National Academies Press: OpenBook
« Previous: Appendix H: IOM's Statements of Task and Conclusions for Agent Orange and Gulf War Reports
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Appendix I
Case Studies

CASE STUDY 1:
MENTAL DISORDERS’ PRESUMPTIONS

War and combat have long been considered to have acute and chronic impacts on the emotional well-being and mental health of those exposed. Indeed, of the two initial presumptive disabilities recognized in 1921, one was “neuropsychiatric disease,” a mental disorder (Veterans’ Bureau Act. 1921. Public Law 67-47. 67th Cong., 1st Sess., p. 154). The scientific understanding of mental disorders, as well as the role of war and combat in triggering them, have evolved over time. These factors have influenced the inclusion of mental disorders as presumptive disabilities among U.S. veterans. This case study is intended to review factors influencing the inclusion of various mental disorders as presumptive disabilities among U.S. veterans to date.

Scientific and Legislative History

Following World War I, two of the most common causes for hospitalizations identified among U.S. veterans were “neuropsychiatric disease” and tuberculosis. At that time, neuropsychiatric disease was considered to consist of a combination of delusions, hallucinations, and illogical thinking. Because of the lack of effective treatments at the time, neuropsychiatric disease was considered to be chronic (VA, 2006). In 1921, Senator Walsh of Massachusetts proposed an amendment to pending legislation, which was adopted and later modified, that removed the burden of proof for connecting military service and the development of a disorder from the veteran by granting presumptions of service connection for veterans diagnosed with neuropsychiatric disease or tuberculosis causing at least 10 percent disability within 2 years of military service (Veterans’ Bureau Act. 1921. Public Law 67-47. 67th Cong., 1st Sess., p. 154). In support of these presumptions, Senator Walsh stated that

It is very apparent to me that this wave of tuberculosis and of nervous and mental disease that has taken such a deadly hold and grip of late upon our ex-servicemen must have been contracted in the service. I feel, therefore, that we ought not continue this requirement of endless affidavits, necessarily involving long delay, in demonstrating the fact that their illness is of service origin.

(61 Cong. Rec. 4105 [daily ed. July 20, 1921], as referenced in VA, 1993a, p. 8)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

To emphasize the difficulty in particular for veterans with neuropsychiatric disease he added that

I think really that the most human feature of this amendment is the assistance it will render to those afflicted with nervous and mental disease in obtaining their compensation. When it is considered that the most important proof, the essential proof, to establish a claim for compensation must come from the man himself, and when it is realized that he is mentally afflicted and therefore can not, for instance, file affidavits from officers and servicemen with whom he served—since memory is usually defective and he can not remember whom his officers or comrades were—it becomes apparent how important is the change made by the bill.

(61 Cong. Rec. 4105 [daily ed. July 20, 1921], as referenced in VA, 1993a, p. 9)


At the time the legislation was enacted in 1921, there were few scientific studies to support or refute these assertions, and a flood of demands for presumptions for other disorders and for enhancement of those already approved followed soon thereafter.

By early 1923, the 2-year period from the time of discharge within which the neuropsychiatric disorder had to manifest itself was extended to 3 years despite administration objections that the extension was not supported by scientific evidence (VA, 1993a, p. 12). Later that same year veterans groups, including the Disabled American Veterans, the Veterans for Foreign Wars, and the American Legion, began calling to extend the period from 3 to 5 years. The lack of scientific evidence to justify the extension was noted by Dr. Earl Holt of the Veterans Bureau. In testimony to Congress, Dr. Holt stated that available statistics showed that neuropsychiatric disorder was just as common among the civilian population as among the military population, and that the extension of presumptions was unwarranted due to uncertainty over the causes of psychiatric disabilities arising after service (VA, 1993a, p. 14). In response to the Depression, the Economy Act of 1933 eliminated all benefits based on presumptive service connection. In its place the President was given broad authority to prescribe rules concerning eligibility for disability compensation including “the nature and extent of proofs, and presumptions” (emphasis added) for various classes of veterans (Economy Act of 1933 ch. 3 § 4, 48 Stat. at 9, as referenced in VA, 1993a, p. 17).

Following strong protests from World War I veterans, Congress subsequently moved to reenact presumptive service-connection conditions. President Roosevelt vetoed the legislation stating that he thought the Economy Act had settled the issue that a service-connected disability was a “question of fact rather than a question of law” in which each individual case would be “considered on its merits” rather than by

legislative dicta which, contrary to fact, provide that thousands of individual cases of sickness that commenced 4, 5, or 6 years after the termination of the war were caused by war services.

(VA, 1993a, p. 19)


Congress overrode the veto, and the measure was enacted into law (Independent Offices Appropriations Act. 1935. Public Law 141. 73rd Cong., 2d Sess.). The tension between those intent on being inclusive and generous in presumptive benefits to veterans and those wanting presumptions to be more firmly grounded by evidence of causation continues today.

The pertinent provisions of that 1935 Act are now found in 38 USC § 1702, “Presumption Relating to Psychosis,” and provide that eligible veterans who developed an active psychosis

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

within 2 years of military discharge “shall be deemed to have incurred such a disability in the active military, naval, or air service.” That language has remained essentially unchanged except for the expansion of eligibility for veterans of subsequent periods of military conflict up to and including those serving in today’s current war. The Committee was unable to obtain information as to how many veterans rated for mental disabilities have been service connected by use of the presumption set forth in section 1702.

Additional legislation concerning psychiatric presumptions has been confined to former prisoners of war (POWs). These psychiatric presumptions were the result of strong concern about the well-being of a group of veterans who suffered extreme privations and of POW studies undertaken to ascertain the effect of their captivity.

Psychiatric disorders, by their very nature, have created challenges for decision making on presumptions. For many of these disorders, symptoms can only be obtained by self-reporting methods, making it more difficult to distinguish those who truly exhibit symptoms and those who do not. Many early studies focused on excessive hospitalizations for mental disorders among veterans and associations of these disorders with military service rather than causation (Beebe, 1975; Cohen and Cooper, 1954). In addition, as the medical community’s understanding of mental disorders has evolved over time so have the definitions of various mental disorders. Congress, the Department of Veterans Affairs (VA), the National Academy of Sciences (NAS), and a large collection of veterans groups and scientists have all been involved in questions related to presumptions and psychiatric disorders in veterans.

The first legislation to create statutory presumptions specifically applicable to POWs was enacted in 1970 (Veterans Disability Compensation Increase Act. 1970. Public Law 91-376. 91st Cong., 2d Sess.). Under Public Law 91-376, psychosis was to be considered a service-connected disability provided it became manifest within 2 years of separation from service in the military. In justifying the statutory presumptions, the accompanying report said that because of the conditions of captivity and the “kinds of long-range harm that may have been caused,” it was “sometimes difficult for a former prisoner of war to establish some time after completion of military service” that a disability is related to military service (U.S. Congress, House of Representatives, Committee on Veterans’ Affairs, 1970, p. 7).

From October 1972 through August 1976, VA issued a number of program guides with respect to Public Law 91-376. These guides were characterized as suggestions for the guidance of personnel in the handling of disability claims filed by former POWs. The program guides instructed that in light of the “frequent paucity of records” in POW claims, special attention should be given to POW experiences in determining the relationship of the disability to service (VA, 1980, p. 118). The duration and circumstances of imprisonment were to be associated with pertinent medical principles in making determinations.

The burden of proof as to the occurrence of a POW episode was shifted from the claimant to the government. The “unusual hardship and isolation from society” resulting from POW life meant that an “extended period of readjustment to ordinary conditions of life is essential.” Claims for individual unemployability (IU) were to receive liberal construction, and if the threshold disability percentage for the IU benefit was not met, the claim was to be submitted to the VA central office for further consideration. Finally, it was emphasized that presumptions were rebuttable only where there was “affirmative evidence to the contrary” (VA, 1980, p. 119).

