sample is not currently large enough to provide information specific to veterans of the Gulf War. Similarly, local facilities may conduct studies of their success with a particular medical treatment (e.g., a coronary bypass), but these are unlikely to provide specific information on Gulf War veterans.19 Both VA and DOD have applied some traditional quality assurance and Efforts to Ensure measurement strategies to the initial examination of Gulf War veterans.20 Quality and Measure In response to a recommendation of the IOM, DOD has asked for feedback Satisfaction With from CCEP participants; however, we have found some problems with DOD’s approach to analyzing Gulf War veterans’ responses to questionnaires Initial Examination relating to their satisfaction with CCEP evaluations. Specifically, Are Not Adequate (1) Gulf War veterans’ responses were not compared to responses from other groups of patients seeking diagnosis and treatment. 18 See Long-Term Care: Status of Quality Assurance and Measurement in Home and Community-Based Services (GAO/PEMD-94-19, Mar. 1994). 19 For previous GAO reviews of performance monitoring and outcomes measurement in VA and DOD health care programs, see Defense Health Care: New Managed Care Plan Progressing, but Cost and Performance Issues Remain (GAO/HEHS-96-128); VA Health Care: Trends in Malpractice Claims Can Aid in Addressing Quality of Care Problems (GAO/HEHS-96-24); VA Health Care: Physician Peer Review Identifies Quality of Care Problems, but Actions to Address Them Are Limited (GAO/HEHS-95-121); Veterans’ Health Care: Veterans’ Perceptions of VA Services and VA’s Role in Health Care Reform (GAO/HEHS-95-14); VA Health Care for Women: In Need of Continued VA Attention (GAO/HEHS-94-114); and VA Health Care: VA Medical Centers Need to Improve Monitoring of High-Risk Patients (GAO/HRD-94-27). 20 DOD identified the following quality assurance measures that are in place for all DOD eligible individuals: The National Quality Management Program, Clinical Quality Management Program for the Military Health Service System. The National Quality Management Program comprises the following seven components: Medical Readiness, Accreditation of Healthcare Organizations, Licensure, Credentials and clinical Privileges, National Practitioner Data Bank, Utilization Management Oversight, and Special Studies. Page 28 GAO/NSIAD-97-163 Gulf War Illnesses Appendix II No Systematic Approach for Monitoring Veterans’ Treatment After Initial Examination (2) DOD combined into a single category the “no opinion” and missing responses. (3) It is unclear to what extent responses were included from patients who voluntarily declined participation in the program or were placed in an “administrative declination” category for “failure to become actively involved in the CCEP program.” In addition, no information is available on the extent to which active duty veterans have sought evaluations outside the CCEP program, which could be an indicator of patient dissatisfaction. VA plans to initiate a satisfaction measure for its PGHREP by incorporating a feedback postcard in the examination process. However, this plan has been under development for months, and VA officials told us it remained so as of March 24, 1997. Both VA and DOD have applied some quality measurement to their registry examination processes. A VA directive requires VA medical centers to use the PGHREP Quality Management/Self-Assessment Monitor to review at least a 10-percent sample of all Persian Gulf registry physical examinations conducted from January 1, 1996, through September 30, 1996. The results of this process were submitted to VA’s Environmental Agents Service and summarized for 167 of the VA’s 173 medical centers. They showed a fairly high self-reported compliance with various aspects of the examination among reporting facilities. Although the results of this self-assessment process suggested that the overwhelming majority of veterans who were symptomatic at the time of the examination had a follow-up examination scheduled, 20 percent were not assigned to a primary care team, and 17 percent of their records lacked evidence of a follow-up letter containing examination results and recommendations. Similarly, DOD tracks the aging of requests for CCEP examinations and provides feedback to its regional facilities on examinations that are overdue or reports that remain incomplete. DOD does not consider an examination complete until certain fields in the examination report have been filled in and submitted. Through supervisory personnel at 13 regional treatment centers, DOD also conducts some oversight of examination activities at its 148 CCEP administration sites. Page 29 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach The approach to collecting data on Persian Gulf veterans and the factors that might have caused their health problems has not been proactive, and articulation of a research agenda came years after the war. The subsequent research, which is largely epidemiological and still ongoing, has pursued some hypotheses more aggressively than others and faces formidable methodological problems. President Clinton established PGVCB on January 21, 1994, to coordinate Goals of the federal efforts to address health concerns raised by veterans of the Persian Epidemiological Gulf War. Various federal agencies had previously independently Research addressed these concerns. The Research Working Group (RWG) of the PGVCB has primary responsibility for managing research into Gulf War illnesses. In August 1995, PGVCB identified three broad goals for research on veterans’ illnesses: (1) determine the nature and prevalence of symptoms, illnesses, and unexplained conditions among Persian Gulf veterans; (2) identify possible causes for any illnesses found among Persian Gulf veterans; and (3) identify diagnostic tools, treatment methods, and prevention strategies for illnesses found among Persian Gulf veterans. These are generally accepted goals for the epidemiologic study of poorly understood conditions. To support these research goals, PGVCB identified 21 more specific research objectives. (See table III.1.) Table III.1: Research Objectives Identified by PGVCB Number Objective 1 What is the prevalence of symptoms/illnesses in the Gulf War veterans’ population? How does this prevalence compare to that in an appropriate control group? 2 What was the overall exposure of troops to leishmania tropica? 3 What were the exposure concentrations to various petroleum products and their combustion products in typical usage during the Gulf War? 4 What was the extent of exposure to specific occupational/environmental hazards known to be common in the Gulf War veterans’ experience? Was this exposure different from that of an appropriate control group? 5a What were the potential exposures of troops to organophosphate nerve agents and/or sulfur mustard as a result of allied bombing at Muhammadiyat and Al Muthanna or the demolition of a weapons bunker at Khamisiyah? 6a What was the extent of exposure to chemical agents, other than at Khamisiyah in the Gulf War as a function of space and time? 7 What was the prevalence of pyridostigmine bromide use among Gulf War troops? (continued) Page 30 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Number Objective 8 What was the prevalence of various psychophysiological stressors among Gulf War veterans? Is the prevalence different from that of an appropriate comparison population? 9 Are Gulf War veterans more likely than an appropriate comparison group to experience nonspecific symptoms and symptom-complexes? 10 Do Gulf War veterans have a greater prevalence of altered immune function or host defense when compared with an appropriate control group? 11 Is there a greater prevalence of birth defects in the offspring of Gulf War veterans than in an appropriate control population? 12 Have Gulf War veterans experienced lower reproductive success than an appropriate control population? 13 Is the prevalence of sexual dysfunction greater among Gulf War veterans than among an appropriate comparison population? 14 Do Gulf War veterans report more pulmonary symptoms or diagnoses than persons in an appropriate control population? 15 Do Gulf War veterans have a smaller baseline lung function in comparison to an appropriate control group? Do Gulf War veterans have a greater degree of nonspecific airway reactivity in comparison to an appropriate control group? 16 Is there a greater prevalence of organic neuropsychological and neurological deficits in Gulf War veterans compared to appropriate control populations? 17 Can short-term, low-level exposures to pyridostigmine bromide, the insect repellent DEET, and the insecticide permethrin, alone or in combination, cause short-term and/or long-term neurological effects? 18 Do Gulf War veterans have a significantly higher prevalence of psychological symptoms and/or diagnoses than do members of an appropriate control group? 19 What is the prevalence of leishmaniasis and other infectious diseases in the Gulf War veteran population? 20 Do Gulf War veterans have a greater risk of developing cancers of any type when compared with an appropriate control population? 21 Are Gulf War veterans experiencing a mortality rate that is greater than that of an appropriate control population? Are specific causes of death related to service in the Persian Gulf? a Objective was added in 1996, following DOD’s announcement of potential exposures to chemical warfare agents in postwar operations at three sites. In connection with these research objectives, 91 specific studies were identified in the most recent annual report to Congress on federally Page 31 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach sponsored research on Gulf War veterans’ illnesses.21 (See table III.2.) According to figures reported by the agencies, through fiscal year 1996, DOD expended $19.9 million on these research efforts, VA spent $11.6 million, and HHS attributes expenditures of $5.5 million to the four projects it has pursued. In all, these agencies spent about $37 million for research through fiscal year 1996. Additional amounts have been made available for ongoing and future projects. Table III.2: PGVCB Research Projects Start Completion Expenditures Project no. Title Status Location date date through FY96a DoD-1A Epidemiologic Studies of Morbidity Among Gulf War Veterans: A Search for Etiologic Agents and Risk Factors Ongoing NHRC 6/1/94 10/1/96 $3,985,000 DoD-1B A Search for Etiologic Agents and Risk Factors: Study 2 Complete NHRC 7/1/94 10/1/96 See DoD-1A DoD-1C A comparative study of pregnancy outcomes Complete NHRC 9/1/94 6/1/96 See DoD-1A DoD-1D Infertility and Miscarriage in Gulf War Veterans Ongoing NHRC 11/1/94 9/1/97 See DoD-1A DoD-1E Seabee Health Study Ongoing NHRC 1/1/96 10/1/2011 See DoD-1A DoD-1F A Comparison of Nonfederal Hosp. Experience Among Veterans in California ... Ongoing NHRC 6/1/95 6/1/97 See DoD-1A DoD-1G Epidemiologic Studies of Morbidity —Study 7: Prevalence of Congenital Anomalies Among Children of Persian Gulf War Veterans Ongoing NHRC 6/1/95 6/1/97 See DoD-1A DoD-2 Physiological and Neurobehavioral Effects in USAMRD Rodents from Exposure to PB, Fuels, and (Wright- DEET Ongoing Patterson) 7/1/94 10/1/97 $90,000 DoD-4 The General Well-Being of Gulf War Era Service Personnel from the States of PA and HI Complete WRAIR 9/1/92 5/20/94 $1,200,000 DoD-6A Combat Stress Pharmacotherapy Ongoing WRAIR 10/1/88 9/30/99 $30,000 DoD-6B Combat Stress Diagnosis: PTSD Prevention Ongoing WRAIR 10/1/87 9/30/98 0 DoD-7A Health Risk Assessment of Embedded Depleted Uranium Ongoing AFFRI 12/1/94 1/30/98 $703,000 DoD-7B Carcinogenicity of Depleted Uranium Fragments Ongoing ITRI 1/26/95 10/30/98 $549,000 DoD-8A Serologic Diagnosis of Viscerotropic Leishmaniasis Complete WRAIR 10/1/93 9/1/96 $10,000 DoD-8B Development of Leishmania Skin Test Antigen Ongoing WRAIR 10/1/93 1/31/2000 $421,000 (continued) 21 The Research Working Group of the Persian Gulf Veterans Coordinating Board, Annual Report to Congress: Federally Sponsored Research on Gulf War veterans’ Illnesses (Washington, D.C.: Department of Veterans Affairs, Apr. 1997). Also see PGVCB, A Working Plan for Research on Persian Gulf Veterans’ Illnesses (Washington, D.C.: VA, Nov. 1996). Page 32 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Start Completion Expenditures Project no. Title Status Location date date through FY96a DoD-9 Identification of the Genetic Factors Which USAMRU- Control Tropism in Leishmania Ongoing Brazil 7/1/94 7/1/98 $150,000 DoD-10 Pyridostigmine Synergistic Toxicity Study Complete CHPPM 11/1/94 3/1/94 $44,000 DoD-11 Male/Female Differential Tolerances to South Florida Pyridostigmine Bromide Ongoing Drug Research 10/1/94 2/1/97 $908,000 DoD-12 Forward Deployable Diagnostics for Infectious Diseases Ongoing MRMC 10/1/93 9/30/2001 $1,546,000 DoD-13 Effects of Persian Gulf War Service on Military Working Dogs Ongoing AFIP 4/1/94 12/1/98 $200,000 DoD-14 Risk Factors Among U.S. Soldiers for Enrolling on the Department of Veterans Affairs Gulf War Registry Ongoing WRAIR 10/1/93 9/1/96 $70,000 DoD-15 Comparative Mortality Among US Military Personnel Worldwide During Operations Desert Shield and Desert Storm Complete WRAIR 5/1/94 1/1/95 $20,000 DoD-16 Kuwait Oil Fire Health Risk Assessment Complete CHPPM 5/5/91 2/18/94 $1,805,000 DoD-17 Retrospective Studies Involving Military Use of PB as a Pretreatment for Nerve Agent Poisoning Complete CHPPM 1/1/91 5/21/92 $21,000 DoD-18 Kuwait Oil Fires Troop Exposure Assessment Model (TEAM) Complete CHPPM 5/31/93 12/31/96 $1,500,000 DoD-19 Persian Gulf Veterans Health Tracking System Ongoing CHPPM 4/1/96 12/31/97 $25,000 DoD-20 A Statistical Study Correlating the Reported Cases of Gulf War Syndrome to Battlefield Locations of Afflicted US Army Personnel U.S. Army During the Iraq-Kuwait War Complete Research Lab. 1/1/94 7/1/95 $50,000 DoD-21 Study of Variability in Pyridostigmine Inhibition WRAIR and of Blood Cholinesterases in Healthy Adults ... Ongoing WRAMC 7/11/95 6/30/97 $138,000 DoD-22 Chronic Organophosphorus Exposure and University of Cognition Ongoing GA 4/15/95 5/14/98 $187,000 DoD-23 Acute and Long-Term Impact of Deployment to Southwest Asia on the Physical and Mental Health of Soldiers and their Families Ongoing WRAIR 10/1/93 9/1/98 $326,000 DoD-30 Epidemiological Studies Persian Gulf War Klemm Illnesses, PG Women’s Health Linkage Study Ongoing Analysis, DC 6/1/96 1/31/99 $779,000 DoD-31 Dysregulation of the Stress Responses in the Georgetown Persian Gulf Syndrome Ongoing Univ., DC 5/6/96 6/6/99 $162,000 DoD-32 Neuropsychological Functioning in Persian Gulf Era Veterans Ongoing VAMC Boston 5/1/96 5/1/99 $353,000 DoD-33 Effects of Pyridostigmine in Flinders Line Rats Differing in Cholinergic Sensitivity Ongoing Chapel Hill 7/1/96 6/30/98 $44,000 DoD-34 Characterization of Emissions from Heaters Lovelace Burning Leaded Diesel Fuel in Unvented Tents Biomedical Albuquerque, Ongoing NM 6/7/96 7/6/98 $36,000 (continued) Page 33 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Start Completion Expenditures Project no. Title Status Location date date through FY96a DoD-35 Feasibility of Investigating whether there is a March of Relationship Between Birth Defects and Dimes, Service in the Gulf War Sacramento, Ongoing CA 6/1/96 6/30/98 $110,000 DoD-36 Fatigue in Persian Gulf Syndrome-Physiologic Mechanisms Ongoing Dallas 6/15/96 7/14/98 $138,000 DoD-37 Neurobehavioral and Immunological Toxicity of Pyridostigmine, Permethrin, and DEET in Male and Female Rats Ongoing Gainesville 5/1/96 6/1/99 $132,000 DoD-38 Diagnostic Antigens of Leishmania tropica Infectious Disease Research Inst.,Seattle, Ongoing WA 6/1/96 5/31/98 $612,000 DoD-39 A Controlled Epidemiological and Clinical Study into the Effects of Gulf War Service on ... UK Armed Forces Ongoing UK 6/1/96 6/30/99 $865,000 DoD-40 Psychological and Neurobiological VAMC West Consequences of the Gulf War Experiences Ongoing Haven 6/7/96 7/6/99 $90,000 DoD-41 Evaluation of Muscle Function in Persian Gulf Veterans Ongoing Philadelphia 6/15/96 11/14/99 $906,000 DoD-42 The Symptomatic Persian Gulf Veterans Protocol: An Analysis of Risk Factors with an Immunologic and Neuropsychiatric VAMC Assessment Ongoing Birmingham 10/31/96 2000 $700,000 DoD-44 Investigation of Seminal Plasma Univ. Hypersensitivity Reactions Ongoing Cincinnati 11/1/96 10/1/99 $634,000 DoD-45 Physical and Emotional Health of Gulf War Veterans Women Ongoing Ann Arbor, MI 9/1/96 8/31/99 $100,000 DoD-46 Exploratory Data Analysis with the CCEP NPGS - Database Ongoing Missouri 10/1/95 9/1/97 $60,000 DoD-47 Study of Mycoplasmal Infections in Gulf War Veterans Ongoing WRAMC 10/10/95 8/30/97 $112,000 DoD-48 Assessment of Genomic Instability via Chromosome 7 Inversion Frequency in a Gulf-War Syndrome Cohort vs. Selected Control Groups Ongoing AFIP 10/10/95 5/31/97 $74,000 DoD-49 Diagnosis and Dosimetry of Exposure to Sulfur Mustard Ongoing Netherlands 10/1/96 2/28/2000 Not available DoD-50 Toxicokinetics of VX Ongoing Netherlands 10/15/96 4/30/98 Not available DoD-51 Transgenic Engineering of Cholinesterases Ongoing Israel 10/1/96 2/28/2000 Not available HHS-1 Health Assessment of Persian Gulf War Veterans from Iowa Ongoing CDC 12/1/94 1/31/98 $4,772,000 HHS-2 Disease Cluster in a Pennsylvania Air National Guard Unit, EPI-AID 95-18 Ongoing CDC 12/1/94 3/31/95 $750,000 (continued) Page 34 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Start Completion Expenditures Project no. Title Status Location date date through FY96a HHS-3 Biomarkers of Susceptibility and Polycyclic Aromatic Hydrocarbon Exposure in Urine and Blood Cell DNA from US Army Soldiers NIH/NCI/DCE b Exposed to Kuwaiti Oil Well Fires Ongoing /LCTP/IVP 1/1/92 1/31/97 HHS-4 Suspected Increase of Birth Defects and Health Problems Among Children Born to Persian Gulf War Veterans in Mississippi Complete CDC 5/1/94 2/1/96 $15,981c VA-1 Mortality Follow-up Study of Persian Gulf Veterans Complete VACO 7/1/94 7/31/96 $150,000 VA-2 National Health Survey of Persian Gulf Veterans Ongoing VACO 7/1/94 6/30/98 $2,716,000 VA-3 Use of Roster of Veterans Who Served in Persian Gulf Area Ongoing VACO 7/1/94 7/31/96 0 VA-4A Evaluation of Cognitive Functioning of Persian Gulf Veterans Ongoing VAMC Boston 10/1/94 9/30/99 $2,572,500 VA-4B Evaluation of Neurological Functioning of Persian Gulf Veterans Ongoing VAMC Boston 10/1/94 9/30/99 See VA-4A VA-4C Gulf War and Vietnam Veterans Cancer Incidence Surveillance Ongoing VAMC Boston 10/1/94 9/30/99 See VA-4A VA-4D Evaluation of Respiratory Dysfunction Among Gulf War Veterans Ongoing VAMC Boston 10/1/94 9/30/96 See VA-4A VA-4E The Aromatic Hydrocarbon Receptor as a Biomarker of Susceptibility Ongoing VAMC Boston 10/1/94 9/30/99 See VA-4A VA-4F Validity of Computerized Tests Ongoing VAMC Boston 10/1/94 9/30/99 See VA-4A VA-5A Health and Exposure Survey of Persian Gulf VAMC E. Veterans Ongoing Orange 10/1/94 9/30/96 $2,572,500 VA-5B Physiological and Psychological Assessments VAMC E. of Persian Gulf Veterans Ongoing Orange 10/1/94 3/1/97 See VA-5A VA-5C Effects of Exertion and Chemical Stress on VAMC E. Persian Gulf Veterans Ongoing Orange 4/1/97 9/30/99 See VA-5A VA-5D Effects of Genetics and Stress on Responses VAMC E. to Environmental Toxins Ongoing Orange 10/1/94 9/30/97 See VA-5A VA-6 Portland Environmental Hazards Research Center: Environmental, Veterans Health and the Gulf War Syndrome Ongoing VAMC Portland 10/1/94 9/30/99 $2,572,500 VA-6A Psychosocial, Neuropsychological and Neurobehavioral Assessment Ongoing VAMC Portland 10/1/94 9/30/99 See VA-6 VA-6B Clinical and Neuroendrocrine Aspects of Fibromyalgia (Project II) Ongoing VAMC Portland 10/1/94 9/30/99 See VA-6 VA-6C Neurotoxicity of Environmental Pollutants and Warfare Agents (Project III) Ongoing VAMC Portland 10/1/94 9/30/99 See VA-6 VA-6D DNA Damage from Chemical Agents and its Repair (Project IV) Ongoing VAMC Portland 10/1/94 9/30/99 See VA-6 VA-7 Desert Storm Reunion Survey Complete VAMC Boston 4/1/91 9/30/95 $122,500 (continued) Page 35 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Start Completion Expenditures Project no. Title Status Location date date through FY96a VA-8 Psychological Test Data of Gulf War Veterans VAMC over Time Mountain Ongoing Home 7/9/91 9/30/97 Not available VA-9 Evaluation of Cognitive Functioning in Persian Gulf War Veterans