In 1975, the second phase of the Follow-Up Studies of World War II and Korean War Prisoners, entitled Morbidity, Disability, and Maladjustment, was issued (Beebe, 1975). Its author, Gilbert W. Beebe, observed at the outset that studies of the long-term effects of catastrophic

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

stress are “difficult to make and the frequently multidimensional character of such stress severely limits inferences about the etiologic role of specific components, such as malnutrition, social isolation, sensory deprivation, physical punishment, compulsory reeducation, and the like.” He further noted that recent studies of survivors of World War II German concentration camps provided useful information on a persistent defective state marked by “severe permanent psychiatric residuals and by nonspecific somatic symptoms” (Beebe, 1975, pp. 400-401).

Beebe’s findings were that morbidity, some types of maladjustment, and disability were elevated in POWs relative to controls especially for Pacific theater veterans. Beebe stated that the “most remarkable and long-lasting differentials” were seen in hospitalization for psychoneurosis and for psychosis (schizophrenia). Data obtained supported a finding that many Pacific theater and Korean War POWs had “permanent psychologic impairments.” Beebe also noted that European theater POWs did not go “unscathed” with respect to psychoneurosis hospitalization (Beebe, 1975, p. 421).

In the Veterans Disability Compensation and Survivor Benefits Act of 1978 (Public Law 95-479. 95th Cong., 2d Sess.), Congress included provisions requiring VA to carry out a “comprehensive study on the compensation awarded to, and the health-care needs of” former POWs. The results of the study were to include such administrative and legislative recommendations as “may be necessary to assure that former prisoners of war receive compensation and health-care benefits for all disabilities which may reasonably be attributed to their internment” (Public Law 95-479, sec. 305; emphasis added). The legislative history indicates that the study was prompted by questions about the adequacy of repatriation examinations and by concerns that health conditions that may have appeared minor at the time were becoming progressively more debilitating. It also cited Beebe’s 1975 follow-up study for the proposition that POWs had excess morbidity, and many were suffering from what was termed a “POW syndrome” (VA, 1993a, p. 51).

As VA was preparing its study, the third phase of the follow-up studies entitled Mortality to January 1, 1976, authored by Robert J. Keehn (1980), was issued. That study continued to find increased risks of mortality among World War II Pacific theater and Korean War POWs, though the excess diminished over time. For Pacific theater veterans the principal cause for the mortality increase was tuberculosis and trauma, while for former Korean War POWs it was trauma. In describing this trauma, the report found that for Pacific theater POWs suicide was responsible for two-thirds and accidents one-third of the excess deaths (Keehn, 1980). The study also reported that anxiety neurosis accounted for 12.7 percent of all service-connected conditions for former POWs, compared to a rate of approximately 4 percent for all veterans receiving compensation.

Given a consistent finding of persistent psychologic impairment and reported problems of adjustment to civilian life, the report concluded that

The finding of increased mortality due to trauma, both accidental (including cases of masked suicide) and suicide, in former prisoners of war is not surprising. Increased feelings of frustration, anger, and tension lead to impatience and impulsive actions that are likely to contribute to both the risk and severity of accidental and self-inflicted injury.

(Keehn, 1980, p. 209)


Shortly thereafter VA delivered its Study of Former Prisoners of War (VA, 1980) to Congress as required by Public Law 95-479. The principal finding that VA stated was “essential to the entire study” was that

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

[T]he POW experience—characterized by starvation diet, poor quality or nonexistent medical care, “death marches,” executions, and tortures—has historically been an extremely harsh and brutal experience.

(VA, 1980, p. 161)


Concerning the quality of POW repatriation procedures, exams, and resultant medical records, the study found that the prescribed procedures were generally well designed and reflected “state-of-the art medical knowledge and technique at the time they were designed,” which “if followed would have provided former POWs with adequate records” (VA, 1980, p. 71). Examination of VA claims folders, however, found that less than 20 percent of European theater POWs who filed disability claims had evidence of a repatriation examination. For Pacific theater and Korean War veterans, records of repatriation examinations were located in 60 percent and 85 percent of their files, respectively. In those cases where records were located, VA reviewer physicians judged 67 percent of World War II and 85 percent of Korean War POW repatriation exams as “providing a good or adequate basis for evaluating physical or psychiatric conditions.” The limitations were that over half of the examinations contained either no medical history or poor history of health status prior to capture. In addition, about one-third of the examinations had “inadequate evaluations of the POW’s mental status and psychiatric conditions” (VA, 1980, p. 72).

The “most remarkable finding” of the study, according to VA, was that anxiety neurosis was “the most prevalent service-connected condition of the former POWs under study, from the time of their repatriation to the present.” Anxiety neurosis accounted for 12.7 percent of all service-connected conditions of former POWs, which was three times the rate of all veterans receiving compensation (VA, 1980, p. 95). VA stated that

The significance of this disability relative to veterans controls remains regardless of the length of internment. This is especially apparent among former European theater POWs, in which those POWs interned less as well as more than 6 months exhibit significantly higher rates of anxiety neurosis compared to other service-connected World War II veterans.

(VA, 1980, p. 95)


The study had also required VA to analyze procedures used to determine eligibility for benefits with a particular emphasis on the statutory and regulatory provisions unique to POWs. VA concluded that

Former POWs generally have received special consideration in keeping with statutory and procedural provisions in terms of medical evaluations and disability compensation. Limitations in knowledge as to the long-term effects of the stresses and deprivations experienced by prisoners of war is a major obstacle for decision makers.

(VA, 1980, p. 128)


VA also reported that over the years it had changed its approach to the adjudication of POW claims by gradually developing flexibility in such areas as substantiation of claims in the absence of medical records for periods of internment and in the presumption of service incurrence for certain disabilities. This changed approach reflected the “evolution of the law,” and the degree of flexibility roughly coincided with “advancements in medical knowledge” concerning the serious effects of imprisonment on health (VA, 1980, p. 121).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The literature review of health problems of former POWs, a fourth requirement of the study, included eyewitness accounts of disabilities during captivity, epidemiologic follow-up studies, analysis of concentration camp populations, and discussions of former POW family and social problems. VA said that its review showed that the higher rate of health problems experienced were “related to the malnutrition, torture, climatic exposure, and other deprivations of internment.” The epidemiologic follow-up studies indicated that “residuals of these and other disabilities have persisted until the present time.” According to VA, it was particularly noteworthy that the psychological problems of former POWs, especially those of World War II, “closely resemble[d]” those of concentration camp survivors of the same period (VA, 1980, p. 154). The K-Z syndrome, as discussed in Beebe’s 1975 report, included the symptoms:

General anxiety and nervousness, “startle” reaction, insomnia and nightmares, phobias, psychosomatic complaints, memory lapses, moodiness, inferiority complex, obsession with the past, depression, apathy, and survivor guilt.

(VA, 1980, p. 154)


The psychological literature on K-Z syndrome and what VA termed “other forms of psychic stress” revealed a significantly higher amount of family and social maladjustment as evidenced by inadequate functioning in father and parent roles, and higher rates of unemployment and disability compensation among POWs (VA, 1980, p. 154).

Two major legislative recommendations were contained in the study submitted to Congress. First, VA recommended that the law be amended to authorize eligibility for VA health care to former POWs for any disease or neuropsychiatric disability. VA observed that studies by the National Research Council (NRC) and NAS showed former POWs generally had higher mortality and morbidity rates and that this was reflected in their higher rates of service-connected disabilities. Yet, despite the special consideration given to POW claims, their adjudication was complicated by the frequent absence of medical information at the time of repatriation and by the fact that “medical science cannot, at this time, conclusively determine on an individual basis the origins of some disabilities particularly prevalent among former POWs” (emphasis added). Authorizing comprehensive VA inpatient and outpatient medical care for any disease or neuropsychiatric condition “would remove access barriers to VA medical care for those former POWs currently classified in a lower than 50 percent service-connected priority category” (VA, 1980, pp. 163-164).

The second legislative recommendation was to modify the existing statutory presumption of service connection for psychosis. VA proposed to “eliminate the requirement that psychoses suffered by POWs must become manifest within 2 years following service separation before the rebuttable presumption of service connection arises” (VA, 1980, p. 164). VA said that its literature review indicated that psychosis related to the POW experience “frequently appears years after service, not just immediately after separation,” citing NAS/NRC follow-up studies published between 1946 and 1980 in support thereof (VA, 1980, p. 164).

VA also reported that it was undertaking several administrative actions as a result of its study. First, forthcoming guidelines on “post-traumatic stress neurosis” would have “explicit reference to former POWs as well as other combat veterans.” Post-traumatic stress neurosis was a term scheduled on October 1, 1980, to become part of VA’s official diagnostic classification system to describe this anxiety neurosis. The guidelines would “specifically be used to diagnose, treat, and rate former POWs with anxiety neurosis or similar neurotic disorders” (emphasis added). VA said this change was justified because former POWs had experienced a wide

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

range of psychological problems. In addition, anxiety neurosis had been the most prevalent disability among former POWs according to NRC epidemiologic studies and VA compensation data. VA also added that an analysis comparing anxiety neurosis and length of internment demonstrated that it remained a “statistically significant service-connected disability among former POWs regardless of the amount of time in prison camp” (VA, 1980, pp. 165-166; emphasis added).

VA announced that it would adopt a standardized protocol for disability compensation examinations for all former POWs similar to that developed by the military for former Vietnam POWs (VA, 1980).

Congressional hearings followed the receipt of VA’s study of POWs in both 1980 and 1981. Legislation was reported in June 1981 and enacted into law on August 14 as the Former Prisoner of War Benefits Act of 1981 (Public Law 97-37. 97th Cong., 1st Sess.). Among its provisions, the requirement that psychosis manifests itself within 2 years of separation of service in order to qualify for a service-connection presumption was deleted as recommended by VA. An additional presumption termed any of the anxiety states was added to the statute. As noted in House testimony, “… psychosis related to the POW experience frequently appears years after service, and not just immediately after separation. This is understandable in view of the psychological torture and ‘brainwashing’ to which these POWs were subjected” (U.S. Congress, House of Representatives, Committee on Veterans’ Affairs, 1981, p. 6).

Three additional presumptions for POWs were added between 1984 and 1988. The first, in the Veterans’ Compensation and Program Improvements Amendments of 1984 (Public Law 98-223. 98th Cong., 2d Sess.) added “dysthymic disorder” to the list of disabilities developing any time after a POW’s separation from service for which a presumption of service connection would apply. Senator Alan K. Simpson, chairman of the Senate Committee on Veterans Affairs, termed the inclusion a clarification of the original intent of the Former Prisoner of War Benefits Act of 1981. Speaking on the floor of the Senate during consideration of the measure he said that

The complexity of anxiety states, anxiety neuroses, posttraumatic stress disorder, and dysthymic disorders and their associated and sometimes interrelated diagnoses inadvertently resulted in a lack of clarity regarding the granting of service connection for depression.

(VA, 1993a, p. 55)


Further explanation was contained in the committee’s report accompanying the measure that states that at the time the Senate reached agreement with the House on the 1981 Act:

[The Senate] was not aware that there would be cases in which former POWS suffering from nonpsychotic depressive disorders would not be diagnosed as suffering from posttraumatic stress disorder and therefore not adjudged under VA guidelines to be service-connected disabled…. The committee intends that this addition would correct the inadvertent oversight in the original legislation and establish a presumption for a mental disorder that is linked in scientific literature to the POW experience.

(VA, 1993a, pp. 55-56)


Since the late 1980s, a number of well-designed studies have supported increased psychiatric morbidity among former POWs as well as among other veterans experiencing combat. In 1988,

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

results from the Vietnam Experience Study (VES) conducted by the Centers for Disease Control and Prevention (CDC) were released. The VES selected a random sample of Army personnel discharged between 1965 and 1971. Those who served a single tour in Vietnam (n = 9,324) were compared to a random sample of Army personnel discharged during the same period and who served a single tour of duty elsewhere (n = 8,989). Veterans who were alive at the time of study completed a telephone interview, in-person psychological examinations, and assessments for reproductive outcomes (CDC, 1988, p. 2702).

The VES found that Vietnam veterans had a 45 percent excess of deaths in the first 5 years after discharge in comparison to non-Vietnam veterans. These deaths were largely due to motor vehicle accidents, suicides, and homicides. After the initial 5-year period, the death rates among the two groups of veterans remained approximately the same. Nevertheless, Vietnam veterans were more likely to meet diagnostic criteria for alcohol abuse or dependence, generalized anxiety disorder, and depression. Among Vietnam veterans, 14.7 percent met criteria for posttraumatic stress disorder (PTSD) previously in life and 2.2 percent met criteria for PTSD during the month prior to the examination (CDC, 1988, p. 2705). Though a constellation of psychiatric symptoms among this Vietnam group was similar to studies previously conducted on World War II POWs, this study was significant in the fact that the veterans enrolled in this study were selected due to their combat theater rather than their POW status.

Several studies published in 1991 examined chronic depression among former POWs. Engdahl et al. (1991) found that long-term chronic depressive symptomatology persisted over 40 years and was elevated among POWs of all theaters when compared to control groups. Age, education, medical symptoms during captivity, and level of social support were found to be related to later levels of adjustment. A second, longitudinal study published the same year by Page et al. (1991) elaborated on previous research to show that not only was depressive symptomatology highly elevated in World War II and Korean War POWs, but it was elevated to the point where these populations closely resembled a clinical population of recovering depressives. Two major conclusions from this longitudinal study were that treatment during captivity is statistically linked with depressive symptoms, and that differences in these symptoms were attributable to captivity-related treatment, even when age at capture and education level were considered.

Conducting a 40-year follow-up of U.S. World War II and Korean War former POWs, Engdahl and Page et al. (1991) measured captivity trauma variables and individual protective variables (i.e., age, education, medical symptoms during captivity, social support) to compare with current depressive symptoms. Although depressive symptoms persisted more than 40 years with the knowledge that PTSD and generalized anxiety disorders are known to occur with elevated frequency in POW populations, the degrees of individual protective variables were related to levels of adjustment. This study made a case for the need to examine former POWs that adjusted well in order to understand both the role of specific protective variables and posttrauma adjustment and resiliency.

Page’s (1991) work continued with respect to the validity and reliability of some of the work cited above. Despite the heavy reliance on survey data, a noticeable shortage of reports on the effects of nonresponse bias on the measurement of depression existed. Longitudinal data presented opportunities for different types of nonresponse bias, but these data could also be useful in modeling for bias because of previously collected data. Page found that a predictive model shows nonresponse bias on the reporting of depressive symptoms among former World War II and Korean War POWs to be small.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

In 1992, the Institute of Medicine (IOM) produced a report entitled The Health of Former Prisoners of War. This longitudinal study focused on morbidity and was initiated in 1986 and built on the earlier work of Cohen and Cooper (1954), Nefzger (1970), and Keehn (1980), which has been previously discussed. Veterans were invited to a medical center to undergo the VA protocol exam, which included a comprehensive physical and psychiatric examination. In addition, a face-to-face psychiatric interview and a battery of psychological tests were administered. A caution was provided at the outset of the report that due to low response rates there could be “no confidence … that the group of respondents accurately reflects the composition of all former POWs” (IOM, 1992, p. 4). Nevertheless, the results were presented as descriptive data that constituted the largest national collection of POW examinations ever gathered and analyzed. Some of the findings confirmed earlier studies while other findings were suggestive and served the purpose of generating more definitive research studies. The report urged a “maximum of reasonable caution in (the) interpretation” of the results (IOM, 1992, p. 5). The report brought attention to the high prevalence of a number of medical conditions for POWs as compared to the controls, especially with regard to psychiatric illness. Prevalence rates for over 20 different medical conditions were discussed, with key results including the following:

  • Pacific theater POWs had higher prevalence rates of PTSD, ulcer, schizophrenia, and generalized anxiety than European theater and Korean War POWs.

  • Visual symptoms were associated with higher prevalence rates of cerebrovascular disease, ulcers, asthma, and PTSD.

  • Korean War POWs showed higher prevalence rates for schizophrenia.

(IOM, 1992, pp. 6-11)


The IOM report stated that many of the organ-specific findings were familiar and that the increased prevalence of depressive disorders, PTSD, and generalized anxiety among POWs was not unexpected.