Reporting War-related VAMC New Health Problems Complete Orleans 4/1/94 9/30/95 $49,000 VA-10 Memory and Attention in PTSD VAMC New Ongoing Orleans 2/1/94 9/30/98 $156,065 VA-11 Neuropsychological Functioning in Veterans VAMC New Complete Orleans 2/1/92 3/1/95 Not available VA-12 Psychological Assessment of Operation VAMC New Desert Storm Returnees Ongoing Orleans 8/1/91 9/1/97 Not available VA-13 Neurobehavioral Aspects of Persian Gulf VAMC Experiences: A Pilot Study Complete Pittsburgh 4/1/94 3/31/95 $122,500 VA-15 Vaccine-Mediated Immunity Against VAMC Leishmaniasis Ongoing Cleveland 1/1/93 9/30/99 $315,070 VA-16 Protective Immunity in Experimental Visceral VAMC San Leishmaniasis Ongoing Antonio 10/1/94 9/30/97 $296,205 VA-17 Immunological Evaluation of Persian Gulf VAMC Veterans Complete Birmingham 4/1/94 5/1/95 Not available VA-18 Chronic Gastrointestinal Illness in Persian Gulf Veterans Ongoing VAMC Boston 10/1/94 10/1/96 Not available VA-20 Psychological Adjustment in Operation Desert VAMC Shield/Storm Veterans Complete Gainesville 7/1/91 7/1/93 Not available VA-21 A Comparison of PTSD Symptomatology among Three Army Medical Units Involved in ODS Complete VAMC Phoenix 1/8/92 12/3/94 Not available VA-30 Female Gender and Other Potential Predictors of Functional Health Status Among Persian Gulf War Veterans Ongoing VAMC Boston 9/11/95 3/19/98 Not available VA-36 Stress Symptoms and their Causal Attribution VAMC in Desert Storm Veterans Ongoing Clarksburg 12/1/95 12/31/96 Not available VA-40 Musculoskeletal Symptoms in Gulf War VAMC Long Syndrome Ongoing Beach 1/1/94 1999 Not available VA-46 Diarrhea in Persian Gulf Veterans: An Irritable VAMC Bowel-Like Disorder Ongoing Gainesville 11/1/95 1996 Not available VA-47 Retrospective Validation of Mustard Gas VAMC Exposure Ongoing Louisville 1/1/97 2000 Not available $37,067,821 (Table notes on next page) Page 36 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach a Information on funding was unavailable for some VA intramural projects because these projects were budgeted from medical center overhead. Additional funds may have been obligated for ongoing projects in fiscal year 1997 and later years. b HHS conducted testing on blood samples from persons exposed to the Kuwait oil well fires to assess evidence of exposure to volatile organic compounds. The cost incurred for this work was $33,000. c Federal costs only. The vast majority of federal research was initiated during or after 1994, Federal Research Was and relatively few of these studies have been completed. Seventy-two Delayed and Lacked a projects (79 percent) were ongoing when we reviewed them in early 1997. Proactive Approach Figure III.1, which shows the proportion of completed studies, is based on the rate of progress toward completion of projects based on actual and projected completion dates provided in the most recent research working plan. In some instances, the projected completion dates have not been met. Therefore, figure III.1 slightly overstates the number of projects actually completed. In fact, 28 studies have estimated completion dates prior to 1997, yet only 19 have been flagged as complete in the most recent (April 1997) report to Congress. Page 37 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Figure III.1: Cumulative Percentages of 91 Federally Funded Studies to Be Completed as a Function of Time Percent of 91 studies 100 80 60 40 20 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Begin Estimated completion Note: An additional two studies are slated to continue beyond the year 2000. Source: GAO analysis of information published by the PGVCB. Many federal research efforts were started in association with legislation enacted by Congress. For example, legislation enacted in December 1991 required DOD to establish a registry of troop members who were exposed to fumes from oil well fires and to report annually on its studies of the health effects of such exposure. Legislation enacted in late 1993 Page 38 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach authorized DOD to make grants for studies on the effects of veterans’ exposure to low levels of hazardous chemicals, including chemical warfare agents, and on the effects of exposure to depleted uranium. In 1994, Congress required DOD, in consultation with the VA and HHS, to conduct studies and administer grants for studies to determine the nature and causes of Gulf War illness, including, among others, studies to determine the effects of exposure to pyridostigmine bromide. In 1996, Congress directed DOD to provide for research into the possible exposure of troops to chemical warfare agents or other hazardous materials and the use by DOD of combinations of various vaccines and investigational new drugs. As noted by external reviewers, since federal research goals and objectives were not identified until 1995, after most research activities had been initiated, they appear to reflect a rationalization of ongoing activity rather than a research management strategy. In March 1995, the Department of Veterans Affairs issued a report to Congress, entitled Federal Activities Related to the Health of Persian Gulf Veterans, that identified most of the projects now covered by the PGVCB’s research agenda but noted that “this is a list of activities and is not intended to be construed as a comprehensive research strategy.” Five months later, PGVCB issued A Working Plan for Research on Persian Gulf Veterans’ Illnesses, which linked the previously identified projects to 19 specific research objectives. VA officials acknowledged that the research strategy was articulated in response to criticism from the IOM, which had said that “VA and DOD should determine the specific research questions that need to be answered.” Table III.3 documents other events coincident with changes in research activity. Page 39 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Table III.3: Events Coincident With Changes in PGVCB Research Agenda 1991 1992 1993 1994 1995 1996 Events and Operation Reports of A staff report to the The Defense IOM interim DOD reports Desert “unexpected signs Senate Banking Science Board report criticizes acknowledges Storm and symptoms” Committee states report states that agencies for lack chemical weapon (winter) from the 123rd that “there is illnesses were not of coordination. release at Army Reserve Unit substantial due to exposure to Khamisiyah. in Indiana. evidence chemical (January) supporting claims weapons. (June) that U.S. servicemen and Presidential women were Advisory Committee exposed to low NIH Technology report finds that level chemical Assessment “The government’s warfare agents Workshop finds current research and possibly that (1) a portfolio on Gulf War biological toxins collaborative veterans’ illnesses is from a variety of government- appropriately possible sources. supported weighted toward This exposure may program had not epidemiologic account for many been established studies and studies of the Gulf War and (2) the on stress-related Illness symptoms.” absence of disorders.” well-designed studies had hampered the development of an appropriate case definition. (April) (continued) Page 40 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach 1991 1992 1993 1994 1995 1996 Research Six new Four new studies Seven new studies 39 new studies Eleven new studies 20 new studies program studies started started started (including started started started 14 at the environmental health centers) Four on Three on stress Four on infectious Shift to include a President $2.5 million stress and and one on diseases, one greater range of announces allocated by VA to one each on methods each on stress, oil symptoms and formation of three new projects oil well fires fires, and methods exposures Presidential examining the health and PB Advisory effects of chemical Committee (March) weapons. VA establishes three environmental DOD issues a DOD launches a hazards “Broad Agency $15 million research research centers. Announcement” for program into the (July) research in PB, possible effects of epidemiology and low-level exposure clinical research to chemical agents as mandated by using $10 million Congress in PL made available by 103-337. $5 million Congress and $5 allocated. There million were more than committed by DOD. 100 responses. (September) (June) VA issues a request for proposals for a fourth Environmental Hazards Research Center for Reproductive Outcomes. (May) Clinical care VA develops DOD starts the DOD issues its DOD reports on the programs the Persian Comprehensive report on the first first 18,000 Gulf Health Clinical Evaluation 10,000 participants participants in CCEP Registry Program (CCEP) in CCEP (August) (April) (April) (June) Although research activity has recently been accelerated and broadened, opportunities have been missed to collect critical data that researchers cannot accurately reconstruct. Even efforts to measure the chemical content of the oil-fire smoke, begun only 3 months after the fires began burning, were initiated after most troops had left the affected areas and the climatological dynamics may have been different. Consequently, researchers were forced to use statistical models of the behavior of smoke plumes in order to infer the ground-level exposures experienced by the Page 41 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach large numbers of troops who had departed by the time they began collecting data. Even if such models accurately explain the behavior of the smoke plumes, they have not been validated as measures of individual exposure, and their accuracy for this purpose cannot be presumed. Similar and even more serious problems were caused in the measurement of other exposures by the failure to collect data promptly and maintain adequate records.22 The delay in starting research has also hindered accurate reporting of exposures by Gulf War veterans. Questionnaires are being distributed today (6 years after the war ended) requesting information from veterans on their exposure to certain agents during Operation Desert Storm. Regarding one study, the IOM concluded, “This is a well-designed and well-intended study, but it has started at least several years too late. Recall problems and the inability to obtain accurate information on those who died before the study started are major threats to its validity.” (IOM, Final Report, 1996, p. 91) Early federal research appeared to emphasize risks associated with Some Hypotheses psychological factors such as stress. To support this emphasis, DOD Received Early pointed out that the psychological state of mind can influence physical Emphasis well-being. DOD also pointed to a recent argument that from the American Civil War onward (and perhaps even earlier), a small number of veterans have reacted to the stress of war by suffering symptoms similar to those reported by some Gulf War veterans.23 Of the 19 studies initiated before 1994, roughly half focused on exposure to stress or the potential for posttraumatic stress disorder (PTSD) among returning troops.24 As late as December 1996, the Presidential Advisory Committee (PAC) on Gulf War Veterans’ Illnesses noted that 25 studies centered on stress or PTSD. However, some early research reflected immediate postwar concerns about other issues, for example, the potential effects of the oil fires set by Iraqi troops departing Kuwait and an unusual form of parasitic infection that had been identified in a small number of patients at Walter Reed Army Medical Center (WRAMC). 22 See Defense Health Care: Medical Surveillance Has Improved Since the Gulf War, but Results in Bosnia Are Mixed (GAO/NSIAD-97-136, May 13, 1997) and Institute of Medicine, Final Report, p. 5. 23 K. C. Hyams et al., “War Syndromes and Their Evaluation: From the U.S. Civil War to the Persian Gulf War,” Annals of Internal Medicine, vol. 125 (1996), pp. 398-405. 24 An additional 3 of the 19 studies did not provide information about veterans’ illnesses but were instead building databases or methods to be used in later studies. Notably, according to the PGVCB, none of these 3 studies has yet been completed. Page 42 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach While research on exposure to stress received early emphasis, other Some Hypotheses hypotheses received scant support. In its Final Report, IOM discusses the Were Not Initially evidence for a number of disease hypotheses, including multiple chemical Pursued sensitivity and organophosphate-induced delayed neuropathy (OPIDN). IOM found the evidence for none of the hypotheses to be highly compelling when it conducted the review, but it nevertheless highlighted the importance of exploring “all possible avenues to increase our knowledge of such illnesses and to reduce suffering and disability.” Nonetheless, aside from studies examining stress-related symptoms, relatively few studies have been supported to evaluate alternative disease hypotheses. For example, prior to October 1996, only one study focused on the health effects of potential exposure to chemical warfare agents.25 While multiple studies of the role of stress in the veterans’ illnesses have been supported with federal research dollars, some other hypotheses have been pursued largely outside the federal research program. Although veterans raised concerns about potential chemical exposures soon after the war, and DOD had acknowledged one soldier’s accidental exposure to a mustard agent in 1994, the federal research plan was not modified to include an investigation of concerns about such agents until 1996, when DOD acknowledged potential exposures to chemical agents at Khamisiyah, Iraq. The failure to fund such research cannot be traced to an absence of investigator-initiated submissions. According to DOD officials, three recently funded proposals on low-level chemical exposure had previously been denied funds.26 (See DOD studies 49, 50, and 51 in table III.2) A substantial body of privately funded research suggests that low-level Additional exposure to certain chemical warfare agents or chemically related Hypotheses Were compounds, such as certain pesticides, is associated with delayed or Pursued in the Private long-term health effects. Regarding delayed health effects of organophosphates, the chemical family used in many pesticides and Sector chemical warfare agents, there is evidence from animal experiments, studies of accidental human exposures, and epidemiological studies of humans that low-level exposures to certain organophosphorus 25 This study of the impacts of sulfur mustard agent is a collaborative effort between the Portland Veterans Affairs Medical Center and the Oregon Health Sciences University. The principal investigator for the study pointed out that the possibility of chemical warfare exposure seemed plausible even in 1994 when he sought initial funding for this research. 26 The three previously unfunded proposals address central nervous system targets for organophosphates, development of a DNA-based method for assessing mustard agent exposure, and work on the pharmacokinetics of the nerve agent VX. Page 43 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach compounds, including sarin nerve agents to which some of our troops may have been exposed, can cause delayed, chronic neurotoxic effects.27 This syndrome is characterized by clinical signs and symptoms manifested 4 to 21 days after exposure to organophosphate compounds. The symptoms of delayed neurotoxicity can take at least two forms: (1) a single large dose may cause nerve damage with paralysis and later spastic movement, or (2) repetitive low doses may damage the brain, causing impaired concentration and memory, depression, fatigue, and irritability. These delayed symptoms may be permanent. As early as the 1950s, studies demonstrated that repeated oral and subcutaneous exposures to neurotoxic organophosphates produced delayed neurotoxic effects in rats and mice. In addition, German personnel who were exposed to nerve agents during World War II displayed signs and symptoms of neurological problems even 5 to 10 years after their last exposure. Long-term abnormal neurological and psychiatric symptoms as well as disturbed brain wave patterns have also been seen in workers exposed to sarin in sarin manufacturing plants.28 The same abnormal brain wave disturbances were produced experimentally in primates by exposing them to low doses of sarin.29 Delayed, chronic neurotoxic effects were also seen in animal experiments after the administration of organophosphate.30 These effects include difficulty in walking and paralysis. In recent experiments, animals given a low dosage of the nerve agent sarin for 10 days showed no signs of immediate illness but developed delayed chronic neurotoxicity after 2 weeks.31 27 Sarin has been used as a chemical warfare agent since World War II, most recently during the Iran-Iraq war, and by terrorists in Japan. 28 F. H. Duffy et al., “Long-Term Effects of an Organophosphate Upon the Human Electroencephalogram,” Toxicology and Applied Pharmacology, vol. 47 (1979), pp. 161-176, and F.R. Sidell, “Soman and Sarin: Clinical Manifestations and Treatment of Accidental Poisoning by Organophosphates,” Clinical Toxicology, vol. 7 (1979), pp. 1-17. 29 J. L. Burchfiel et al., “Persistent Effect of Sarin and Dieldrin Upon the Primate Electroencephalogram,” Toxicology and Applied Pharmacology, vol. 35 (1976), pp. 365-379. 30 M. B. Abou-Donia, “Organophosphorus Ester-induced Delayed Neurotoxicity,” Annual Review of Pharmacological Toxicology, vol. 21 (1981), pp. 511-548, and M. K. Johnson, “The Target for Initiation of Delayed Neurotoxicity by Organophosphorus Esters: Biochemical Studies and Neurotoxicological Applications,” Review of Biochemistry and Toxicology, vol. 4 (1982), pp. 141-212. 31 K. Husain et al., “Assessing Delayed Neurotoxicity in Rodents after Nerve Gas Exposure,” Defence Science Journal, vol. 44 (1994), pp. 161-164; K. Husain et al., “Delayed Neurotoxic Effect of Sarin in Mice After Repeated Inhalation Exposure,” Journal of Applied Toxicology, vol. 13 (1993), pp. 143-145; and K. Husain et al., “A Comparative Study of Delayed Neurotoxicity in Hens Following Repeated Administration of Organophosphorus Compounds,” Indian Journal of Physiology and Pharmacology, vol. 39 (1995), pp. 47-50. Page 44 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach It has been suggested that the ill-defined symptoms experienced by Gulf War veterans may be due in part to OPIDN.32 This hypothesis was tested in a privately supported epidemiological study of Gulf War veterans.33 In addition to clarifying the patterns among veterans’ symptoms by use of statistical factor analysis, this study demonstrated that vague symptoms of the ill veterans are associated with objective brain and nerve damage compatible with the known chronic effects of exposures to low levels of organophosphates.34 It further linked the veterans’ illnesses to exposure to combinations of chemicals, including nerve agents, pesticides in flea collars, N,N-diethyl-m-toluamide (DEET) in highly concentrated insect repellents, and pyridostigmine bromide tablets. Toxicological research indicates that agents like pyridostigmine bromide, which Gulf War veterans took to protect themselves against the immediate, life-threatening effects of nerve agents, may alter the metabolism of organophosphates in ways that activate their delayed, chronic effects on the brain.35 Moreover, exposure to combinations of organophosphates and related chemicals like pyridostigmine bromide or DEET has been shown in animal studies to be far more likely to cause morbidity and mortality than any of the chemicals acting alone.36 32 R. W. Haley et al., “Preliminary Findings of Studies on the Gulf War Syndrome,” Presentations to the Intergovernmental Coordinating Board for the Gulf War Illness and the Staff of the Presidential Advisory Committee on Gulf War Veterans’ Illnesses,” September 16, 1995; R. W. Haley, “Organophosphate-Induced Delayed Neurotoxicity,” Internal Medicine Grand Rounds, University of Texas Southwestern Medical Center, Dallas, Texas, October 10, 1996; and G. A. Jamal et al., “The Gulf War Syndrome: Is There Evidence of Dysfuction in the Nervous System?” Journal of Neurology, Neurosurgery and Psychiatry, vol. 60 (1996), pp. 449-451. 