To better understand the characteristics that affect POWs’ reintegration into civilian life, Engdahl et al. published a report in 1993 that investigated long-term responses to captivity trauma among former POWs. Engdahl et al. reported that symptoms at 20 years following release were related to those at 40 years following release. Many factors known to affect POWs’ long-term adjustment were not included in the study (e.g., combat exposure, postwar social support). This was due in part to their statistical infrequency or skewed nature (i.e., family history of mental illness, marital status at capture, military rank at capture). Trauma response was found to be determined by an interaction of characteristics of the individual and characteristics of the trauma, not primarily one over the other. The authors suggested that trauma response, from an evolutionary standpoint, may be better understood as adaptive due to its persistent nature.

As previously discussed, Congress has delegated to the VA Secretary the general authority to prescribe “all rules and regulations … with respect to the nature and extent of proof and evidence and the method of taking and furnishing them in order to establish the right to benefits under such laws” (38 USC § 501[a]). Pursuant to that authority, VA published evidentiary presumptions with respect to establishing PTSD claims in 1993 (VA, 1993b). The regulation initially observes that service connection for PTSD requires “medical evidence establishing a clear diagnosis of the condition, credible supporting evidence that the claimed in-service stressor actually occurred, and a link, establishing by medical evidence, between current symptomatology and the claimed in-service stressor” (VA, 1993b, p. 3). The regulation provided that

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

If the claimed stressor is related to combat, service department evidence that the veteran engaged in combat or that the veteran was awarded the Purple Heart, Combat Infantryman Badge, or similar combat citation will be accepted, in the absence of evidence to the contrary, as conclusive evidence of the claimed in-service stressor.

(VA, 1993b, pp. 3-4; emphasis added)


In addition, the regulation provided that status as a POW would similarly be regarded as evidence of an in-service stressor.

PTSD continued to gain great attention. Page et al. (1997) examined the literature on PTSD in POWs and compared lifetime PTSD prevalence among POWs and control subjects. After a follow-up period of 40 years, differences in prevalence rates existed between POWs and control subjects for depressive disorders as well as generalized anxiety. All groups of POWs shared nearly the same lifetime and current PTSD rates. Among World War II POWs, however, roughly half of the POWs who once suffered from PTSD were not currently diagnosed with that condition, suggesting the possible presence of chronic, stable PTSD in the other half of the POWs evaluated. However, the authors stated that this may be explained by the symptoms causing less stress on the first group of individuals. The authors suggested that those with higher distress levels should be evaluated for secondary symptoms of PTSD, such as depression. They concluded that

Sensitivity toward older war veterans is vital. An awareness that their PTSD may have gone unnoticed by other health-care professionals for decades should encourage direct clinical inquiries about possible PTSD symptoms. We strongly recommend a structured interview…. PTSD symptoms have been all too common, yet undiagnosed among older war veterans, especially POWs.

(Page et al., 1997, p. 157; emphasis added)


Work on PTSD continued into the new millennium when World War II and the Korean War POW interviews were examined for two separate index measures at two points in time—1965 and 1990. Results from Gold et al. (2000) supported previous research highlighting the severe psychological consequences of POW status 40-50 years following captivity. Trauma severity during captivity was found to be the best predictor of current PTSD symptomatology.

The Veterans Benefits Act of 2003 (Public Law 108-183. 108th Cong., 1st Sess.) included provisions that removed the 30-day minimum confinement requirement for 5 of the 16 POW presumptive conditions. Included in those 5, for which no minimum confinement was required, were (a) psychosis, (b) any of the anxiety states, and (c) dysthymic disorder (or depressive neurosis). In justifying the change, a Senate committee report on a similar bill observed that POWs were often treated brutally and, even if treated humanely, often suffered extreme mental anguish. Thus, the “30-day minimum requirement for purposes of presumptive service connection may be too restrictive for certain conditions” (U.S. Congress, Senate, Committee on Veterans’ Affairs, 2003, p. 10).

Lessons Learned

Presumptive decisions for mental disorders have been established for veterans who are former POWs and for veterans who developed chronic mental problems during or shortly after military service. The subjective nature and self-reporting aspects of mental disorders have made it

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

more difficult to determine the mental disorders that should be presumptively service connected. Although legislation has been informed by the scientific evidence available at the time (Beebe, 1975; CDC, 1988; Cohen and Cooper, 1954; Engdahl and Page, 1991; Keehn, 1980; Nefgzer, 1970; Page et al., 1997), the scientific evidence has been limited by inconsistency surrounding the disorders that have been included in the research. For example, if the limited and suggestive evidence led to presumptive decisions for PTSD, dysthymic disorders, and any anxiety state among former POWs, then there does not appear to be a clear basis for excluding other mental disorders with equal or stronger evidence of connection to being a POW, such as major depression or substance abuse. The presumptive decisions with regard to these mental disorders demonstrate that these decisions have been influenced not only by the evidence, but also by political and social considerations that apply to these veterans and the specific mental disorders they manifest. The need to develop stronger evidence and consistency with regard to these disorders is great, particularly in light of evidence of high rates of disorders among military personnel currently assigned to Iraq. This case study illustrates the need for a process that can continually carry out research while updating the scientific evidence used in presumptive disability decision making. Improved future studies will be aided by pre- and postdeployment mental health assessments of Service members and more thorough assessment and documentation of exposures while deployed. This information will facilitate a deeper understanding of the relationship between military service and the subsequent development of mental health disorders.

References

Beebe, G. W. 1975. Follow-up studies of World War II and Korean War prisoners: Morbidity, disability, and maladjustments. American Journal of Epidemiology 101(5):400-422.

CDC (Centers for Disease Control). 1988. Vietnam experience study—psychosocial characteristics. Journal of the American Medical Association 259(18):2701-2707.

Cohen, B. M., and M. Z. Cooper. 1954. A follow-up study of World War II prisoners of war. Washington, DC: Government Printing Office.

Engdahl, B. E., and W. F. Page. 1991. Psychological effects of military captivity. In Epidemiology in military and veteran populations. Proceedings of the Second Biennial Conference, March 7, 1990. Washington, DC: National Academy Press. Pp. 49-66.

Engdahl, B. E., W. F. Page, and T. W. Miller. 1991. Age, education, maltreatment, and social support as predictors of chronic depression in former prisoners of war. Social Psychiatry and Psychiatric Epidemiology 26(2):63-67.

Engdahl, B. E., A. R. Harkness, R. E. Eberly, W. F. Page, and J. Bielinski. 1993. Structural models of captivity trauma, resilience, and trauma response among former prisoners of war 20 to 40 years after release. Social Psychiatry and Psychiatric Epidemiology 28(3):109-115.

Gold, P. B., B. E. Engdahl, R. E. Eberly, R. J. Blake, W. F. Page, and B. C. Frueh. 2000. Trauma exposure, resilience, social support, and PTSD construct validity among former prisoners of war. Social Psychiatry and Psychiatric Epidemiology 35(1):36-42.

IOM (Institute of Medicine). 1992. The health of former prisoners of war: Results from the medical examination survey of former POWs of World War II and the Korean Conflict. Washington, DC: National Academy Press.

Keehn, R. J. 1980. Follow-up studies of World War II and Korean Conflict prisoners: Mortality to January 1, 1976. American Journal of Epidemiology 111(2):194-211.

Nefzger, M. D. 1970. Follow-up studies of World War II and Korean War prisoners: Study plan and mortality findings. American Journal of Epidemiology 91(2):123-138.

Page, W. F. 1991. Using longitudinal data to estimate non-response bias. Social Psychiatry and Psychiatric Epidemiology 26:127-131.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Page, W. F., B. E. Engdahl, and R. E. Eberly. 1991. Prevalence and correlates of depressive symptoms among former prisoners of war. Journal of Nervous and Mental Disease 179(11):670-677.

Page, W. F., B. E. Engdahl, R. E. Eberly, C. S. Fullerton, and R. J. Ursano. 1997. Persistence of PTSD in former prisoners of war. In Posttraumatic stress disorder: Acute and long-term responses to trauma and disaster, edited by C. S. Fullerton and R. J. Ursano. Washington, DC: American Psychiatric Press. Pp. 147-158.