33 This research, conducted at the University of Texas Southwestern Medical Center, has been supported in part by funding from the Perot Foundation. 34 R. W. Haley et al., “Is There a Gulf War Syndrome? Searching for Syndromes by Factor Analysis of Symptoms,” Journal of American Medical Association, vol. 277 (1997), pp. 215-222; R. W. Haley et al., “Evaluation of Neurologic Function in Gulf War Veterans: A Blinded Case-Control Study,” Journal of American Medical Association, vol. 277 (1997), pp. 223-230; and R. W. Haley et al., “Self-reported Exposure to Neurotoxic Chemical Combinations in the Gulf War: A Cross-sectional Epidemiologic Study,” Journal of American Medical Association, vol. 277 (1997), pp. 231-237. 35 C. N. Pope and S. Padilla, “Potentiation of Organophosphorus Delayed Neurotoxicity,” Journal of Toxicology and Environmental Health, vol. 31 (1990), pp. 261-273. 36 M. B. Abou-Donia et al., “Increased Neurotoxicity Following Concurrent Exposure to Pyridostigmine Bromide, DEET, and Chlorpyrifos,” Fundamentals of Applied Toxicology, vol. 34 (1996), pp. 201-222, and M. B. Abou-Donia et al., “Neurotoxicity Resulting From Coexposure to Pyridostigmine Bromide, DEET, and Permethrin,” Journal of Toxicology and Environmental Health, vol. 48 (1996), pp. 35-56. Page 45 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Federal Research Emphasis Most Studies Use an Sixty-one of the 91 federally sponsored studies (67 percent) are classified Epidemiological Approach as epidemiological by the Persian Gulf Veterans Coordinating Board. The remaining 30 studies are classified as basic (20 percent), applied (10 percent), and clinical (3 percent) research. Table III.4 shows that the epidemiologic emphasis is present across most major health effects and risk factors under investigation. Table III.4: Number of Studies by Primary Research Focus and Study Type Research type Primary research focus Applied Basic Clinical Epidemiological Total Birth and reproductive effects 0 0 0 4 4 Cancer 0 0 0 1 1 Chemical weapons 1 3 0 1 5 Depleted uranium 0 2 0 0 2 Fibromyalgia 0 0 0 1 1 Gastrointestinal 0 0 0 2 2 Genitourinary 0 0 0 1 1 Immunological 0 0 0 2 2 Infectious diseases 4 3 0 1 8 Methods 1 2 1 6 10 Mortality 0 0 0 2 2 Multiple symptoms/diseases 1 0 0 18 19 Muscular 0 0 0 3 3 Multiple organophosphates (including pyridostigmine 0 6 1 2 9 bromide) Neurological/cognitive 0 0 0 3 3 Oil-well fires 2 0 0 0 2 Pulmonary 0 0 0 1 1 Stress and PTSD 0 2 1 13 16 Total 9 18 3 61 91 Little Research on As indicated in table III.5, federal research is currently centered on studies Treatment of the prevalence, nature, and risk factors associated with veterans’ illnesses. Few studies are focusing primarily on identification and Page 46 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach improvement of treatments for Gulf War veterans’ illnesses. Results of our interviews with principal investigators of ongoing epidemiological projects are generally consistent with this distribution; none of the investigators we interviewed identified the primary goal of his work as developing treatment strategies. Table III.5: Primary Emphasis of 91 Federally Sponsored Research Objective Number Percenta Projects Identified by PGVCB Prevalence 26 29 Nature 17 19 Cause 18 20 Diagnosis 6 7 Treatment 3 3 Methodology 14 15 Combination 7 8 a The individual percentages do not add to 100 due to rounding. Source: GAO analysis of information provided by PGVCB. Descriptive studies are useful for providing information about an illness. But the principal value of doing descriptive studies is to aid in generating hypotheses that, through careful analytical studies, can lead to isolating the nature of the illness and developing treatments. Because so little was initially known about Gulf War veterans’ health, there was a need for descriptive studies. Most of the epidemiological studies thus far have focused on descriptive studies of prevalence. With the exception of the studies that explore the hypothesis that combat stress explains a portion of Gulf War veterans’ symptoms, research has, by and large, been stuck at the beginning of the study cycle presented in appendix I, perhaps partly as a result of a failure to identify hypotheses for further testing, the absence of exposure data, and a failure to identify one or more case definitions. If research on treatments must follow the descriptions of illnesses and causes provided through epidemiological research, then improved treatments for the illnesses afflicting Gulf War veterans might never be found. In 1994, Congress directed DOD and VA to research treatments for ailing Gulf War veterans. Our report shows that such research has largely not taken place, even though more focused research can be done without having first answered general descriptive questions. Page 47 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Our review indicated that most of the ongoing epidemiological studies Formidable focusing on the prevalence or causes of Gulf War-related illnesses have Methodological been hampered by data problems and methodological limitations and Problems consequently may not be able to provide conclusive answers in response to their stated objectives, particularly in identifying risk factors or potential causes. Problems With Prevalence All but one of the research objectives identified by PGVCB (as noted earlier Studies in table III.1) concern establishing the prevalence of symptoms, exposures, morbidity, or mortality. In fact, the PGVCB research plan states that “the most important question about the health of Persian Gulf veterans is: Are Persian Gulf veterans experiencing a greater prevalence of symptoms and illnesses in comparison with an appropriate control population?” The research plan suggests that the direction of additional exploration is contingent on the answer to this question (for example, greater priority will be given to investigating excess health outcomes). It should be noted that Gulf War veterans, even in theater, may have experienced broadly different sets of circumstances and exposures. For example, according to press reports, none of the French troops have complained of similar illnesses. Some notable differences were that French forces were not in the same places as the other allied forces; the French camps were not sprayed with insecticides; and the French did not vaccinate against anthrax, take preventive measures against botulinum toxin, or administer pyridostigmine bromide. None of the federally funded studies used French troops as a comparison group. In contrast, most of the ongoing studies designed to assess the prevalence of various conditions of Gulf War veterans and others were making broad comparisons between deployed and nondeployed veterans, rather than specific types and levels of exposures. For example, our interviews found that 12 of 13 ongoing cohort studies had defined the exposed cohort with reference to nothing more than deployment status. That is, in almost all cases, the exposure of interest was defined simply as “Gulf War service,” and the prevalence of symptoms or illnesses among Gulf War veterans is being compared to the prevalence of symptoms or illnesses among troops who were not deployed to the Gulf. Such comparisons may have value for providing basic assurances to veterans regarding widespread and severe health consequences of Gulf War service. However, many service-connected illnesses could be obscured by broad comparisons of deployed and nondeployed veterans Page 48 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach without regard to their specific exposure histories. At the same time, illnesses that were not actually service connected could appear to be linked to deployment status due to preexisting group differences. For example, some troops were not deployed for health reasons, potentially biasing the comparison group in the direction of greater illness. Also, due to the failure to compare the prewar health of the groups, the absence of differences is no assurance that one of the groups has not experienced a significantly steeper decline in health. Some investigators have attempted to address some of the problems of systematic differences between deployed and nondeployed veterans by comparing Gulf War veterans to servicemembers who were deployed to locations other than the Gulf. To the extent that such group differences are measured, they can also be statistically controlled. While these are potentially promising solutions, such comparisons must still be carefully evaluated in the absence of evidence of prior similarity between the groups and greater specificity regarding exposure. Problems With Studies of As we noted earlier, to ascertain the causes of illnesses, it is imperative Risk Factors or Causes of that investigators have valid and reliable methods to collect information Illness on exposures as well as their effects. The need for accurate, dose-specific information is particularly critical for low-level or intermittent exposure(s) to drugs, chemicals, or biological agents. In addition, the investigators must specify diagnostic criteria to (1) reliably determine who has the disease or condition being studied and who does not and (2) select appropriate controls (people who do not have the disease or condition). To the extent that individuals are misclassified regarding disease or exposure, conclusions would be misleading and relationships would be obscured. Measurement of Exposure Is The research program to answer basic questions about the illnesses that Problematic afflict Gulf War veterans has at least three major problems in linking exposures to observed illness or symptoms. First, it is extremely difficult to gather information about the unplanned exposures (for example, oil-fire smoke and insects) that may have occurred in the Gulf, and DOD has acknowledged that records of planned or intentional exposures (for example, the use of vaccines and pyridostigmine bromide to protect against chemical/biological warfare agents) were inadequate. Second, the veterans were typically exposed to a wide array of agents with commonly accepted health effects, making it difficult to isolate and characterize the effects of individual factors or study their combined effects. Third, the Page 49 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach passage of time following these exposures has made it increasingly difficult to have confidence in any information gathered through retrospective questioning of veterans.37 In part, the latter difficulty was created by the delayed release of information about detection of chemical warfare agents during the war as well as the delayed collection of exposure data. Five years passed before it was acknowledged that American soldiers may have been exposed to chemical warfare agents shortly after the war ended in 1991 (at the Khamisiyah site). Moreover, although chemical detections by Czech forces have been deemed “credible” by the Central Intelligence Agency (CIA), the source of these detections remains unknown. In the face of denials by DOD officials, a few researchers told us that they had considered it pointless to pursue hypotheses that the symptoms may have been associated with exposure to chemical weapons. When we asked investigators responsible for ongoing federally funded epidemiological projects about how they were collecting data on the various factors to which Gulf War veterans may have been exposed, we found that most projects had no means other than self-reports for measuring most of the factors to which troops may have been exposed. (See table III.6.) This reliance on self-reports was present even for exposures such as vaccines for which records might have existed. 37 Large numbers of veterans questioned during their participation in the VA’s revised health registry examination program reported they did not know whether they were exposed to certain agents. “Don’t know” responses were greatest for nerve gas (64.9 percent), mustard gas (60.2 percent), depleted uranium (52.5 percent), chemical-agent resistant coating (47.8 percent), microwaves (32.8 percent), paints or solvents (24.9 percent), and pyridostigmine (21.1 percent). To the extent that a response of some kind reflects greater certainty, veterans were more confident in their reports regarding smoke from tent heaters, passive smoking, diesel or other petrochemical fumes, skin exposure to fuel, pesticides in cream or spray form, and burning trash or feces, each of which resulted in fewer than 11 percent of respondents reporting “don’t know.” However, the provision of a response does not necessarily connote that the reports are accurate. Page 50 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Table III.6: Ongoing Epidemiological Studies Using Measures Other Than Is the exposure measured Self-Reports to Assess Key Exposures Is the study collecting data through any means other on this exposure?a than self-report?b Exposure Yes No No response Yes No No response CARC 10 10 2 1 7 2 Biological warfare agents 10 9 3 4 3 3 Depleted uranium 14 6 2 2 8 4 DEET 11 7 4 1 7 3 Permethrin 11 7 4 1 7 3 Other pesticides or repellents 12 5 5 1 8 3 Pyridostigmine bromide 15 5 2 0 11 4 Vaccines 13 5 4 6 4 3 Petroleum products 14 5 3 4 7 3 Oil-fire smoke 16 5 1 3 11 2 war stressors 15 5 2 1 10 4 Infectious diseases 11 7 4 6 5 0 Chemical warfare agents 15 4 3 5 5 5 Note: The survey incorporated responses from 31 of the 43 studies identified as ongoing epidemiological studies by PGVCB in its November 1996 plan. Of these, 22 indicated they were collecting exposure information. a Among the 22 collecting any exposure data. b Among those collecting data on the exposure named in the first column. Source: GAO’s survey of investigators charged with ongoing epidemiological studies. There are three problems associated with reliance on self-reports for exposure assessments. First, recalled information may be inaccurate after such a long time period; that is, some veterans may not remember that they were exposed to particular factors, while others may not have been exposed but nonetheless inaccurately report that they were. Second, recollections also may be biased if, for example, veterans who became sick following the war recall their exposures earlier, more often, or differently than veterans who did not become sick. Third, there is often no straightforward way to test the validity of self-reported exposure information, making it impossible to separate biased recollections from actual differences in exposure frequency. Page 51 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach Some investigators are also relying on a model developed by the U. S. Army Environmental Hygiene Agency for assessing exposures to components of oil-fire smoke through the combination of unit location data and information from models of the distribution of oil-fire smoke. However, this method requires the use of unit location as a proxy for exposure, and the validity of this approach is unknown. As PAC noted, DOD’s Persian Gulf Registry of Unit Locations “lacks the precision and detail necessary to be an effective tool for the investigation of exposure incidents.” (See PAC’s Final Report (Washington, D.C.: GPO), p. 35.) Case Definition Is Complicated Another major hurdle to the development of a successful research agenda by Presence of Nonspecific has been the difficulty in classifying symptoms into one or more distinct Symptoms illnesses. Some veterans complain of gastrointestinal pain, others report musculoskeletal pain or weakness, and still others report emotional or neurological symptoms. As explained previously, a specific case definition is essential to conducting certain types of epidemiological studies. Although some data on symptoms were collected beginning in 1992 with the initiation of the VA registry, initial efforts to collect information about symptoms and exposures from registry participants were limited and nonspecific, constraining their potential use for improving understanding of the patterns of veterans’ complaints. The limitations in early registry data are unfortunate insofar as detailed information about symptoms and exposures might have yielded earlier, more reliable analyses of the nature and causes of veterans’ complaints that could have also assisted in arriving at working case definition(s). Furthermore, clinical effects of a transitory nature that may have been manifested soon after the war would have been missed due to delays in setting up and developing studies and registries. We also found that both the federally supported projects and the federal registry programs have generally failed to study the conjunction of multiple symptoms in individual veterans. Articles and briefing documents that we have obtained report findings that address the incidence of single symptoms and diagnoses. There are two exceptions. First, the Center for Disease Control (CDC) and Prevention developed an operational case definition, which is quite similar to the case definition of chronic fatigue syndrome. Obviously, this definition cannot be generalized beyond the population from which it was derived. Second, the studies conducted by Haley et al. also focused on identifying symptom clusters. For those ongoing epidemiological projects that are built on case-control designs, we inquired about how a case was defined. The specificity of this Page 52 GAO/NSIAD-97-163 Gulf War Illnesses Appendix III Federal Strategy to Research Gulf War Illnesses Lacks a Coherent Approach definition is important because a vague case definition can lead to considering multiple kinds of illness together. When this is done, it is not surprising to find no commonality of experience among the cases. Moreover, the use of specific case definition is particularly critical to achieving meaningful results within this type of research design. However, in the ongoing studies we surveyed, case definition was quite broad, even among studies that depended upon case-control research designs. For example, among 13 case-control or nested case-control studies, case definitions included such broad descriptors as registry participants, Gulf War veterans who are symptomatic without diagnosable illness, and veterans with complaints of chronic fatigue and muscle weakness. Sample Size Most investigators we interviewed in our survey took steps to estimate the size of sample they would require to have a reasonable expectation of detecting differences between deployed and nondeployed veterans or exposures to hazardous substances. However, many variables are involved in such calculations, for example, the size of the investigated exposure’s expected impact on health (consistent lethal effects can be detected in a smaller sample than more subtle problems) and the prevalence of exposure, some of which were unknown at the time the studies were planned. Thus, they had to be estimated within somewhat broad parameters. Although steps were clearly taken to plan for an adequate sample size, some investigators reported difficulty in locating subjects due to factors beyond their control, such as the rate of referrals from VA examination centers or the rate of identification of subjects that fit highly specific case definitions. Moreover, other studies, such as those on specific birth defects, require extremely large samples. An investigator on a principal study of birth defects indicated that the number of births to Gulf War veterans and problems with data collection would mean that data would not be sufficient to draw conclusions about a particular defect (Goldenhaar syndrome) for 6 years or more. Page 53 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations A key measure of the effectiveness of a research program is the extent to which it has yielded verifiable conclusions regarding the subject of study. We previously reviewed findings contained in the November 1996 revision of A Working Plan for Research on Persian Gulf Veterans’ Illnesses and concluded that PGVCB had formed few strong conclusions based on the research that it had sponsored and coordinated. To gauge the extent of knowledge about Gulf War illnesses, we also reviewed other recent documents and spoke to VA and DOD officials to determine what would represent the best statement of conclusions. This review indicated that the most extensive and detailed review of the evidence about Gulf War illnesses was done by the Presidential Advisory Committee on Gulf War Veterans’ Illnesses. The 12-member Committee held 18 public meetings between August 1995 and November 1996 before reaching its conclusions. In its final report, PAC presents its conclusions about the likelihood that 10 commonly cited exposure agents have contributed to the explained and unexplained illnesses being suffered by Gulf War veterans. (See table IV.1.) The PAC report was reviewed by DOD, which endorsed many of the findings.38 Table IV.1: PAC Conclusions on Health Effects of Different Individual Exposure Agents Exposure agent PAC’s conclusion Reasons Our assessment Biological warfare “It is unlikely the health effects “There were no verified detections We have noted the limitations of the agents reported today by Gulf War of anthrax or botulinum toxin during U.S. detection capability for veterans are the result of the war. Second, stateside biological warfare agents. We exposures to biological warfare examination of soil samples and agree with PAC that the effects of agents” enzyme assays did not reveal the at least one of the agents that Iraq presence of BW agents.” weaponized might not be observed for many years. (continued) 38 In endorsing PAC’s conclusion that it is “unlikely” that the symptoms and diseases are due to exposure to agents during the Gulf War, DOD also noted that “there may still be small groups of Gulf War veterans that may have illnesses related to exposures during the Gulf War [and that DOD] will continue...our clinical investigation and research efforts.” According to PAC, VA, HHS, veterans’ service organizations, and individual veterans and veterans’ advocates also reviewed its report. However, PAC did not provide information on the extent to which these reviewers agreed with its findings, or whether it incorporated their comments in its reports. Page 54 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations Exposure agent PAC’s conclusion Reasons Our assessment Chemical warfare “It is unlikely the health effects “Available scientific evidence does We dispute this conclusion. There agents reported by Gulf War veterans not indicate that such long-term is evidence from various sources today are the result of exposure to effects occur in humans following that chemical weapons were OP or mustard chemical warfare low-level exposures, but the released at Khamisyah and agents during the Gulf War.” amount of data from either human elsewhere on the battlefield. Some or animal research on low-level evidence from animal and exposures is minimal.” epidemiological studies documents the potential for delayed or chronic effects from such exposure. Thus, we cannot exclude the possibility that such health effects could impact exposed veterans. Depleted uranium “It is unlikely that health effects “Toxic effects are likely to be similar We have no comment on this issue.a reported by Gulf War veterans to the kidney toxicity observed from today are the result of exposure to inhaled or ingested uranium. To depleted uranium during the Gulf date, VA has reported no kidney War.” toxicity among soldiers wounded by DU fragments in friendly fire episodes.” Infectious diseases “It is unlikely that infectious “While viscerotropic leishmaniasis Owing to the invasive character of diseases endemic to the Gulf can be difficult to confirm, it is not current screening tests for region are responsible for long considered to be a cause of viscerotropic leishmaniasis it has term health effects in Gulf War widespread illness in Gulf War been impossible to test broadly for veterans, except in a small, known veterans. All veterans diagnosed infection. Although some sources number of individuals.” with viscerotropic leishmaniasis, have suggested that the rate of except one, have experienced the leishmania infection may be as signs characteristic of the disease. high as 5% of certain groups From August 1990 through July deployed to the Persian Gulf, there 1991, the U.S. Army deployed is currently no means of screening approximately 347,000 individuals for asymptomatic infections which to the Gulf region. Based on can re-emerge during immune information from U.S. Army field system failure. The Center for hospitals, the only infectious Disease Control and Prevention diseases that caused 30 or more has found evidence of previous Q each of approximately 14,000 fever and sandfly fever infection in admissions were pneumonia, a subsample of Gulf War veterans, intestinal infections, inflammation of which would indicate exposure to the testes and/or epididymus, the sandfly that caries leishmania. chicken pox, and kidney infections.” Oil-well fire smoke “It is unlikely exposure to oil-well “Toxic gases that can be found in We have no comment on this issue.a fire smoke is responsible for oil-well fire smoke-such as symptoms reported today by Gulf hydrogen sulfide and sulfur War veterans.” dioxide-can cause eye and nose irritation, decreased pulmonary function, and increased airway reactivity. These toxic gases were not detected at high levels during the fires.” (continued) Page 55 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations Exposure agent PAC’s conclusion Reasons Our assessment Pesticides “It is unlikely that health effects and “According to DOD, after-action Our review of the literature symptoms reported today by Gulf reports from in-theater medical identified evidence that exposure War veterans are the result of personnel did not reveal any U.S. to organophosphate agents can exposure to pesticides during the troops reporting symptoms that induce delayed neuropathy without Gulf War” would indicate pesticide poisoning. causing immediate symptoms. Evidence from studies of humans Moreover, it has been suggested poisoned by organophosphate that treatment with pyridostigmine pesticides suggests that low-level bromide following exposure to exposures that do not cause signs organophosphates (either OP and symptoms of immediate and pesticides or chemical weapons) severe poisoning will not result in may actually enhance the potential long-term health effects.” for delayed effects. Petroleum products “It is unlikely that health effects We have no comment on this issue.a reported today by Gulf War veterans are due to exposure to petroleum products during the war.” Psychological and “Stress is likely to be an important “Animal studies demonstrate that Although the evidence that we physiological stress contributing factor to the broad stress can have measurable effects reviewed indicates that stress can range of illnesses currently being on the brain, immune system, have an important role in symptoms reported by Gulf War veterans.” cardiovascular system, and various of many physical illnesses, when “The entire federal research hormonal responses. Although the stress is present in a patient with portfolio should place greater human body can adapt to normal untreated and undiagnosed diffuse emphasis on basic and applied stresses, if the stress lasts longer it physical symptoms, care must be research on the physiologic effects can be expressed in a variety of taken to determine whether the of stress and stress-related physical illness symptoms. Some stress is the cause or the effect of disorders.” researchers suspect that the the physical symptoms. We found inadequate production of stress weak support for the conclusion hormones and stress response that stress is an important occurs in some (not all) humans contributing factor in the broad with chronic fatigue syndrome and range of illnesses being reported post-traumatic stress disorder. by Gulf War veterans; most of the Based on this understanding and evidence cited by PAC addressed supported by decades of clinical the effects of stress solely on PTSD. observations, physicians recognize that many physical, as well as psychological, diagnoses are the consequences of stress.” (continued) Page 56 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations Exposure agent PAC’s conclusion Reasons Our assessment Pyridostigmine bromide “It is unlikely that health effects PB is used in much higher doses in Experiments in animal models reported today by Gulf War patients with myasthenia gravis (including one study sponsored by veterans are the result of exposure than was administered to military DOD) show that PB has toxic simply to PB (emphasis added). personnel. effects in combination with other Ongoing federally funded studies elements, such as DEET and should help the scientific permethrin, in the Gulf War community draw conclusions about environment. This may be the synergistic effects of PB and particularly true for animals with a other risk factors.” genetic predisposition. Cases of such delayed neurotoxic effects in humans exposed to PB and DEET have been epidemiologically inferred and reproduced in hens. We note that PB was intended for use only when other agents were believed to be present or imminent. PB remains classified as an investigational new drug for the purposes for which it was used in the Gulf War. Vaccines “It is unlikely that health effects “The human immune system has DOD has not adequately monitored reported by Gulf War veterans evolved the capability to deal with the effects of receiving multiple today are the result of exposures to thousands of foreign substances, vaccines. the BT or anthrax vaccines, used to sort them out, and to regulate alone or in combination.” immune response. Humans live among a vast population of hostile microorganisms, and vaccinations—even multiple, contemporaneous vaccinations—are a small part of total immune stimulation. Individual vaccines can cause adverse effects, but several studies of the effects of giving multiple vaccinations at one time have found no adverse effects associated with the practice.” a This does not mean that we believe that it is not a risk factor. PGVCB has stated that “some symptoms may be related to PTSD. Published Extent of findings suggest an increased prevalence of PTSD and other psychiatric Posttraumatic Stress diagnoses, such as depression in some Persian Gulf veterans....stressors Disorder May Be during the Persian Gulf conflict were sufficient to cause significant psychiatric morbidity.”39 In testimony before the House Appropriation Overestimated Committee, the Assistant Secretary of Defense for Health Affairs has stated that 39 PGVCB, A Working Plan for Research on Persian Gulf Veterans’ Illnesses (Nov. 1996), p. 36. Page 57 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations “one of the most striking findings of our clinical work has been the recognition of psychological conditions and stress-related symptoms as a major diagnostic category among veterans cared for in our facilities. Our clinicians have been impressed that stress experienced during the Gulf War and in its aftermath appears to be a major contributing factor in the development of psychological conditions as well as the manifestation of symptoms associated with non-psychological conditions.” Similarly, PAC has stated that “epidemiological studies to assess the effects of stress invariably have found higher rates of PTSD in Gulf War veterans than among individuals in nondeployed units or in the general U.S. population of the same age.”40 However, the studies to which PAC refers have not excluded other conditions that produce symptoms similar to PTSD and can also elevate scores on key measures of PTSD. Although the reported rates of PTSD in various studies range from 4 to 32 percent, these rates were based on widely different populations, with high rates of nonparticipation, and little information on selection bias. Moreover, as with most scales and tests, a certain number of people will test positive on any given measure of PTSD even though they do not have PTSD; they may have a related disorder or no disorder at all. Based on the large numbers of individuals to whom these scales were administered, such false positives may be a significant portion of all those who obtained scores indicative of PTSD. In a CDC-sponsored study of Iowa veterans that achieved a 76-percent response rate and used a relatively inclusive criterion for identification of presumptive PTSD, observed rates were quite low, although they were higher among Gulf-deployed than nondeployed veterans.41 Only 15 percent of the diagnoses categorized as psychological (according to the International Classification of Diseases-9th Revision (ICD-9)) among CCEP registrants are clear cases of PTSD. Owing to the breadth and heterogeneity of ICD-9 categories used to report CCEP data, high percentages of primary or secondary “psychological conditions” are reported, but the most frequently diagnosed “psychological condition” was tension headache. Investigators from the Department of Military Psychiatry at WRAMC reported, “The major conclusion concerning physical health of these veterans is that for those who deployed to the Gulf War and currently report physical symptoms, neither stress nor exposure to combat 40 PAC, Final Report (Dec. 1996), p. 79. 41 Iowa Persian Gulf Study Group, “Self-reported Illness and Health Status Among Gulf War Veterans: A Population Based Study,” Journal of the American Medical Association, vol. 227 (1997), pp. 238-245. Page 58 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations or its aftermath bear much relationship to their distress; only the fact of deployment differentiates them from their less-burdened counterparts.”42 Alternative causes for stress-related symptoms may not have been fully explored. For example, just following the war, experts from Walter Reed Army Institute of Research (WRAIR) and WRAMC noted, “Sandfly fever (phlebotomus fever)....has caused substantial epidemics in foreign military forces in the Middle East. It is an acute, self-limited viral disease with a course of two to five days and an incubation period of less than one week, whose acute manifestations will be unlikely in those who have returned from the region. Convalescence, however, is frequently complicated by depression, fatigue, and weakness that can last months. The evaluation of a chronic fatigue or post-traumatic stress-like syndrome in those who have returned from the Persian Gulf should therefore include serologic testing to rule out an earlier sandfly fever virus infection.”43 Such serologic testing is available only from CDC, in Fort Collins, Colorado, and from U. S. Army Medical Research Institute of Infectious Disease (USAMRIID) in Fort Detrick, Maryland. Thus, it is unlikely that testing has been broadly done to assess veterans’ fatigue symptoms. However, a CDC analysis of blood taken from 158 volunteer Pennsylvania Air National Guardsmen found that 5.7 percent showed evidence of previous sandfly fever infection. For various reasons, including false positives and the absence of preexposure blood samples for comparison, such evidence can be difficult to interpret but suggests the importance of reviewing alternative explanations for diagnoses of PTSD and chronic fatigue syndrome. Although widely cited work has argued that ill-defined syndromes have been observed following many previous military conflicts, it is difficult to compare current and historical findings due to differences in the diagnostic capabilities previously available.44 It is highly likely that these historical groups contained a mix of ailments that would now be differently diagnosed. Moreover, even if these postwar syndromes contained overlapping symptoms, it is not a foregone conclusion that commonalities reflect the common experience of stress. 42 R. H. Stretch et al., “Physical Health Symptomatology of Gulf War-era Service Personnel From the States of Pennsylvania and Hawaii”, Military Medicine, vol. 160 (1995), pp. 131-136. 43 R. A. Gasser et al., “The Threat of Infectious Disease in Americans Returning From Operation Desert Storm,” The New England Journal of Medicine, vol. 324 (1991), p. 862. 44 K. C. Hyams et al., “War Syndromes and Their Evaluation: From the U.S. Civil War to the Persian Gulf War,” Annals of Internal Medicine, vol. 125 (1996), pp. 398-405. Page 59 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations PGVCB concluded that “the likelihood of Leishmania tropica as an Extent of important risk factor for widely reported illness has diminished.”45 While Asymptomatic this is the case for observed symptomatic infection with the parasite, the Leishmania Infection prevalence of asymptomatic infection is unknown, and such infection may reemerge in cases in which the patient’s immune system becomes Is Unknown deficient. Leishmaniasis is an infectious disease caused by a microscopic parasite that invades certain types of white blood cells. While leishmaniasis occurs in Southwest Asia and certain other parts of the world, it is very rarely seen in the United States. The disease is transmitted by sandflies, and a number of different leishmania species are known to infect humans. Personal protective methods are relatively less effective against sandflies than against mosquitoes. Sandfly populations were monitored during the Gulf War and were found to be high from August to November 1990 and again from April to June 1991. Forms of disease that involve low levels of parasite infection can be particularly difficult to diagnose using currently available methods. According to briefings we received by experts at WRAIR, accurate diagnosis of leishmaniasis is important because effective treatment involves the use of potentially toxic drugs currently being investigated as new drugs and not yet approved by the Food and Drug Administration. They noted that such diagnosis is problematic because • most clinicians would fail to recognize classic forms of leishmaniasis, much less atypical clinical presentations; • accurate laboratory diagnosis of suspected cases (detection of parasites in biopsy or culture) is not available to most physicians; and • blood tests can provide supportive evidence of infection but cannot be used alone to establish a diagnosis of leishmaniasis. While blood testing for leishmania infection is problematic, it is the only means currently available of assessing the potential prevalence of such infection. In testing blood collected since the war from 158 Air National Guardsmen, CDC researchers reported positive results for exposure to 45 PGVCB, A Working Plan for Research on Persian Gulf Veterans’ Illnesses (Nov. 1996), p. 20. Page 60 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations leishmania donovani in 4.9 percent and leishmania tropica in 4.3 percent.46 Most of these individuals were also among the 5.7 percent showing evidence of exposure to the sandfly vector that carries leishmania through positive results on a well-characterized test for sandfly fever. However, the CDC sample was composed of Air National Guardsmen who volunteered for a particular study and were deployed from the same area, so the tests do not represent estimates of the prevalence of the infection for Gulf War veterans at large. The study also found no clear association between results for leishmania infection and the presence of a set of symptoms characteristic of chronic fatigue syndrome.47 Although PGVCB officials told us that the symptoms typical of leishmaniasis, including enlargement of the liver, were not being observed, not all ill veterans would show such symptoms. In commenting on a report on a new form of leishmaniasis, CDC noted that five Gulf War veterans had been diagnosed with the infection, even though their symptoms were nonspecific, and none had the marked symptoms typical of visceral leishmaniasis.48 Approval continues to be pursued for a skin test to assess the prevalence of asymptomatic infection. 46 In 1991, tests were run on blood samples from 119 military working dogs that had been in Saudi Arabia. Five dogs (4.2 percent) were positive for the disease. One of these dogs subsequently developed the infection, which was confirmed by autopsy. Symptomatic disease and demonstrated infection have been observed in individuals with serological titers of 1:16. While none of 50 Marines showed a result at this level before deployment, tests of 488 Desert Storm veterans conducted after the war showed 5 percent had results of 1:32 or higher. Roughly 5 percent of a sample of troops tested after the initial identification of viscerotropic leishmaniasis showed positive results using a skin test involving a slightly different parasite. However, few of those who tested positive were symptomatic, and the accuracy and appropriateness of the tests for this purpose is controversial. Finally, a Seattle organization attempting to develop a test for viscerotropic leishmaniasis has reportedly found positive responses among asymptomatic subjects. WRAIR officials view this test as a highly specific indicator of exposure to leishmania tropica, but not a specific indicator for the type of the parasite associated with viscerotropic infection. 