U.S. Congress, House of Representatives, Committee on Veterans’ Affairs. 1970. Service-connected compensation increase for veterans. House Report 91-1166. 91st Cong., 2nd Sess. June 9.

U.S. Congress, House of Representatives, Committee on Veterans’ Affairs. 1981. Prisoner of war benefits and health-care services act of 1981. House Report 97-28. 97th Cong., 1st Sess. May 4.

U.S. Congress, Senate, Committee on Veterans’ Affairs. 2003. Veterans’ Benefits Enhancements Act. Senate Report 108-169. 108th Cong., 1st Sess. October 21.

VA (Department of Veterans Affairs). 1980. Study of former prisoners of war. Washington, DC: Government Printing Office.

VA. 1993a. Analysis of presumptions of service connection. Paper presented to the IOM’s Committee on the Evaluation of the Presumption Disability Decision-Making Process for Veterans, Washington, DC.

VA. 1993b. Direct service connection (post-traumatic stress disorder). Final Rule. Federal Register 58(95):29109.

VA. 2006. VA history in brief. http://www1.va.gov/opa/feature/history/index.asp (accessed July 5, 2007).

CASE STUDY 2:
MULTIPLE SCLEROSIS PRESUMPTION

This case study examines the 1962 decision to grant presumptive service-connected disability to any veteran with multiple sclerosis (MS) whose disease is diagnosed within 7 years of separation from the military. This case study illustrates the challenges of evaluating scientific evidence and compensating veterans when the etiology of a disease is unknown and the possibility of a service-related cause cannot be excluded with certainty.

Background

MS is a neurological disease characterized by inflammation, destruction, and scarring of myelin cells that protect the neurons in the central nervous system. Symptoms of MS typically present in young adulthood, with a prevalence in the United States of 1 in 1,000 (IOM, 2001, p. 17; Noseworthy et al., 2000). Women have higher rates of both incident and prevalent disease, and studies in the United States and Canada suggest that the female-to-male disease ratio has been increasing over time (Figures I-1 and I-2).

Two features of MS have proven challenging in evaluating the scientific basis for a presumption of service connection. First, the point of onset of MS can be difficult to determine. There are no clinical symptoms unique to MS; rather, a variety of neurological symptoms are possible. These symptoms often occur in distinct episodes that may at least partially resolve, and the nature of the symptoms can vary over time. Diagnosis of MS can be delayed either because an affected individual may not seek medical attention for a neurological symptom that may be resolving or because physicians may not recognize symptoms as suggestive of MS.

Second, the etiology and pathogenesis of MS remain largely unknown. A variety of environmental, genetic, and autoimmune factors have been implicated, and perhaps the most intriguing observation in the epidemiology of MS is the finding that individuals at higher latitudes are at increased risk of disease (IOM, 2001). In the Nurses’ Health Study, for example, a graded increase in risk of MS was observed in those in the northern United States compared with the

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

FIGURE I-1 Prevalence of multiple sclerosis in the United States (from the National Health Interview Survey).

SOURCE: Noonan et al., 2002.

FIGURE I-2 Female-to-male ratio of multiple sclerosis in a Canadian registry.

SOURCE: Orton et al., 2006.

southern United States (Hernán et al., 1999). Individuals migrating from a high-risk to low-risk area appear to adopt the risk of their new home, although the age at migration may influence how this risk is reassigned (IOM, 2001). Why location may be important in the epidemiology of MS

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

is unknown, and the nature of the environmental exposure associated with higher latitudes that potentially contributes to MS risk is unclear. Triggers or exposures ranging from viral illness to lack of vitamin D related to lower levels of sun exposure have been postulated (IOM, 2001).

The Need for a Presumption

The rationale for the MS presumption is different from that of other presumptions discussed in the case studies and appears to have shifted over time. While most other presumptions serve to fill gaps in evidence for exposure (e.g., the Agent Orange presumptions) or gaps in evidence for association (e.g., cardiovascular disease among amputees), the original basis for this presumption fits in neither of these two categories. This presumption stems from the VA’s interest in compensating disease and disability that has its onset during military service. A defined list of chronic conditions have long been granted service connection if diagnosed within 1 year of separation from the military. These diseases are insidious in onset and difficult to diagnose. Presumption is granted because the possibility that these conditions were present during military service cannot be excluded with certainty. The presumption for MS, therefore, does not fill gaps in evidence for exposure or association, but rather gaps in evidence for the timing of disease onset. This distinction is important, because the scientific evidence needed to inform this presumption is neither the information about exposure during military service nor for association between military service and disease; rather the scientific evidence needed for this presumption is evidence for an insidious disease course that makes onset during military service nearly impossible to exclude.

Over time, Congress created a special exception for MS in comparison to other presumptive health outcomes, extending the presumptive period for diagnosis of MS first from 1 to 2 years following military service (Veterans’ Benefits Act of 1957. Public Law 85-56. 85th Cong., 2d Sess.; VA, 1949; Act of October 12, 1951. ch. 499, 65 Stat. 421, as cited in VA, 1993). Subsequently, Congress extended the presumptive period for diagnosis of MS to 3 years (Act of August 25, 1959. Public Law 86-187. 86th Cong., 1st Sess., as cited in VA, 1993) and then finally to 7 years following military service in 1962 (Veterans’ Disability Compensation Increase Act of 1962. Public Law 87-645. 87th Cong., 2d Sess., as cited in VA, 1993). The rationale for this exception was based in part on the continued evidence supporting the insidious onset of this disease. However, with this extension the debate in the Federal Register also shifted to the inability to exclude a military exposure as the possible etiological cause of MS in veterans. This shift is significant because the scientific evidence necessary to support this type of presumption would include evaluation of evidence for possible exposures during military service and evidence linking military service with MS.

Because the debate surrounding the MS presumption has taken place over many years (since the 1930s) in congressional committees, the subtle but important shift in the rationale for this presumption and its implications for the scientific evidence needed to support it have not been systematically evaluated. As a result, considerable disagreement and uncertainty remain as to whether this presumption is supported by the scientific evidence.

A Brief History of the MS Presumption

A 1933 executive order first granted presumptive service connection to a defined list of chronic diseases if these were diagnosed within 1 year of separation from the military. The rationale for this order was that these conditions were

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

of “such an insidious nature” that the disease did not become manifest to a ten-percent degree immediately upon inception but required as much as a year from the date of inception to become manifest.

(VA Solicitor, 46 Op. Sol. 140 [9-15-39] as cited in VA, 1993, p. 17)


The assumption was that these conditions were eligible for direct service connection because of their temporal relation to military service; their insidious nature required this lag period outlined in the presumption. These chronic diseases were all characterized by their insidious onset and not by any direct evidence that military service caused these conditions. Other illnesses covered in this presumption, in addition to MS, include hypertension, diabetes, and atherosclerosis. The MS presumption was debated by Congress in 1948 and codified in 1949. Again, the language of the debate suggested that this presumption was meant to be an extension of direct service connection for a limited number of diseases that, though present during service, were difficult to diagnose during the period of service (Veterans’ Chronic and Tropical Diseases Act of 1948. Public Law 80-748. 80th Cong., 2d Sess.; VA, 1949).

An internal VA memorandum acknowledges this logic as the basis for this presumption and comments on the low likelihood that military service actually caused these illnesses. It states that

The diseases … are indeed of chronic type, and their presence within a year after discharge raises a strong probability that part of the course of the disease … coincided with the period of military service, but the likelihood that any of their course was influenced by the facts or circumstances of service is extremely remote.

(Internal VA memorandum, as cited in VA, 1993, p. 28)


In 1951, Congress singled out MS from the other chronic illnesses and extended the period of diagnosis from 1 to 2 years following military service. In 1959, Congress again extended this period to 3 years, and in 1962 the period was extended to 7 years following military service. The congressional decision was based in part on scientific testimony from the National Institutes of Health and the National Multiple Sclerosis Society that 7 years “was not an unreasonable period to recognize as the interval between onset and diagnosis” (VA, 1993, p. 33). The VA opposed these multiple extensions, noting that

Although the exact cause of the disease is unknown, there is nothing in the circumstances of military service in time of war which from a medical and scientific standpoint would warrant a presumption of fact that a manifestation of the disease … years after discharge is in any way related to the factor or circumstances of services. In this connection it does not appear that the disease is any more prevalent among the veteran population than the nonveteran population.