47 Based on concerns about the potential for transmission of this disease through the blood supply, blood donations were temporarily deferred for all Gulf War veterans returning from Southwest Asia since August 1, 1990. The blood donation ban was lifted on January 1, 1993. However, an accurate and noninvasive screening test for this form of leishmaniasis remains commercially unavailable. Although a study of transfused animals has demonstrated that the parasite retains its infectivity under blood bank conditions, in lifting the ban, DOD officials observed that there had been no documented case of transfusion-acquired leishmania tropica. 48 “Viscerotropic Leishmaniasis in Persons Returning from Operation Desert Storm—1990-1991,” MMWR, vol. 41, pp. 131-134. Reprinted in Journal of American Medical Association, vol. 267(11) (1992), pp. 1444-46. Page 61 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations DOD has consistently denied that Gulf War veterans were intentionally or Evidence of Exposure unintentionally exposed to biological warfare agents, and prior to to Biological and June 1996, it denied any exposure to chemical warfare agents. If Chemical Weapons servicemembers were exposed, exposure would have occurred in one of three ways: (1) through intentional Iraqi use of chemical or biological Has Not Been warfare agents, (2) through theaterwide contamination resulting from air Aggressively Pursued war bombings of Iraq, or (3) through site-specific events. As has been pointed out by the Presidential Advisory Committee, the United States currently has no system that can detect and identify biological warfare agent aerosols rapidly enough to enable troops to take protective measures. Regarding chemical warfare agents, while the United States has a detector/alarm system, according to DOD, it is not as sensitive as some other systems, such as those operated by Czechoslovakian coalition partners. DOD has taken the position that chemical and biological agent exposures can be confirmed only through evidence of mass incidents of morbidity and mortality. Since there were no such instances, DOD asserted that Gulf War veterans were not exposed. Biological Warfare Agents According to the CIA, the Presidential Advisory Committee, and others, the Iraqis had weaponized several biological agents at the time of the Gulf War, including Bacillus anthracis, Clostridium botulinum, and aflatoxin.49 Apart from aflatoxin (a potent liver carcinogen), these agents are known to have immediate and life-threatening toxic effects. Although the United States took steps to vaccinate troops against anthrax and botulism, according to PAC, “after the war, new data revealed that Iraq had also weaponized aflatoxin.” This agent’s effects may not be observed until decades after low-level exposure via ingestion, and the effects of aerosolized aflatoxin are poorly understood. PAC notes that any effects (notably liver cancer) from exposure to aflatoxin would not be expected until several years passed. PAC also recommended that DOD and VA monitor the Gulf War veteran population. PAC reviewed U.S. Army hospital admission records and identified only one admission for anthrax (a disease indigenous to the Gulf region) and none for botulism. In addition, although Navy and Army researchers tested over 800 pairs of prewar and postwar blood samples from Navy Seabees for antibody to anthrax, they found no evidence of acute infections. While many blood samples showed evidence of vaccine-induced immunity, only 49 J. D. Walker, “Biological Weapons: Attempts to Verify” In Ranger, R. (Ed.) (1996). The Devil’s Brews I: Chemical and Biological Weapons and Their Delivery Systems (Lancaster, UK: The Center for Defence and International Security Studies), pp. 36-8. Page 62 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations one showed evidence of exposure to the wild antigen or similar bacteria. PAC reported that other evidence it examined also failed to support the notion that biological weapons were used. The PAC report documents that Iraq had weaponized aflatoxin. In our discussions with agency officials, the potential use of aflatoxin was dismissed because it would not immediately incapacitate coalition forces and would therefore have no strategic value. Prior to the war, the United States told Iraq that any use of biological or chemical weapons on coalition forces would have devastating consequences for Iraq. The United States did not deploy a real-time detection system for biological weapons.50 Therefore, one cannot be certain that such weapons were not used, particularly since the United Nations Special Commission on Iraq (UNSCOM) has not been able to confirm Iraq’s self-declared destruction of these weapons.51 Similarly, a USAMRIID official indicated that tests were not available to detect low-dose (i.e., asymptomatic) exposures to various biological agents that the Iraqis had weaponized. While biomarkers may be available for exposure to some of these agents, interpreting the results of such testing in the absence of symptoms is complex, and little such testing appears to have been done. Chemical Warfare Agents As with exposures to biological weapons, there were no massive incidents of mortality or morbidity observed in theater that were consistent with known acute effects of exposure to chemical warfare agents. The U.S. Army officer responsible for medical surveillance of chemical/biological warfare agents during the war has testified to the PAC that only one accidental casualty was treated. However, it is important to note that detections of the nerve agent sarin occurred on January 19, 1991, and of mustard gas on January 24, 1991, by coalition partners from Czechoslovakia in areas near Hafir al Batin. DOD has verified the reliability 50 The Army fielded the interim Biological Integrated Detection System (BIDS) in September 1996. A total of 38 systems have been produced, with a total of 35 located collectively with the 310th Army Reserve Chemical Company (Biological Detection) and the 20th BIDS Detachment (Active Army), at Ft. McClellan, AL. The current BIDS can detect and identify up to four biological agents at a time in 45 minutes. Future improvements are expected to enable BIDS to detect and identify more agents in less time. (Sources: Chemical and Biological Defense: Emphasis Remains Insufficient to Resolve Continuing Problems (GAO/NSIAD-96-103, Mar. 29, 1996), p.6; Chemical and Biological Defense: Protection of Critical Overseas Ports and Airfields Remains Largely Unaddressed (GAO/NSIAD-97-9, June 13, 1997), pp. 20-21. 51 See Chemical and Biological Defense: Emphasis Remains Insufficient to Resolve Continuing Problems (GAO/NSIAD-96-103, Mar. 29, 1996). Page 63 GAO/NSIAD-97-163 Gulf War Illnesses Appendix IV Support for Key Official Conclusions Is Weak or Subject to Different Interpretations of the Czech equipment but has never identified the source of these detections, although both DOD and CIA have deemed the detections credible. One cannot rule out the possibility that these detections were the result of fallout from coalition bombings. During late January and February 1991, DOD records indicate that coalition forces successfully conducted a series of aerial bombings on suspect nuclear, biological, and chemical weapons storage and production sites. UNSCOM has not inspected all suspect and targeted sites. As a result, the magnitude of exposures to chemical warfare agents has not been fully resolved. With regard to site-specific exposures identified at Khamisiyah, uncertainties surround the extent of potential exposure. A contractor for CIA had attempted to model the dispersion of chemical warfare agents. But the uncertainties were too great to complete the model. These uncertainties stem from (1) the lack of pertinent meteorological data; (2) gross uncertainties about the amount of chemical warfare material present at the time of demolition; and (3) the behavior of the material on demolition (e.g., vaporization or evaporation) in an open pit. The 1995 PGVCB research plan noted that investigations of chemical Impact of DOD weapons effects were not done because there was no evidence of Denials on Federal exposure.52 Noting that there had been no mass casualties to indicate Research chemical weapons exposure, DOD failed to fund research on the possible long-term health consequences of low-level exposure to chemical warfare agents. In fact, a few researchers told us that, as a result of DOD’s strong position, they believed it would be fruitless to request funding for such research. PGVCB reversed its position in its 1996 plan, following the revelations regarding Khamisiyah. A broad agency announcement seeking research on this issue was subsequently issued and some work has been commissioned. Experts in the field of toxicology told us that had such information been made available earlier, the direction and outcome of research would have been different. 52 We could not assess this statement, as relevant data were not available for us. Page 64 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense Note: GAO comments supplementing those in the report text appear at the end of this appendix. See comment 1. Page 65 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense See comment 2. Page 66 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense See comment 3. Page 67 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense See comment 3. Page 68 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense See comment 4. See comment 5. See comment 6. See comment 7. Page 69 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense See comment 8. Page 70 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense Page 71 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense Page 72 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense Page 73 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense The following is GAO’s response to the Department of Defense’s (DOD) comments, dated June 9, 1997. 1. DOD offers selected excerpts from reports of the Institute of Medicine GAO Comments (IOM) and the Presidential Advisory Committee (PAC) that leave the impression the reports were uncritical of its actions, but this was not the case. These reports point out multiple problems and contain numerous recommendations for improvement. 2. The national defense authorization act for fiscal year 1997 requested that we conduct an independent and objective review of federal clinical care and medical research efforts into Gulf War illnesses. That directive included gathering and analyzing information and coming to our own conclusions on the matters under review. Our information sources included previous reports, such as those by the IOM Committee to Review the Health Consequences of Service During the Persian Gulf War and the Presidential Advisory Committee on Gulf War Veterans’ Illnesses. However, the conclusions presented in our report are ours, and not those of other bodies. DOD’s assertion that our assessment was somehow less careful or thoughtful than those provided by the PAC and IOM is groundless. 3. DOD’s comments do not address our specific finding that it has no information on whether Gulf War veterans are any better or worse today than when they were initially diagnosed. DOD suggests that its current approach provides adequate oversight for Gulf War veterans’ care but then indicates that it is reviewing a draft proposal on health outcome measures. We found that DOD relies on quality assurance mechanisms that do not ensure a given level of effectiveness for the care provided. Given the fact that DOD has no way to track changes in veterans’ health status, we continue to believe that DOD and VA should develop and implement plans to monitor the clinical progress of veterans. 4. DOD incorrectly infers that we have taken the position that a single illness or a few illnesses with specific correct treatments account for veterans’ complaints. We repeatedly stated in our draft report that veterans are experiencing a wide array of symptoms and disabling conditions. 5. DOD’s conclusion that research on treatments should await the results of epidemiological studies belies the fact that several illnesses suffered by these veterans have already been identified but that imperfect treatment Page 74 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense exists for these illnesses. Additionally, DOD and VA were directed to conduct research on treatments for ailing Gulf War veterans in the national defense authorization act for fiscal year 1995. Our report does not recommend that any ongoing research be discontinued; rather, it points out that as a result of the misplaced focus and formidable methodological problems, much of the ongoing epidemiological research will not be able to provide precise, accurate, and conclusive answers regarding the potential causes of the Gulf War veterans’ illnesses. Moreover, given that the majority of federal research already covers epidemiological issues, we recommend that DOD give greater priority to research on treatment for ill veterans and on low-level exposures to chemicals and their interactive effects and less priority to further epidemiological studies. 6. IOM also commented that any additional nationwide epidemiologic studies of Gulf War veterans are likely to be of limited scientific value at this time. At this stage, greater emphasis is warranted on studies that explore plausible disease hypotheses rather than large-scale population-based studies of prevalence. While the large-scale federal studies cited by DOD have yielded descriptive information on the health profile of Gulf War veterans, they have shed less light on why Gulf War veterans report more health complaints than nondeployed veterans. 7. Regarding research on low-level exposures to various chemical agents, DOD refers to an allocation of slightly more than $15 million for this purpose and describes the process that would be followed to obligate these funds to specific research projects. However, its comments on our recommendation provide no detail on its progress in distributing these funds. In its final report, PAC noted that, “DOD’s intransigence in refusing to fund [research on possible long-term health consequences of low-level exposure to chemical warfare agents] until summer 1996 has done veterans and the public a disservice.” 8. DOD partially concurs with our recommendation that it refine current approaches of the clinical and research programs for diagnosis of PTSD, consistent with recent IOM suggestions. IOM recently found that, “In view of potential exposure to low levels of nerve agents, certain refinements in the CCEP would increase its value.” IOM recommended improved documentation of the screening used during Phase I for patients with psychological conditions such as depression and PTSD, noting that “if there are long-term health effects of nerve agent exposure, it is possible that these effects could be manifested as changes in mood or behavior.” IOM has made other specific recommendations that are consistent with our Page 75 GAO/NSIAD-97-163 Gulf War Illnesses Appendix V Comments From the Department of Defense findings, including the recommendation that physicians take more complete patient histories regarding the onset of health problems and occupational and environmental exposures to rule out alternative explanations for neuropsychological findings.53 In its comments, DOD refers to diagnostic procedures used in Phase II of the CCEP examination, but these cover a small proportion of participants. 53 See Institute of Medicine, Adequacy of Comprehensive Clinical Evaluation Program: Nerve Agents (Washington, D.C.: National Academy Press, 1997), pp. 16-17. Page 76 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Note: GAO comments supplementing those in the report text appear at the end of this appendix. See comment 1. Page 77 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 78 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 79 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 2. Page 80 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 81 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 3. Page 82 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 83 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 84 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 4. See comment 5. Page 85 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 6. Page 86 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 7. Page 87 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 8. See comment 9. See comment 10. Page 88 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 11. Page 89 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 90 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 12. See comment 13. Page 91 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs See comment 14. See comment 15. Page 92 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 93 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 94 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs Page 95 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs The following is our response to the Department of Veterans Affairs’ letter dated June 17, 1997. 1. We have changed the word “illness” in our report title to “illnesses.” GAO Comments 2. VA acknowledges that clinical progress cannot be measured with existing or new databases. VA also notes that “appropriateness and effectiveness of treatment can only be determined for a specific medical condition whose pathogenesis and natural history has been well characterized.” VA agrees that longitudinal tracking of veterans with specific diagnoses could be of value. The majority of veterans have one or more diagnoses, which, in combination with their chief health complaints, should provide the basis for evaluating their care. Nevertheless, VA emphasizes the difficulty of evaluating the clinical progress of individuals with undiagnosed conditions. We are not suggesting randomized clinical trials of new treatments, as VA appears to imply, but do suggest that the Department develop a plan to monitor the condition of undiagnosed individuals in order to promote effective symptomatic treatment. Although VA asserts that those veterans who receive no diagnosis for their illnesses are treated appropriately for their symptoms, they do not indicate that they have any means of ensuring this, and they provide no evidence for the assertion. As VA suggests, augmenting its collection of data on the progress of ill Gulf War veterans with additional comparative data would provide valuable additional information. However, at a minimum, it seems desirable to collect descriptive information on how veterans’ conditions have improved or worsened. 3. VA agrees that research on low-level exposures to chemicals should be given higher priority but does not believe epidemiological studies should be given lower priority. Since VA does not provide evidence to dispute our findings that ongoing epidemiological studies will not provide accurate, precise, or conclusive answers, we continue to believe that emphasis in the research should be shifted. 4. VA does not concur with our recommendation on the basis that it is already making efforts to refine current approaches of its clinical and diagnostic programs for diagnosing posttraumatic stress disorder (PTSD). The emphasis of our recommendation is not upon how PTSD is diagnosed at specialty centers but upon how it is diagnosed in the course of ordinary registry evaluation and the improved validation of diagnostic methods Page 96 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs used in establishing its prevalence. VA’s comments offer no corrective plan. Therefore, we continue to believe that VA should refine the current approaches of the clinical and research programs for diagnosing PTSD. 5. See comment 1. 6. VA’s comments do not address our specific finding that it has no information on whether Gulf War veterans are any better or worse today than when they were initially diagnosed. VA suggests that its current approach provides adequate oversight for Gulf War veterans’ care. We found that VA relies on quality assurance mechanisms that do not ensure a given level of effectiveness for the care provided. VA agrees with us that the 1996 National Customer Satisfaction Survey was not adequate, and it plans to correct those deficiencies. However, VA has not provided any evidence to us to the contrary. Given the fact that VA has no way to track changes in veterans’ health status, we continue to believe that DOD and VA should develop and implement plans to monitor the clinical progress of veterans. 7. While VA agrees that it is indeed possible that a Gulf War-related exposure to agents may never be precisely linked to Gulf War veterans’ illnesses (regardless of how well a study may be designed or what type of research is conducted), it believes that epidemiologic research can provide important information about the health consequences of Gulf War service. We agree that descriptive studies cited by VA are useful in understanding group differences, but it is not clear what hypotheses these studies have generated regarding risk factors. Our conclusion remains valid regarding the inability of ongoing epidemiological research to provide precise, accurate, and conclusive answers regarding the causes of veterans’ illnesses because of formidable methodological problems. 8. Comments from the Presidential Advisory Committee and our responses are in appendix VII. Our methodology is described on pages 13 and 14 of our report. We use an extensive quality assurance process for all of our products, as we did for this report. The expertise of the team who conducted this review is discussed on page 14. 9. VA notes that our report is thinly supported with few references (approximately 35). However, as we note in our report, we reviewed not only the articles that PAC cited in support of its conclusions (which we do not list) but also articles published in peer-reviewed journals that PAC did not take into consideration (which we do list). Page 97 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs 10. Our report does not imply the assertion VA is making. In our evaluation of federal research strategy, we are reporting our findings on the extent to which the strategy is coherent. 11. Our conclusion is based on PGVCB-provided data, which show that the vast majority of federal research was initiated during or after 1994 and relatively few studies have been completed. 12. This is an inaccurate presentation of the statement in our report. We stated in our report that PAC did not provide evidence in support of its assertion that stress is an important contributing factor to the “broad range” of illnesses currently being reported by Gulf War veterans. 13. As our report notes, we reviewed the literature cited by PAC. All but two of these references from peer-reviewed journal articles deal with the putative association with PTSD; only two discuss the role of general life stress. The scientific articles on PTSD do not present convincing evidence that PTSD is common in Gulf War veterans or that it explains the symptoms reported by Gulf War veterans. All but one of the peer-reviewed studies on PTSD in Gulf War veterans relied on psychometric PTSD scales, unaccompanied by psychiatric interviews, and only minimal elevations of scores were found. These do not indicate the presence of PTSD. Virtually any illness that causes primary or secondary emotional concern can produce minimal elevations of scores on the psychometric PTSD scales. The fact that minimal elevations of psychometric scales scores were slightly higher than those of nondeployed veterans proves only that deployed veterans have more illness of some kind, but it does not establish that it is related to PTSD or general life stress. 14. VA failed to understand the central message underlying the two research studies it cited (Jamal et al., 1996, and Haley et al., 1997). These two studies demonstrate that the syndrome of chronic fatigue, cognitive problems, balance disturbances, joint aches, diarrhea, etc., could be neurological injuries from exposure to chemicals in the war. These studies also suggest that routine medical examinations are incapable of detecting chronic neurotoxicity. Thus, the statement that “the majority of VA Registry participants have conventional medical diagnoses and are being treated with appropriate therapies” is undoubtedly sincerely meant and true, but irrelevant. 15. We have stated in our report that while the government found no evidence that biological weapons were deployed during the Gulf War, the Page 98 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VI Comments From the Department of Veterans Affairs United States lacked the capability to promptly detect biological agents, and the effect of one agent, aflatoxin, would not be observed for many years. Page 99 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses Note: PAC’s comments on our draft report, along with our responses to these comments, are reproduced here. Due to the length and highly technical nature of PAC’s comments, our response follows each individual comment in the letter, rather than at the end of the comments. GAO RESPONSE The purpose of our evaluation of PAC’s conclusions with respect to risk factors was to ascertain the amount of knowledge about Gulf War illnesses generated by research 6 years after the Gulf War, and evaluate the evidence supporting conclusions on these issues. We reviewed these conclusions because they are the strongest statements of any official body that we have found on these matters. Moreover, the PAC review panel included a number of recognized experts in scientific questions at issue who were assisted by a large staff of scientists and attorneys. In addition, Page 100 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses PAC extensively reviewed the research on Gulf War veterans’ illnesses. Thus, evaluating the strength of the PAC’s conclusions provides important evidence about how fruitful research on Gulf War illnesses has been to date. We have repeated in our letter the statement on this point that we had made in appendix IV. Our report cites PAC’s recommendations and significant conclusions. We also carefully reviewed the PAC’s interim report, which cites potential problems for federally funded research. We documented that such problems affect large portions of the federally sponsored studies (see app. III). GAO RESPONSE Our study is a systematic evaluation of the matters that Congress directed us to examine. We reviewed the scientific literature and published as well as unpublished work of internal and external bodies. In reviewing conclusions, we examined the support cited as well as additional evidence we gathered and compared these with the official conclusions. Page 101 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses We have added citations to better reflect our use of scientific literature in our review. GAO RESPONSE As we pointed out in our draft report, the number of studies we cite was taken directly from the most recent (April 1997) annual report to Congress by the official sponsoring and coordinating entity for pertinent research, the Persian Gulf Veterans’ Coordinating Board (PGVCB). PGVCB, which coordinates research on Gulf War veterans’ illnesses involving VA, DOD, and HHS, is required under Public Law 102-585 to report annually on the results and progress of pertinent research activities undertaken or funded by the executive branch. GAO RESPONSE We do not confuse PTSD and stress-related effects. Indeed, it is unclear how we could both “juxtapose” the distinct issues and simultaneously “treat them as a single matter,” as the comment alternately suggests. We address Page 102 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses these items in tandem in order to prevent the very type of misinterpretation about which the Committee is concerned. In contrast, PAC’s report blurs the distinction between PTSD (that is, a specific syndrome caused by emotional trauma) and stress (a potential risk factor). To support its conclusion regarding the contribution of stress to veterans’ illnesses, PAC cites 18 reports from peer-reviewed journals, but these largely assess the association between stress and PTSD. Only two peer-reviewed articles were presented in support of the broader effects of stress and neither included measurements of Gulf veterans. Some studies, while intending to assess the extent of PTSD, found little and instead discussed “stress symptomatology,” “trauma-related symptoms,” or “PTSD symptoms,” using these terms to refer to measurements on a PTSD scale that did not meet the validated cutoff for indication of PTSD. GAO RESPONSE As our report notes, we reviewed the literature PAC cited in support of its conclusion. Only two references from peer-reviewed journals were provided to substantiate the role of general life stress in the etiology of veterans’ symptoms.54 Neither reference presented measurements of Gulf War veterans. We noted in our report that this quotation is taken from, “R.H. Stretch et al., “Physical Health Symptomatology of Gulf War-era Service Personnel From the States of Pennsylvania and Hawaii,” Military Medicine, vol. 160 (1995), pp. 131-36. (See app. II.) 54 K. C. Hyams & F. S. Wignall, “War Syndromes and Their Evaluation: From the U.S. Civil War to the Persian Gulf War,” Annals of Internal Medicine, vol. 125 (1996), pp. 398-405 and G. P. Chrousos & P. W. Gold, “The Concepts of Stress and Stress System Disorders,” Journal of the American Medical Association, vol. 267 (1992), pp. 1244-1252. Page 103 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE Stress can be associated with a wide range of physical illnesses, and we do not suggest that illnesses that are stress-induced are any less real. However, we did not find evidence that the broad range of Gulf War veterans’ physical symptoms were induced by stress. (See our response to the next comment.) GAO RESPONSE We reviewed the reports from the Fort Devens and the New Orleans studies on PTSD, but these two studies provide little support for Committee’s conclusion that “stress is likely to be an important contributing factor to the broad range of illnesses currently being reported by Gulf War veterans.” First, the primary focus of each of these studies is Page 104 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses the measurement of PTSD. Second, the symptom measures employed in both studies focused selectively on psychological, psychosomatic, and/or stress-related conditions. For example, the New Orleans study employed a checklist inquiring about 20 conditions—11 taken from a scale of psychosomatic complaints and 9 other symptoms that are commonly observed to be stress-related.55 The samples of deployed troops who indicated high war-zone stress checked more items on this symptom list than those classed as having low or no war zone stress, but 4 to 10 months after the war, only three complaints (nervousness, concentration difficulties, and needing medications to sleep or calm down) showed statistically significant differences based on war-zone stress. Third, no physical examination was conducted in these studies, so it is impossible to determine whether the measured symptoms were selectively related to war zone stress. Fourth, in both studies, there is some possibility that the relationships between war zone stress and symptoms are byproducts of similarities in the methods used to measure them; the Fort Devens study acknowledges some research showing that self-reports of stress are vulnerable to bias from a host of event-related and personal characteristics. Finally, in the Fort Devens study, the total amount of variation in reported symptoms that was explained by the combination of combat exposure stress and a variety of other factors was quite modest (13 percent).56 The Committee’s remark concerning diagnostic methods suggests that the inclusion of control groups overcomes bias from faulty methods for measuring PTSD. First, the method recognized by experts in the field of PTSD research for conclusively making the diagnosis of PTSD is a psychiatrist’s or psychologist’s clinical interview following a structured interview protocol, such as the CAPS or SCID. (See app. V, DOD’s response to GAO’s report.) Of the 18 peer-reviewed studies of Gulf War veterans cited by the PAC report in support of its conclusions on stress, only one used this method.57 All others relied on psychometric scales. Second, if the deployed veterans suffered subtle neurological damage, for example from chronic pesticide exposures, their scores on the psychometric PTSD scales could be falsely elevated, while those of the nondeployed controls, not exposed to pesticides, would not be. The use of a control group would not correct for this type of bias. 55 P. B. Sutker, et al., “War-Zone Trauma and Stress-Related Symptoms in Operation Desert Shield/Storm (ODS) Returnees,” Journal of Social Issues, vol. 49 (1993), pp. 33-49. 56 J. Wolfe, et al., “Reassessing War Stress: Exposure and the Persian Gulf War,” Journal of Social Issues, vol. 49 (4) (1993), pp. 15-31. 57 S. Perconte, A. Wilson et al., “Unit-Based Intervention for Gulf War Soldiers Surviving a SCUD Missile Attack: Program Description and Preliminary Findings,” vol. 6 (1993), pp. 225-238. Page 105 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We refer to the use of major diagnostic categories from the International Classifications of Diseases - 9th Revision to report on various types of conditions as a group. For example, the category “psychological conditions” is used to report data from DOD’s clinical program. The diagnoses included in the ICD-9 series for psychological conditions cover everything from relatively common, transient, and easily treated conditions, such as tension headache, to more intractable disorders, like clinical depression. It is not clear what clinical or scientific purpose is served by discussing these varied diagnoses as a group. In the PAC report, under the heading “Data on Stress-Related Disorders” (see p. 71), the Committee notes that “psychological conditions are either the primary or secondary diagnosis in 36.0 percent of CCEP participants,” and that “the most common conditions are: major depressive disorder, neurotic depression (also called dysthymia), depression (not otherwise specified), PTSD, anxiety disorders, adjustment disorders, alcohol-related disorders, and substance-related disorders.” However, as noted in the footnote to the table on page 72 of the PAC report, the single most common condition in this category is actually tension headache (11.3 percent of CCEP participants and 2.3 percent of registry participants). Apart from tension headache, none of the individual diagnoses listed in this category is the primary diagnosis for more than 3 percent of CCEP registrants. Page 106 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We quoted from page 34 of PAC’s Final Report, “Currently, stress is the risk factor funded for the greatest fraction of total studies—32 studies (30 percent).” However, we have now substituted the figure provided by PAC. GAO RESPONSE As PAC notes, we did not conclude that stress was incompatible or incapable of producing physical symptoms; we concur with PAC’s assessment of this matter. However, we do not find that PAC has cited evidence that stress is likely to be an important contributing factor to the Page 107 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses broad range of illnesses that veterans report. We have revised the statement in table IV.1 to clarify this point. Regarding the health effects of low-level exposure to chemical warfare agents, the Committee suggested in 1996 that “the government should plan for further research on possible long-term health effects of low-level exposure to organophosphorus nerve agents, such as sarin, soman, or various pesticides, based on studies of groups with well-characterized exposures, including (a) cases of U.S. workers exposed to organophosphorus pesticides and (b) civilians exposed to the chemical warfare agent sarin during the 1994 and 1995 terrorist attacks in Japan. Additional work should include follow-up and evaluation of an appropriate subset of any U.S. service personnel who are presumed to be exposed during the Gulf War. The government should begin by consulting with appropriate experts, both governmental and nongovernmental, on organophosphorus nerve agent effects. Studies of human populations with well-characterized exposures will be much more revealing than studies based on animal models, which should be given lower priority.” (PAC, Final Report, p. 54) Accidental and occupational exposures like those cited by PAC are rarely “well characterized,” and due to the potentially toxic nature of the exposures, animal studies will be more important to characterizing the effects, particularly synergistic ones. Although PAC concluded that “ongoing federally funded studies should help the scientific community draw conclusions about the synergistic effects of PB and other risk factors” (Final Report, p. 117), we could find no PAC recommendation for additional research on the synergistic health effects of pyridostigmine bromide and other risk factors. Page 108 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE All but one of the 18 studies on PTSD in Gulf War veterans cited by the PAC report based diagnoses of PTSD on psychometric PTSD scales without confirmatory psychiatric interviews. These instruments are validated for screening, not diagnosis. In addition, care must be taken in evaluating elevated scores that do not surpass validated cut-points for discrimination of PTSD and non-PTSD populations.58 GAO RESPONSE The PAC states that its review of studies [on PTSD] was based on those with high participation rates. However, most of the cited studies presented PTSD survey data based on samples that were not statistically generalizable. Among those that did employ generalizable samples, participation rates varied from 25 percent to 58 percent, but no comparison of participants and nonparticipants was presented to assess the likelihood of selection bias. We respond to the Committee’s second and third points elsewhere in this appendix. 58 See, for example, the discussion by T. M. Keane et al., “Mississippi Scale for Combat-Related Posttraumatic Stress Disorder: Three Studies in Reliability and Validity,” Journal of Consulting and Clinical Psychology, vol. 56(1) (1988), pp. 85-90. Page 109 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE Control groups are not a substitute for accurate diagnostic methods. For example, without accurate diagnosis, it is possible that neurological symptoms related to war-related exposures apart from stress will be misattributed to PTSD. In addition, some studies have employed modified PTSD scales incorporating questions that may become markers for recent war participation, rather than evidence of PTSD. These questions would selectively increase scores in the Gulf War group. GAO RESPONSE We quote from testimony provided to PAC on March 26, 1996, by Dr. Peter Spencer, who is the principal investigator of a large, federally funded study. As known since 1995 and acknowledged in the PAC report (p. 118), the incubation period for classical visceral leishmaniasis (usually caused by L. donovani) may exceed 2.5 years.59 In addition, the natural history of a newly recognized form of the illness (viscerotropic leishmaniasis) is 59 W. H. Jopling, “Long Incubation Period in Kala-azar,” British Medical Journal, vol. 2:1013 (1955). Page 110 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses unknown.60 Those whose immune systems become weakened for any reason will be at particular risk. In such patients, the development of visceral leishmaniasis (involving malaise, lassitude, weight loss, splenomegaly, and anemia) up to 20 years after exposure has been documented.61 GAO RESPONSE Insofar as no screening or simple diagnostic test is currently available for newly recognized forms of leishmaniasis, there is an insufficient basis to assess the success of the clinical examinations in detecting it. However, this presumption is the primary basis on which the Committee dismisses the notion that leishmania infection is much of a continuing problem. In a March 13, 1997, briefing, experts from Walter Reed Army Institute of Research told us that “most clinicians will fail to recognize ’classic’ forms of leishmaniasis, much less atypical clinical presentations.” As we note in the report, a CDC analysis appears to concur that the signs of a newly recognized form of the disease are nonspecific and that the diagnosed cases were identified by aggressive case-finding.62 It stands to reason that diagnosis would be difficult insofar as leishmaniasis is generally unknown in the United States. While PAC concludes that viscerotropic leishmaniasis is not considered to be a cause of widespread illness among Gulf War veterans, PAC acknowledges on p. 118 of its Final Report that “viscerotropic leishmaniasis can be difficult to confirm.” 