(VA statement, as cited in VA, 1993, p. 33)


This debate illustrates the two different lines of reasoning that have entered the discussion regarding the MS presumption. On the one hand, the insidious nature of disease may make a presumption necessary because of the lag required for diagnosis; the scientific evidence necessary is simply that of the length of time on average required for diagnosis. On the other hand, because the etiology of MS is unknown and an exposure that occurs during military service cannot be ruled out as the cause of MS, a presumption may be necessary on this basis as well. Here the scientific basis for association between exposure and outcome becomes relevant.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

During the same time in which the presumptive period for the diagnosis of MS was repeatedly extended, this disease was a focus of epidemiologic investigation by scientists at VA and the Medical Follow-Up Agency (MFUA) (Berkowitz and Santangelo, 1999). This research activity was driven largely by neurologists at VA who had become interested in developing treatments for MS (VA Multiple Sclerosis Study Group, 1956, 1957), and leaders of MFUA, who were interested in making use of the extensive medical records on young individuals to learn more about the manifestation of diseases in that population generally (Berkowitz and Santangelo, 1999). The extension of the presumptive period for service connection to 7 years enhanced the ability to conduct these epidemiologic investigations; at one time, two-thirds of the cases of MS estimated to occur among all veterans of World War II were being studied by VA and MFUA investigators (Kurtzke et al., 1979, p. 1233). Studies of veterans with MS yielded insights into the epidemiology of this disease, confirming that MS appears to be a “disease of place” and that environmental exposures in childhood and young adulthood, as yet still unidentified, may likely play a role in the etiology of this disease (Kurtzke and Page, 1997, p. 204). In addition to the important insights into the epidemiology of MS, these studies of veterans notably failed to find any evidence that military service (or particular exposures during military service) placed veterans at a greater risk for MS than the general population (Kurtzke and Page, 1997; Kurtzke et al., 1979, 1985, 1992; Norman et al., 1983; Page et al., 1993, 1995; Wallin et al., 2000).

Lessons Learned

This presumption straddles the line between compensation for disease that manifests during military service and compensation for the future adverse health effects that occurred as a result of specific exposures during military service. Both lines of argument were used to justify or oppose the MS presumption, but the scientific basis for each of these arguments is very different. In the first case, timing is the standard (specifically the possibility of disease onset during military service), and evidence for association between an exposure and outcome is not required. In the second case, evidence from some association between exposure during service and future disease is necessary. Congress did not call for a systematic review of the scientific literature on this topic; such a review might have allowed for more evidentiary discussion of the premise behind this presumption and the type of evidence that might be necessary to support it.

The exception of repeatedly lengthening the presumptive period of diagnosis following military service was justified because MS is difficult to diagnose and because of the possibility that some, as yet still unidentified, environmental exposures may play a role in its etiology. However, these two characteristics are also true of a variety of other chronic conditions for which presumptions have not been established. MS almost certainly received particular attention because of heightened awareness of MS as a result of the numerous high-profile publications on its epidemiology based on the study of veterans.

The epidemiological insight on MS gained from the study of veterans deserves comment. These studies of MS have had far-reaching benefit beyond those realized by the veterans themselves and have formed the basis of much of the understanding of this complicated chronic disease.

References

Berkowitz, E. D., and M. J. Santangelo. 1999. The Medical Follow-Up Agency: The first fifty years 1946-1996. Washington, DC: National Academy Press.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Hernán, M. A., M. J. Olek, and A. Ascherio. 1999. Geographic variation of MS incidence in two prospective studies of US women. Neurology 53(8):1711-1718.

IOM. 2001. Multiple sclerosis: Current status and strategies for the future. Washington, DC: National Academy Press.

Kurtzke, J. F., and W. F. Page. 1997. Epidemiology of multiple sclerosis in U.S. veterans: 7. Risk factors for MS. Neurology 48(1):204-213.

Kurtzke, J. F., G. W. Beebe, and J. E. Norman, Jr. 1979. Epidemiology of multiple sclerosis in U.S. veterans: 1. Race, sex, and geographic distribution. Neurology 29(9):1228-1235.

Kurtzke, J. F., G. W. Beebe, and J. E. Norman, Jr. 1985. Epidemiology of multiple sclerosis in U.S. veterans: 3. Migration and the risk of MS. Neurology 35(5):672-678.

Kurtzke, J. F., W. F. Page, F. M. Murphy, and J. E. Norman, Jr. 1992. Epidemiology of multiple sclerosis in U.S. veterans: 4. Age at onset. Neuroepidemiology 11(4-6):226-235.

Noonan, C. W., S. J. Kathman, and M. C. White. 2002. Prevalence estimates for MS in the United States and evidence of an increasing trend for women. Neurology 58:136-138.

Norman, J. E., Jr., J. F. Kurtzke, and G. W. Beebe. 1983. Epidemiology of multiple sclerosis in U.S. veterans: 2. Latitude, climate and the risk of multiple sclerosis. Journal of Chronic Diseases 36(8): 551-559.

Noseworthy, J. H., C. Lucchinetti, M. Rodriguez, and B. G. Weinshenker. 2000. Multiple sclerosis. New England Journal of Medicine 343(13):938-952.

Orton, S. M., B. M. Herrera, I. M. Yee, W. Valdar, S. V. Ramagopalan, A. D. Sadovnick, and G. C. Ebers. 2006. Sex ratio of multiple sclerosis in Canada: A longitudinal study. The Lancet Neurology 5(11):932-936.

Page, W. F., J. F. Kurtzke, F. M. Murphy, and J. E. Norman, Jr. 1993. Epidemiology of multiple sclerosis in U.S. veterans: 5. Ancestry and the risk of multiple sclerosis. Annals of Neurology 33(6):632-639.

Page, W. F., T. M. Mack, J. F. Kurtzke, F. M. Murphy, and J. E. Norman, Jr. 1995. Epidemiology of multiple sclerosis in U.S. veterans: 6. Population ancestry and surname ethnicity as risk factors for multiple sclerosis. Neuroepidemiology 14(6):286-296.

VA (Department of Veterans Affairs). 1949. Veterans claims miscellaneous amendments. Federal Register 14:568.

VA. 1993. Analysis of presumptions of service connection. Washington, DC: Department of Veterans Affairs.

VA Multiple Sclerosis Study Group. 1956. Isoniazid in the treatment of multiple sclerosis: Report on VA Cooperative Study. Transactions of the American Neurological Assocociation (81st meeting): 128-131.

VA Multiple Sclerosis Study Group. 1957. Isoniazid in the treatment of multiple sclerosis: Report on Veterans Administration Cooperative Study. JAMA 163(3):168-172.

Wallin, M. T., W. F. Page, and J. F. Kurtzke. 2000. Epidemiology of multiple sclerosis in U.S. veterans: 8. Long-term survival after onset of multiple sclerosis. Brain 123(Pt 8):1677-1687.

CASE STUDY 3:
PRISONERS OF WAR PRESUMPTIONS

This study examines the scientific evidence and governmental policy decisions concerning veterans who were held as POWs. Covering a period of 60 years commencing with the end of World War II, the case study examines the policy considerations that resulted in the creation of various legal presumptions to facilitate former POWs’ access to VA health and disability benefits. It focuses particular attention on epidemiologic and other studies examining the residual health effects of the POW experience and their influence on the policy decisions reached.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Background

In most wars a number of combatants are captured and interned as POWs. In our history there have been over 500,000 American POWs. Almost 70 percent of those were captured by both sides during the Civil War (VA, 2005, p. 4).

In the 20th century over 142,000 service personnel were held as POWs, about 17,000 of which died while held prisoner. Ninety-eight percent of these deaths occurred during World War II and the Korean War (VA, 2006). POW deaths during World War II varied greatly depending on the theater of operation. About 40 percent of the nearly 28,000 Pacific theater POWs—most of whom were interned in the Philippines in the early months of 1942—died while in captivity (VA, 2005, p. 5).