60 A. J. Magill et al., “Viscerotropic Leishmaniasis in Persons Returning from Operation Desert Storm — 1990-1991,” Journal of the American Medical Association, vol. 267(11) pp. 1444-46. 61 Badaro, Falcoff et al., “Treatment of Visceral Leishmaniasis With Pentavalent Antimony and Interferon Gamma,” New England Journal of Medicine, vol. 332 (1990), pp. 16-21. 62 “Viscerotropic Leishmaniasis in Persons Returning from Operation Desert Storm—1990-1991” [CDC Editorial Note], Journal of the American Medical Association, vol. 267 (11) (1992), pp. 1444-6. Page 111 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We have modified the statements in our letter summarizing our findings to match the statement in appendix IV, which incorporates the role of the weakened immune system. It is in cases in which the patient’s immune system becomes deficient that such reexpression of previously asymptomatic infection is a concern.63 However, because it is not possible to predict which persons’ immune systems may become weakened, we believe that it is important for all veterans and health care professionals to understand the significance of such potential infection. In addition, the natural history of the viscerotropic form of leishmaniasis is not well understood; that is, little is known about the length of incubation and the course of disease.64 While it could be consistent with some of the Gulf War veterans’ symptoms, we do not contend in our report that leishmaniasis—or any other illness of which we are aware—would explain the range of symptoms currently being reported in the veterans. 63 See, for example, A. J., Magill, et al., “Visceral Infection Due to Leishmania tropica in a Veteran of Operation Desert Storm Who Presented 2 Years After Leaving Saudi Arabia, Clinical Infectious Diseases, vol. 19 (Oct. 19, 1994), pp. 805-6. These authors note, “...the presence of a cofactor depressing cell-mediated immunity (malnutrition, immunosuppressive drug treatments, AIDS, or malignancy) can lead to symptomatic leishmanial disease....” See also , R. Badaro, et al. “Leishmania donovani: An Opportunistic Microbe Associated With Progressive Disease in Three Immunocompromised Patients,” Lancet, vol. 1 (1986), pp. 647-9. 64 Even infection with the same species of parasite (Leishmania donovani) can take widely different courses (see Badaro et al., “New Perspectives on a Subclinical Form of Visceral Leishmaniasis,” The Journal of Infectious Diseases, vol. 154(6), pp. 1003-1011.). Page 112 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses We noted the reemergence of leishmaniasis in Europe in the context of noting that the infection can flare when the immune system is weakened; the comparability of the exposed groups is not relevant to the point that we were making.65 GAO RESPONSE We presume that veterans are concerned about their future health as well as their current health. In the absence of simple diagnostic tests, it is difficult to judge the extent of current illness attributable to leishmania infection. The more important point is whether these veterans are ill as a result of their exposure. The risk of sandfly fever to U.S. troops in the Gulf was believed to be high. Although we recognize that it is possible that some of these veterans may have been deployed to other areas in which they might have contracted this disease, the blood samples that CDC analyzed were taken after their return to the United States from the Gulf. 65 See, for example, Phillip G. Lawyer, “Leishmaniasis Update,” Proceedings of the 1995 DOD Pest Management Workshop (1995), p. 3 (http://www.afpmb.acq.osd.milpubs/present/htm). He states, “The emergence of VL [viscerotropic leishmaniasis] as a serious opportunistic infection in AIDS patients in Europe has alarming implication for leishmaniasis endemic areas where the prevalence of HIV infection is increasing (Africa, Brazil, Indian subcontinent).” Page 113 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE Medical surveillance during the war was imperfect. While some reports indicate no cases of sandfly fever, at least six cases of febrile illness compatible with sandfly fever were reported among soldiers of the 1/505 PIR on September 22, 1990, by a preventive medicine officer. In addition, the risk of sandfly fever was believed to be high. A December 1991 Defense Intelligence Agency report presented tests of blood samples from Iraqi military personnel involved in the Gulf War. These tests were conducted to help identify biological warfare agents in the Iraqi inventory and assess the prevalence of endemic diseases. In discussing naturally occurring diseases, the report notes, “The large percentage of positive reactions to sandfly fever (Sicilian and Naples strain) confirms the high risk this disease poses for US military operations in the region.” For the Sicilian strain, 98 of 125 samples were positive, and 49 of 126 samples were positive for the Naples strain. (In contrast, only 21 of 130 samples were positive for exposure to Q-fever.) Thus, if there were no cases of sandfly fever, it seems difficult to explain their complete absence. It is true that the presence of evidence of exposure to sandfly fever did not distinguish persons with the cluster of fatigue symptoms defined by CDC from persons who did not fit this definition. However, this does not obviate the need to exclude such infection in diagnosing particular veterans’ fatigue and posttraumatic symptoms. Sandfly fever would not consistently result in such complications, though it might sometimes be responsible for them. Page 114 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE First, references have been provided in footnotes 28-36 of appendix III. Second, our specific responses to the claim that we have mischaracterized the Committee’s conclusions are set forth below. Third, we have clarified statements that may have led the Committee to infer that we claimed that all organophosphates compounds produce similar long-term effects. Fourth, we are careful to distinguish between those individuals who are ill today and individuals who may become ill in the future. For example, our discussion of aflatoxin is largely about potential cancers in the future. Fifth, as directed by Congress, we conducted our own independent, objective review. Our review included reviewing reports and scientific literature, interviewing researchers, and analyzing the information Page 115 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses available to us. Through this review we reached different conclusions from those of the PAC. In any event, given PAC’s finding that minimal research is available on the health effects of low-level exposure, it is difficult to understand the rationale for its conclusion that chemical warfare agent exposures are unlikely to be consistent with veterans’ health complaints. Moreover, we note that findings of some studies on such low-level exposures are not supportive of such a conclusion. We respond to PAC’s remarks concerning its recommendations elsewhere in this appendix. Insofar as the Committee clearly feels strongly about the need for additional research, we find it difficult to understand the rationale behind PAC’s conclusion that it is unlikely these exposures could have contributed to veterans’ health complaints. GAO RESPONSE PAC’s(first) comment incorrectly misinterprets our point. As noted, PAC formed some of its conclusions in the absence of exposure data. However, we have removed the quotation. Some of the Committee’s conclusions are inconsistent with the results of applying its analytic framework. For example, it is difficult to understand why the Committee concludes that the agent in question is “unlikely” to have contributed to the health problems reported by veterans even as it recognizes the need for data on the health effects of low-level exposure. Page 116 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE As we noted in an earlier report, “available bomb damage assessments during the war concluded that 16 of 21 sites categorized by Gulf War planners as nuclear, biological, and chemical (NBC) facilities had been successfully destroyed. However, information compiled by the United Nations Special Commission (UNSCOM) since the end of Desert Storm reveals that the number of suspected NBC targets identified by U.S. planners, both prior to and during the campaign, did not fully encompass all the possible NBC targets in Iraq.” UNSCOM has conducted investigations at a large number of facilities suspected by the U.S. authorities as being NBC related. Regarding the few suspected weapons sites that have not yet been inspected by UNSCOM, we have been able to determine that each was attacked by coalition aircraft during Desert Storm and that one site is located within the Kuwait theater of operations in closer proximity to the border, where coalition ground forces were located. However, we have yet to learn why these facilities have not been investigated.66 66 See Operation Desert Storm: Evaluation of the Air Campaign (GAO/NSIAD-97-134, June 12, 1997, p. 2). Page 117 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE PAC has misinterpreted our position on the proper sequencing of studies. Research into the nature of the health effects of agents to which troops may have been exposed during Operation Desert Storm should not wait for accurate answers to questions of the magnitude of actual exposures. We neither state nor imply otherwise. In its reference to a statement we made in table IV.1, we made that statement to provide background for our assessment. We have deleted the word “given,” which may have left the incorrect impression that the Committee did not take account of the presence of chemical warfare agents at Khamisiyah and elsewhere on the battlefield. Page 118 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We cite animal studies showing that exposures to certain organophosphate agents at levels that do not cause acute poisoning are associated with measurable long-term effects. It appears inconsistent for the Committee both to conclude that exposures to organophosphate agents are unlikely to have contributed to veterans’ health problems and simultaneously to recognize the existence of minimal research on low-level exposure—the most likely exposure scenario for organophosphate pesticides. The Committee may have overlooked a set of articles published in peer-reviewed journals by Husain et al. addressing the chronic neurotoxicity of low-level exposure to sarin. In these studies, the investigators exposed hens and mice, in separate experiments, daily for 10 days to sub-acute doses of sarin orally and through inhalation. The animals did not require protection by simultaneous administration of atropine and pralidoxime, often used in high-dose experiments. Fourteen days after the start of the daily exposures, some of the animals developed effects (for example, ataxia, muscular weakness), suggesting that sarin can induce OPIDN. These studies have been discussed since they were published in 1993, 1994, and 1995, and they have received no serious criticism of which we are aware. It appears that DOD, PGVCB, and PAC have not recognized and commented on them, while continuing to insist that there is no evidence that low-level sarin can cause chronic neurotoxicity in the absence of severe immediate effects. Page 119 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE The comment by PAC reflects a selective reading of the work by Duffy and Burchfiel in the area of the EEG effects of organophosphates. Indeed, the authors characterized one of the tests that they conducted on the EEGs of treatment and control groups as “inconclusive.”67 However, they also reported statistically significant differences between the two groups on several other measures, such as the increase in the amount of beta activity in EEGs and an increase in rapid eye movement sleep. The authors concluded that, “Our EEG findings and the psychological reports in the literature provide parallel warnings of possible long-term CNS toxicity of OP agents.”68 Additionally, in recent testimony before the House Committee on Government Reform and Oversight, Professor Duffy made the following observations, which support our conclusions about the effects on the behavior of organophosphates (including sarin): “It is quite possible to have a biologically significant exposure to OP compounds and not be aware of it...Sarin can produce long term alteration of brain function. Levels of exposure capable of producing such late effects may not be recognizable by subjects, especially if they are unaware of what is happening and/or are distracted by other activities.”69 The Armed Forces Epidemiological Board also reviewed these studies and found that, “they represent reasonable evidence that even small doses (exact level is unknown) may result in EEG changes.”70 67 James L. Burchfiel, et al., “Organophosphate Neurotoxicity: Chronic Effects of Sarin on the Electroencephalogram of Monkey and Man,” Neurobehavioral Toxicology and Teratology, vol. 4 (1982), pp. 767-778. 68 Ibid., p. 777. 69 Frank H. Duffy, M.D. (Department of Neurology, Harvard Medical School), “Evidence that Minor Exposures to the Nerve Agent Sarin May Lead to Long Term Difference in Brain Function,” testimony provided to the House Committee on Government Reform and Oversight, Jan. 19, 1997. 70 Environment Committee, Armed Forces Epidemiological Board, Long-term Health Effects Associated with Sub-clinical Exposures to GB and Mustard, p. 6. Page 120 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We have cited the work of Dr. Haley and his colleagues as a positive example of an attempt to refine a case definition in the presence of diffuse and nonspecific symptoms. His approach, to look for patterns of correlation among the reported symptoms and explore their relationship with exposure history, is a reasonable and rational first step upon which others might build. We discussed Dr. Haley’s approach with two leading epidemiologists, who agreed that the approach Dr. Haley had taken was reasonable in an instance in which a case definition was difficult to derive. In fact, elsewhere in the aforementioned editorial, Dr. Landrigan concurs with the major thrust of our position: “Haley et al. suggest that some cases of illness in members of their population may represent chronic neurotoxicity caused by low-dose exposures to chemical warfare agents. This is an important question that demands serious investigation... Further research is Page 121 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses needed to determine whether low-dose exposure to chemical warfare agents can cause chronicneurotoxicity.71 We agree with Dr. Landrigan on this point. It is also our contention, based on the evidence presented in our report, that the federal research program has not pursued this question with sufficient energy. It is apparent to us, based on the literature that we reviewed and the references cited, that the hypothesis that some veterans’ illnesses are OPIDN or similar to OPIDN that may stem from exposure from pesticides, chemical warfare agents, or pyridostigmine bromide while on duty in the Persian Gulf is a plausible hypothesis. We disagree with PAC’s conclusions that these are unlikely exposure scenarios for the illnesses being experienced by veterans. Moreover, we fail to understand the Committee’s rationale for endorsing additional studies in this area after discounting the likelihood of the hypothesis. In fact, it should be noted that CDC took similar steps to construct a case definition in its review of symptoms reported by a large group of Gulf War veterans. Dr. Haley’s work has apparently generated plausible hypotheses for further exploration and testing. 71 P. J. Landrigan, Illness in Gulf War Veterans. Journal of the American Medical Association, vol. 277 (1997), pp. 259-261. Page 122 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE Concerning exposure to aflatoxin, for the reasons cited earlier, we can neither confirm nor rule out veterans’ exposure to this agent. The Central Intelligence Agency has noted that the health effects of exposure to aerosolized aflatoxin are poorly understood. Descriptive studies of clearly defined endpoints may be useful in providing assurance that large numbers of veterans are not suffering the health problems characteristic of aflatoxin exposure. However, it is important to note that only in the instance of widespread exposure would this approach resolve the issue of whether particular veterans’ health problems are attributable to their Gulf War service. In response to the general comment concerning the value of epidemiologic studies of Gulf War veterans, we agree that some basic descriptive information on veterans’ health may be useful, to include the cancer surveillance studies identified by PAC, for the purpose of providing veterans with information about the presence of widespread and serious health effects. However, it will be very difficult for these studies to resolve the issue of whether specific veterans’ health problems are related to their Gulf War service in the absence of (a) widespread exposure; (b) biomarkers for exposure; or (c) better data on who was exposed and at what levels. Page 123 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We have added text to note the Committee’s findings. The PAC concluded, “Aflatoxin...is a liver carcinogen, and increased rates of liver cancer could result decades following low-level exposure, although available evidence reviewed by the Committee does not indicate such exposures occurred during the Gulf War.” (PAC, Final Report, p. 112.) GAO considers exposure to aflatoxin as an unresolved issue. GAO RESPONSE We have added text to clarify our position regarding the likely utility of further epidemiologic research in light of the absence of adequate exposure data. Page 124 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We used the methodological categorizations identified by PGVCB as reported for each study in PGVCB research plans and reports to Congress. The categories identified by PAC appear to be an amalgam of methodological approaches and topical emphases. GAO RESPONSE Appendix IV contains a more detailed discussion of the PAC’s conclusions and our assessments. GAO RESPONSE Our specific rebuttals to PAC’s individual assertions on these matters are set forth below. Page 125 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE Concerning delayed neurotoxic effects of organophosphates, not all organophosphates cause these effects; however the Committee is incorrect in implying that pesticides to which U.S. service people were exposed in the war could not have caused delayed neurotoxicity.72 As PAC reports on p. 97 of its Final Report, Chlorpyrifos (Dursban) was shipped to the Gulf. Dursban has been linked to delayed, chronic neurotoxicity in laboratory animals. In the past, the delayed, chronic neurotoxic potential of chlorpyrifos was overlooked.73 The Environmental Protection Agency has recently penalized the manufacturer for failing to promptly report human injuries from this pesticide and suggested that the pesticide be relabeled to withdraw it from many applications, a decision with which the company acquiesced. In addition, other unknown pesticide chemicals may have been brought from the United States or purchased from local suppliers in Saudi Arabia by troops outside the command structure; these cannot be enumerated by DOD. Regarding pyridostigmine bromide pretreatment, while pretreatment with pyridostigmine bromide does not potentiate chronic neurotoxicity from subsequent organophosphate exposure, studies of similar drugs indicate that treatment after a sufficient organophospates exposure may potentiate 72 M. Lotti, “The Pathogenesis of Organophosphate Polyneuropathy,” Critical Reviews in Toxicology, vol. 21 (1991) pp. 465-487 (esp. pp. 466-7, 472); J.G. Kaplan et al., “Sensory Neuropathy Associated With Dursban (Chlorpyrifos) Exposure,” Neurology, vol. 43 (1993) pp. 2193-96; C.S. Petty, “Organic Phosphate Insecticide Poisoning, American Journal of Medicine (Mar. 1958), pp. 467-70; E. Capodicasa, . et al., “Chlorpyrifos-induced Delayed Polyneuropathy “Archives of Toxicology,” vol. 65 (1991), pp. 150-155; J. Rosenberg, “Organophosphate Toxicity Associated with Flea-Dip Products — California,” Journal of the American Medical Association, vol. 260 (July 1, 1988), pp. 22-3. 73 M. Lotti, “The Pathogenesis of Organophosphate Polyneuropathy,” Critical Reviews in Toxicology, vol. 21 (1991), pp. 465-587 (especially pp. 467 and 473). Page 126 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses chronic neurotoxicity.74 There is evidence from laboratory studies that post-exposure treatment can cause chronic neurotoxicity to occur with organophosphate doses that would ordinarily be too low to cause a problem (pharmacologic “promotion”), or it could turn a mild case of organophosphate-induced chronic neurotoxicity into a severe case (pharmacologic “potentiation”). Several papers have appeared on this subject since 1990, although studies of post-exposure promotion by pyridostigmine per se have not been undertaken as far as we know. In response to the committee’s assertion that the only evidence for synergism among pyridostigmine bromide, permethrin, and DEET is from in vivo bioassays using lethal doses, Haley et al. found epidemiologic evidence that pyridostigmine toxicity and a chemical nerve agent may have acted synergistically to cause a syndrome they labelled “confusion-ataxia,” the most severe of the three primary syndromes they identify.75 They also found both pyridostigmine toxicity and DEET exposures to be strongly associated with their syndrome 3 (arthro-myo-neuropathy). In any event, standard risk analytic practice involves study of interactive effects in laboratory animals at doses that result in acute toxicity and to further characterize the relationship from that point. We have made the point that further study of the effects of this as well as other exposures is warranted. 74 Op cit., M. Lotti, p. 473; C. N. Pope & S. Padilla, “Potentiation of Organophosphorus-Induced Delayed Neurotoxicity by Phenylmethylsulfonyl Fluoride,” Journal of Toxicology and Environmental Health, vol. 31 (1990), pp. 261-73. 75 R.W. Haley and T.L. Kurt, “Self-reported Exposure to Neurotoxic Chemical Contamination in the Gulf War,” Journal of the American Medical Association, vol. 277 (1997) (3), pp. 231-237. Page 127 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We are uncertain what type of evidence PAC would consider sufficient to conclude that some effect might have occurred. While studies testing the toxic effects on humans have not been conducted, available animal studies provide reasonable evidence of negative effects that make it premature to conclude this is not a serious a risk factor and indicate that further research should be conducted. For obvious ethical reasons, it is not possible to conduct experimental studies on humans of effects that are feared to be toxic; thus, standard toxicological approaches used by the government and the private sector focus on research with animals. Researchers have shown that the neurotoxic phenomenon produced by organophosphate nerve agents in some poultry varieties was comparable to the manifestations produced in man.76 In this regard, it is known that organophosphate compounds that are neurotoxic to chickens will also produce neurotoxicity in humans under appropriate conditions. For example, the laboratory studies published by Abou-Donia et al. demonstrated that pyridostigmine, chlorpyrifos, permethrin, and DEET can synergistically act to cause delayed, chronic neurotoxicity. The hen is the EPA-required laboratory model for testing chemicals for the potential to cause OPIDN. Testing these chemicals for synergism in humans would have been highly unethical. The doses of permethrin, DEET, chlorpyrifos and pyridostigmine were intended to be in the range of sublethal human exposure. Given the severity of the OPIDN that occurred with chemical combinations in the doses used, it is possible that lower, but still medically significant, levels of damage would follow even with slightly lower doses of pyridostigmine. The Abou-Donia group administered pyridostigmine bromide orally; Gulf War veterans likewise were administered pyridostigmine orally. The permethrin, chlorpyrifos, and DEET were injected by needle into the skin just under the surface (intradermally) to simulate the probable absorption through the skin. Since under their feathers hens have thicker skin than humans, it is common laboratory practice to deliver skin exposure to hens by intradermal injection. We believe the Committee erred in not considering the findings of the Abou-Donia et al. studies to be indicative of expected effects on humans. 76 Stockholm International Institute for Peace Research (SIPRI), Delayed Toxic Effects of Chemical Warfare Agents (New York: Alonquist and Wiksell International, 1975). Page 128 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We are unaware of studies specifically testing, with sufficiently sensitive neurophysiologic techniques, for long-term neurotoxic side effects of pyridostigmine in any prior population taking pyridostigmine regularly. We are, however, aware that patients taking pyridostigmine are cautioned to avoid exposure to malathion, which was among the pesticides sent to the Gulf.77 Concerning the Committee’s assertion that research is in flux, pyridostigmine bromide may have been used in the presence of contraindicated coexposures (malathion) and may potentiate the effects of Dursban. In response to the Committee’s point about delayed neurotoxic effects, we have modified our statement to read, “delayed neurotoxic effects in humans exposed to PB and DEET have been epidemiologically inferred and reproduced in hens.” Human epidemiologic evidence of the synergistic effects of pyridostigmine, DEET, chlorpyrifos, and chemical nerve agents has been presented for an epidemiologic association between patterns of complaints and reported exposures of these agents in humans.78 77 “Pyridostigmine,” in Clinical Pharmacology (Online—http://www.cponline.gsm.com), Gold Standard Multimedia Inc., 1994. 78 Op cit., Haley et al. Page 129 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE We respond to PAC’s remarks concerning its recommendations elsewhere in this appendix. GAO RESPONSE Concerning the Committee’s observation that DOD was not the only reviewer of its report, we note that DOD publicly endorsed PAC’s findings. However, we have added a note that “PAC has asked us to point out that ’DOD was not the only party involved in [its] external review.....VA, HHS, veterans service organizations, and individual veterans and veterans advocates also reviewed [its] Interim and Final Report.’ PAC does not provide information on the extent to which these reviewers agreed with its findings or had their comments incorporated.” Again, our purpose was not to conduct a review of PAC’s activities but to identify and assess the strength of support for conclusions that had been drawn from the available research. Page 130 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses GAO RESPONSE Where the Committee has requested that specific changes be made to table IV.1, we have incorporated them (see previous comments). We have presented our detailed responses to the Presidential Advisory Committee in this appendix. We summarize the major points below: Page 131 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses Where the Committee has expressed concerns about citations and cross references to other studies, we have provided additional references and have modified the language in some instances to ensure that other readers will not misconstrue the meaning of our report. We double-checked the information that the Committee challenged and generally found that the data had been correctly stated. Moreover, a careful review of PAC comments indicates that they represent a selective presentation of data that tend to bias the reader’s perception of the issues. Therefore, we have not changed the overall thrust of our report. The Committee also takes issue with our reviews of its conclusions regarding psychological stress, leishmaniasis, and chemical warfare agents. Regarding stress, PAC states that we ignored its analytical framework, which was to presume that stress occurred and determine whether the types of symptoms and conditions reported by veterans were consistent with exposure to stress. However, the Committee did not provide evidence that stress is an important contributing factor to the “broad range” of illnesses currently being reported by Gulf War veterans. Although we agree that life stress can be associated with a wide array of physical symptoms, the research cited by the Committee largely assesses the relationship between stress and PTSD, which is not a common diagnosis among Gulf War veterans. Although the Committee notes that scores on PTSD screening questionnaires are higher among Gulf War veterans than among controls, problems are associated with interpreting scores on these scales below the validated cutoff points for PTSD risk. In addition, confirmatory psychiatric interviews to eliminate alternative explanations for elevated PTSD screening scores were not done in most of these studies. Thus, the possibility remains that Gulf War veterans show higher average scores on PTSD screening scales due to other conditions or nonstress-related exposures selectively associated with service. Some studies also modified PTSD screening instruments to make them more applicable to Gulf War veterans; an unintended consequence of this modification is the introduction of information bias (that is, adding questions that selectively affect the scores of Gulf War veterans). Finally, as we note in our report, at least one study that examined the relationship between stress and veterans’ physical symptoms in a large sample found none, and in a separate report, its authors noted that, “Although the stress that the Page 132 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses deployed veterans are experiencing can be characterized as substantial, it is being handled unremarkably.”79 Regarding our evaluation of the conclusion that the likelihood of leishmania infection has diminished as an explanation for widespread illness, the Committee asked, based on the low numbers of cases of leishmaniasis diagnosed in DOD and VA clinical programs, that we justify any continued concern about asymptomatic infection. While there is no fundamental disagreement regarding the available facts or the basic circumstances under which asymptomatic infection is a concern, the Committee’s justification for dismissal of leishmania as a risk factor rests heavily on two assumptions: (1) that diagnostic programs have been highly likely to detect the disease, even in the absence of any widely available screening or diagnostic tests and in the presence of nonspecific symptoms, and (2) that the course of various forms of leishmaniasis is well understood by scientists and by the doctors examining the veterans. As discussed in our report, we find these assumptions remain open to question. Moreover, during our exit conference, DOD and VA officials voiced agreement with our concerns in this regard. Insofar as the prevalence of this infection is still unknown and it is impossible to predict which veterans’ immune systems will be weakened, it is premature to discard leishmaniasis as a risk factor. Finally, regarding our evaluation of the Committee’s conclusion that low-level chemical exposures are unlikely to be associated with veterans’ health conditions, the Committee appears not to contest the fact that laboratory data document specific health effects in animals exposed to one or more organophosphate agents not detectable by the usual clinical tests. While the Committee notes that it recommended additional research in this area, we find it difficult to reconcile the Committee’s dismissal of such exposure as an “unlikely” contributor to veterans’ health problems with its acknowledgement of an absence of data on an important exposure scenario. Moreover, although the Committee apparently found no data to suggest a problem with low-level exposures, we found some data that do pose concerns. While PAC argues that these studies were done on animals, they are consistent with standard toxicological practice employed by the Environmental Protection Agency and others. The Committee’s insistence that such effects be demonstrated in humans appears unreasonable insofar as humans cannot be exposed to toxic substances for experimental research for obvious ethical reasons. Also, comparing occupational or 79 R. H. Stretch, P. D. Bliese et al. Psychological Health of Gulf War-Era Military Personnel. Military Medicine, vol. 161 (1996), pp. 257-61. Page 133 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VII Comments From the Presidential Advisory Committee on Gulf War Veterans’ Illnesses accidental exposures, such as the Rocky Mountain Arsenal exposures to sarin, to the possible exposures experienced by Gulf War veterans would provide some degree of information, but the value of such information is generally limited because each situation is different. PAC appears to have set an unusually restrictive standard for the evidence that would support any concern in this area. Page 134 GAO/NSIAD-97-163 Gulf War Illnesses Appendix VIII Major Contributors to This Report Sushil K. Sharma, Ph.D., Dr.P.H., Assistant Director National Security and Betty Ward-Zukerman, Ph.D., Project Manager International Affairs Dan Engelberg, Ph.D., Senior Evaluator Division, Washington, Nancy Ragsdale, Senior Evaluator (Communications Analyst) Joseph F. Murray, Assistant Director, Report Review D.C. Page 135 GAO/NSIAD-97-163 Gulf War Illnesses Glossary Aflatoxin Any of several carcinogenic toxic substances that are produced especially in stored agricultural crops, by molds. Anthrax An infectious disease of warm-blooded animals caused by a spore-forming bacterium transmissible to man and characterized by external ulcerating nodules or by lesions in the lungs. Biomarker A biological indicator, typically of exposure or of susceptibility to illness. Botulinum A spore-forming bacterium that secretes a toxin that is the cause of botulism, an acute paralytic disease. Chemical Agent Resistant A paintable covering used to protect against chemical and biological Coating attacks. Decontaminating Solution 2 An extremely corrosive and reactive solution used to decontaminate (DS2) material of chemical and biological weapons. Depleted Uranium A mixture of about 0.2 percent radioactive U-235 and the rest U-238 which is used in armor-piercing shells and armor plating because of its extreme density. Fibromyalgia A group of common rheumatic disorders characterized by pain, tenderness and stiffness of muscles, areas of tendon insertions and adjacent soft-tissue structures. Leishmania Any of a genus of flagellate protozoans that are parasitic in the tissues of vertebrates. L. tropica is a member of this genus. Mustard Gas Chemical warfare agents that blister the skin or any other part of the body they contact. They act on the eyes, mucous membranes, lungs, skin and blood-forming organs. They also damage the respiratory tract when inhaled and cause vomiting and diarrhea when ingested. Organophosphate-Induced A neurological condition characterized by enlarged axons (long, single Delayed Neuroathy (OPIDN) nerve cells) and axonal degeneration, caused by exposure to certain chemicals that inhibit cholinesterase, an enzyme important to nervous system functions. Pyridostigmine Bromide A drug that was taken by some U.S. troops during the Persian Gulf war to protect them against the nerve agent soman. Page 136 GAO/NSIAD-97-163 Gulf War Illnesses Glossary Ricin A biological warfare agent extracted from the seed of the castor plant. It blocks protein synthesis by altering the RNA, thus killing the cell. Sarin An extremely toxic chemical warfare agent that affects the nervous system. VX A persistent and extremely lethal nerve agent. Page 137 GAO/NSIAD-97-163 Gulf War Illnesses Glossary Page 138 GAO/NSIAD-97-163 Gulf War Illnesses Related GAO Products VAHealth Care: Observations on Medical Care Provided to Persian Gulf Veterans (GAO/T-HEHS-97-158, June 19, 1997). Chemical and Biological Defense: Protection of Critical Overseas Posts and Airfields Remains Largely Unaddressed (GAO/NSIAD-97-9, June 13, 1997). Operation Desert Storm: Evaluation of the Air Campaign (GAO/NSIAD-97-134, June 12, 1997). Defense Health Care: Medical Surveillance Has Improved Since the Gulf War, but Results in Bosnia Are Mixed (GAO/NSIAD-97-136, May 13, 1997). Chemical and Biological Defense: Emphasis Remains Insufficient to Resolve Continuing Problems (GAO/NSIAD-96-103, Mar. 29, 1996). Operation Desert Storm: Health Concerns of Selected Indiana Persian Gulf War Veterans (GAO/HEHS-95-102, May 16, 1995). Operation Desert Storm: Potential for Reproductive Dysfunction Is Not Being Adequately Monitored (GAO/T-PEMD-94-31, Aug. 5, 1994). Operation Desert Storm: Questions Remain on Possible Exposure to Reproductive Toxicants (GAO/PEMD-94-30, Aug. 5, 1994). Operation Desert Storm: Early Performance Assessment of Bradley and Abrams (GAO/NSIAD-92-94, Jan. 10, 1994). Operation Desert Storm: Problems With Air Force Medical Readiness (GAO/NSIAD-94-58, Dec. 30, 1993). Operation Desert Storm: Army Medical Supply Issues (GAO/NSIAD-93-206, Aug. 11, 1993). Operation Desert Storm: Improvements Required in the Navy’s Wartime Medical Care Program (GAO/NSIAD-93-189, July 28, 1993). Operation Desert Storm: Army Not Adequately Prepared to Deal with Depleted Uranium Contamination (GAO/NSIAD-93-90, Jan. 29, 1993). Operation Desert Storm: Full Army Medical Capability Not Achieved (GAO/NSIAD-92-175, Aug. 18, 1992). Page 139 GAO/NSIAD-97-163 Gulf War Illnesses Related GAO Products Operation Desert Storm: DOD Met Need for Chemical Suits and Masks, but Longer Term Actions Needed (GAO/NSIAD-92-116, Apr. 7, 1992). Defense Health Care: Efforts to Address Health Effects of the Kuwait Oil Well Fires (GAO/HRD-92-50, Jan. 9, 1992). Reproductive and Developmental Toxicants: Regulatory Actions Provide Uncertain Protection (GAO/PEMD-92-3, Oct. 2, 1991). Chemical Warfare: Soldiers Inadequately Equipped and Trained to Conduct Chemical Operations (GAO/NSIAD-91-197, May 29, 1991). Chemical Protective Suits: No Basis to Question Procuring Agency’s Acquisition Strategy (GAO/NSIAD-90-162, May 1, 1990). Hazardous Materials: DOD Should Eliminate DS2 From Its Inventory of Decontaminants (GAO/NSIAD-90-10, Apr. 25, 1990). Army Procurement: Unnecessary Restriction on Competition for New Chemical Protective Masks (GAO/NSIAD-88-66, Mar. 2, 1988). (713002) Page 140 GAO/NSIAD-97-163 Gulf War Illnesses Ordering Information The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Orders by mail: U.S. General Accounting Office P.O. Box 6015 Gaithersburg, MD 20884-6015 or visit: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (301) 258-4066, or TDD (301) 413-0006. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. For information on how to access GAO reports on the INTERNET, send an e-mail message with "info" in the body to: email@example.com or visit GAO’s World Wide Web Home Page at: http://www.gao.gov PRINTED ON RECYCLED PAPER United States Bulk Rate General Accounting Office Postage & Fees Paid Washington, D.C. 20548-0001 GAO Permit No. G100 Official Business Penalty for Private Use $300 Address Correction Requested
Gulf War Illnesses: Improved Monitoring of Clinical Progress and Reexamination of Research Emphasis Are Needed
Published by the Government Accountability Office on 1997-06-23.
Below is a raw (and likely hideous) rendition of the original report. (PDF)