In the European/Mediterranean theater about 1 percent of the over 94,000 POWs perished while interned. One in four of the European/Mediterranean theater POWs were captured during the Battle of the Bulge in late December 1944 and January 1945. During the Korean War 2,700, or nearly 38 percent, of slightly more than 7,100 service personnel died while in captivity (VA, 2005, p. 5).

Of the 125,000 POWs returned to U.S. military control in the last century, almost 4,000 were veterans of World War I, none of whom are alive today. The 116,000 World War II veterans released at the end of hostilities constitute the largest cohort (93 percent) of 20th-century POWs with the remainder either veterans of Korea, Vietnam, or later conflicts (VA, 2006).

VA estimates that at the end of 2005 there were just over 29,000 living POWs of whom 91 percent are World War II veterans and 7 percent veterans of the Korean War. Almost 17,000 POWs were receiving VA disability compensation as of August 2006, and 13,000 of them were rated 100 percent disabled (VA, 2006).

World War II and Its Aftermath

Congressional concern about the difficulties that combat veterans might encounter in establishing claims for service-connected disability benefits was manifested in legislation enacted within 2 weeks of the U.S. entry into World War II. That law, the Veterans Determination of Service Connection of Disabilities Act of 1941 (Public Law 77-361. 77th Cong., 1st Sess.), directed the VA Administrator, in considering benefit claims, to give “due consideration” to the “places, types and circumstances” of the claimant’s service and statutorily codified VA’s liberal evidentiary rules for establishing proof of service connection (VA, 1993, pp. 22-23). The legislative history reflects that it was intended to overcome the possible lack of official records that might occur under wartime conditions.

Consistent with the 1941 legislation, VA, in issuing its 1945 Schedule for Rating Disabilities (the Rating Schedule), acknowledged the difficulties that former POWs might have in establishing service-connected disability claims. The rating schedule provided that in considering claims with respect to tropical diseases, dysentery, and other gastrointestinal diseases, “great weight must be assigned to tropical service and to imprisonment or internment under unsanitary conditions, or food deprivation” (VA, 1993, pp. 43-44; emphasis added). Additional instructions issued in 1946 concerning the Rating Schedule provided that “prisoner of war experience requires special consideration” with respect to claims involving malnutrition, intestinal parasites, weakness and fatigability, and neuropsychiatric disorders. Claims examiners were instructed that the “existence of any chronic disease that may be associated with the circumstances of imprisonment should be carefully checked and reported on” (VA, 1993, pp. 43-44; emphasis added).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The 1950s

Growing concern by members of Congress about the well-being of POWs and the privations they experienced, particularly in the Pacific theater, led to the introduction of legislation and hearings in 1948. One measure would have provided that any veteran held prisoner for a continuous period of at least 2 years would have been conclusively presumed to be totally disabled for a period of 5 years from his release. In addition, any death occurring during that 5-year period would be presumed service connected in the absence of clear and convincing evidence to the contrary. After the initial 5-year period the presumption of total disability would be rebuttable but again only by clear and convincing evidence (VA, 1993).

VA opposed the legislation on the ground that service connection should not be granted in the absence of a factual showing of disability. Additionally, it detailed the special consideration for POWs contained in its operating procedures. Although the measure was not acted upon by the House Veterans Affairs Committee, concern continued about the mental and physical effects of imprisonment as various members of Congress introduced bills in the early 1950s to require studies of POW mortality and morbidity. The sponsors anticipated that the studies might provide evidence to support a conclusive presumption for service connection for the purposes of VA hospitalization benefits as well as standards that could be employed in evaluations of POW disability claims. Again, VA demurred, noting the special consideration it was giving to POW claims. VA also pointed out that it was already conducting studies related to the POW experience (VA, 1993).

Ultimately, Congress enacted the War Claims Act Amendments of 1954 that incorporated provisions of these bills with respect to studying mortality and morbidity experiences of POWs, while omitting any of the provisions relating to the proposed development of standards for conclusive presumptions (VA, 1993, p. 45).

In September 1954, VA published A Follow-Up Study of World War II Prisoners of War, coauthored by Bernard Cohen of the NRC and Maurice Cooper of VA (Cohen and Cooper, 1954). The study, which was initiated prior to the passage of the War Claims Act, was characterized by the authors as an exploratory record follow-up study of POWs released at the end of World War II that “does not purport to be more than an examination of the surface of the problem.” Rather, its purpose was to “delineate the broader consequences of imprisonment that manifest themselves in increases in mortality, hospitalization, disability, health status, and work adjustment, to measure the magnitude of these changes and to describe the gross findings” (Cohen and Cooper, 1954).

A key finding of the study was that Pacific theater veterans had a high rate of mortality in the first 2 years following liberation. By contrast, returned European theater POWs did not have a significant difference in mortality compared with control groups. Some 64 percent of Pacific theater POW deaths were attributable to either tuberculosis or accidents. The frequency of tuberculosis deaths was not unexpected given malnutrition and crowded, unsanitary living quarters, but unanswered questions remained as to whether the degree and duration of malnutrition that Pacific theater POWs experienced could be sufficient to “result in obscure irreversible structural and functional changes that might be expected to affect longevity.” With respect to excess accidental deaths, the authors recommended more detailed examination in future studies, particularly as to whether there might be a psychological basis involved (Cohen and Cooper, 1954).

The authors concluded that

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

[T]he mortality findings and their implications suggest the existence of organic and emotional residuals of imprisonment severe enough to be factors affecting survival, and raise many questions concerning the future survival potential of those still alive and their long run morbidity and disability expectations.

(Cohen and Cooper, 1954; emphasis added)


The early morbidity and disability observations were found to be “quite consistent with those of mortality.” Pacific prisoners exhibited a wide variety of illness that occurred with frequencies in nearly every major category of disease that were far in excess of those shown by European POWs or control groups. Chronic conditions that were noticeably frequent and persistent were tuberculosis, residuals of malnutrition, psychoneurosis, ophthalmologic changes, gastrointestinal disorders, and cardiovascular conditions. Among European POWs there was not found “a great deal more illness” than control groups although there was a “relative excess in malnutrition, psychoneurosis, and gastrointestinal disorders.” The authors suggested that “[s]ome if not all” of these conditions deserved “more intensive investigation” (Cohen and Cooper, 1954).

The authors concluded their report by recommending a “continuation of the follow-up [study] to detect late effects or confirm early findings, and the desirability of more intensive study in certain areas” (Cohen and Cooper, 1954). The 1954 study was followed by a Department of Health, Education, and Welfare (HEW) study, Effects of Malnutrition and Other Hardships on the Mortality and Morbidity of Former United States POWs and Civilian Internees (HEW, 1956), issued in 1956 in response to the direction of the War Claims Act Amendments of 1954 (Public Law 83-774. 83rd Cong., 2d Sess, as referenced in VA, 1993). It relied heavily on the NRC/VA study and its findings and conclusions mirrored that report. It recommended, “in conformity with the stated policy of VA,” that “[p]articular attention” should be given in evaluation of POW claims for claimed disabilities resulting from “prolonged malnutrition and other conditions shown by the NRC/VA study to exist in high incidence in Pacific POWs” as well as to the difficult question of “complaints which cannot today be evaluated by objective measurements or test” (HEW, 1956).

For the remainder of the decade, there was little activity regarding establishment of statutory presumptions for POWs with the exception of a measure to officially declare all POWs to be disabled and entitled to full retirement pay, which did not receive favorable consideration from Congress (VA, 1993, pp. 45-46).

The general issue of statutory presumptions, however, did receive considerable attention as a result of a 1956 report of the President’s Commission on Veteran’s Pensions. Established by executive order, the commission—typically referred to as the Bradley Commission after its chairman, General Omar Bradley who headed VA at the conclusion of World War II—addressed statutory presumptions in one of the 70 recommendations it issued in 1956. Specifically, it recommended withdrawing the “presumption of service connection for chronic disease, tropical diseases, psychoses, tuberculosis, and multiple sclerosis as now listed” (President’s Commission on Veterans’ Pensions, 1956, p. 178). Relying in large part on the result of a survey of physicians, the Bradley Commission asserted that as to those conditions:

Accepted medical principles can reasonably and accurately establish the onset of a disease and the disability process. Where there is reasonable doubt, the law provides for the doubt to be resolved in favor of the veteran…. The physicians surveyed were in general

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

agreement that service connection should be determined in accordance with sound medical principles and not by fiat.

(President’s Commission on Veterans’ Pensions, 1956, p. 178)


Hearings on the Bradley Commission’s recommendations by the House Committee on Veterans Affairs in 1956 elicited strong reactions from national veteran service organizations (VSOs). Witnesses from the Disabled American Veterans challenged the assertion that physicians were in agreement about whether certain chronic diseases discovered 1 or 2 years after separation had their inception in service, and it was for this reason that Congress authorized statutory presumptions (VA, 1993).

The American Legion representative testified that the question was “not entirely a medical one” and that the “purpose of a presumption is to free the veteran from carrying an unconscionable burden of proof” in establishing disability compensation entitlement. The witness added that it “is a rebuttable presumption … [which] enables the government, if medical knowledge is what the Bradley Commission says that it is, to rebut the presumption.” He concluded that until American medicine can “determine with more than a reasonable degree of accuracy” whether or not certain types of diseases had their inception in service, the veteran should be entitled “to a presumption that his disease or disability, within reasonable periods now or to be specified, was the result of his service” (VA, 1993, p. 41).

This position was essentially endorsed by the committee which declined to act on the Bradley Commission recommendation. In hearings the following year, its chairman, Olin “Tiger” Teague from Texas stated that presumptions were enacted “because Congress did not agree with many of the medical findings” of VA physicians and consequently it was the “only way we can force proper administration” (VA, 1993, p. 42). The American Legion’s senior medical consultant testified that the Legion would not ask for presumptive service connection “unless the same was justified on reasonable medical grounds.” Presumptions were often required, he said, because of inadequately trained VA personnel which resulted in a lack of uniformity in the application of the law (VA, 1993, p. 42).

The 1960s

Legislation seeking to establish presumptions for POWs began to increase in the 1960s. Some measures would have established a presumption of service connection for any disability for which a veteran was seeking service connection if that veteran had been a Pacific theater POW confined for more than 2 years. Other bills would have increased the existing presumptive period for chronic and tropical diseases to 5 years in the case of any veteran who had been a POW. Still other proposals would have created an irrebuttable presumption that all former World War II and Korean War POWs who had been interned for at least 3 years were service-connected disabled to a 50 percent degree. Variants on that proposal would have reduced the minimum period of confinement to 1 year. None of these POW-specific presumptions received favorable consideration from Congress (VA, 1993).

The 1970s

In the late 1960s, a three-phase program of research, Follow-Up Studies of World War II and Korean War Prisoners, was undertaken by the NRC with funding from VA. The first phase, Study Plan and Mortality, published in 1970, was authored by M. Dean Nefzger who acknowledged difficulties in obtaining generally useful information because

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Prison experience is a manifold of inadequate food, exposure, disease, physical abuse, and emotional torment. These various ingredients appear to be thoroughly mixed and interdependent. Their relative intensities probably vary from time to time, from place to place, and from man to man; and it seems impossible to separate any one component of this complex experience for the total.

(Nefzger, 1970, p. 124)


Although this “limits the scope and specificity of inferences that may be drawn statistically,” the author said that follow-up studies might offer the “hope of learning more about the late effects of stress that is recognized as severe even though not well differentiated.” They may also “suggest areas for more penetrating study by other methods” (Nefzger, 1970, p. 124).

The report generally confirmed the earlier findings and trends of the 1954 report by Cohen and Cooper. Pacific theater POWs had excess mortality compared to control groups, although that significantly diminished over time. European POWs showed no significant excess mortality. The report did note, however, that mortality experience for a small group of European prisoners hospitalized for malnutrition immediately after release suggested causes similar to Pacific theater POWs (Bard, 1994).

Korean War POWs, who were included in this report, also experienced excess mortality rates throughout a 12-year follow-up when compared to controls. For Korean War veterans, trauma was the most common cause of death whereas Pacific POW deaths were attributed to tuberculosis, accidents, and cirrhosis of the liver (Nefzger, 1970).

On the question of excess mortality due to disease, Nefzger wrote

There is a great temptation to conclude that the apparent excess of deaths from diseases of the digestive system, including cirrhosis, resulted from malnutrition during imprisonment. That might be the case, but so specific an interpretation is hard to defend. It is possible that such diseases are an indirect consequence of imprisonment if that experience has contributed to a different standard or manner of living since repatriation.

(Nefzger, 1970, p. 137)


Finally, the report concluded that the contrasts between European prisoners generally and those European POWs hospitalized for malnutrition, and the contrast between European and Pacific prisoners, “suggest a positive association of stress in prison with later mortality,” although no general conclusion could be drawn from the data (Nefzger, 1970, pp. 137-138).

The first legislation to create statutory presumptions specifically applicable to POWs was enacted in 1970 (Veterans Disability Compensation Increase. 1970. Public Law 91-376. 91st Cong., 2d Sess.). The act had its origins in S. 3348 that, as introduced, would have presumed service connection for any disability suffered by any POW held captive for more than 6 months. VA opposed the measure, reiterating that it gave special consideration to claims by former POWs, and arguing that the mere fact of 180 days confinement by itself did not justify service connection for “any disability the veteran may acquire at any time during the balance of his life,” VA also maintained that it would be “discriminatory” to those POWs who may have sustained equal or greater privations but were confined for less than the specified period—a point agreed to by the American Legion (U.S. Congress, Senate, Committee on Finance, 1970, p. 10).

As finally reported to and passed by the full Senate, the bill had been modified to eliminate any minimum confinement period. Instead, the Senate substituted a requirement that the veteran

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

had to have suffered from “dietary deficiencies, forced labor, or inhuman treatment in violation of the terms of the Geneva Convention of July 27, 1929” (U.S. Congress, Senate, Committee on Finance, 1970, p. 12). If former POWs met these criteria and had any of seven specified nutritional or gastrointestinal diseases, those diseases would be “considered to have been incurred in or aggravated by such service notwithstanding that there is no record of such disease during the period of service” (U.S. Congress, Senate, Committee on Finance, 1970, p. 12).

Under the act, “psychosis” was also to be considered a service-connected disability provided it became manifest within 2 years of separation from service.

In justifying the statutory presumptions, the accompanying report said that due to the conditions of captivity and the “kinds of long-range harm that may have been caused,” it was “sometimes difficult for a former prisoner of war to establish some time after completion of military service” that a disability is related to military service (U.S. Congress, Senate, Committee on Finance, 1970, p. 4).

When the Senate-passed measure was considered by the House, the bill was further amended to add back the provision of not less than 6 months of confinement in addition to the Geneva Code violations required by the Senate. No explanation was given as to the rationale for the 6-month requirement. Following Senate concurrence in the amendments, the measure was enacted as Public Law 91-376 on August 12, 1970 (Veterans Disability Compensation Increase. 1970. Public Law 91-376. 91st Cong., 2d Sess.).

From October 1972 through August 1976, VA issued a number of program guides with respect to Public Law 91-376. These guides were characterized as suggestions for the guidance of personnel in the handling of disability of claims filed by former POWs. The program guides instructed that in light of the “frequent paucity of records” in POW claims, “special attention should be given to POW experiences in determining the relationship of the disability to service.” The duration and circumstances of imprisonment were to be associated with pertinent medical principles in making determinations (VA, 1980, p. 118).

The burden of proof as to the occurrence of a POW episode was shifted from the claimant to the government. A veteran’s statement as to wounds or injury just prior to imprisonment was to be accepted as proof of actual incurrence when residual disability attributable to service was found. The “unusual hardship and isolation from society” resulting from POW life meant that an “extended period of readjustment to ordinary conditions of life is essential.” Claims for IU were to receive liberal construction and if the threshold disability percentage for the IU benefit was not met, the claim was to be submitted to the VA Central Office for further consideration. Finally, it was emphasized that presumptions were rebuttable only where there was “affirmative evidence to the contrary” (VA, 1980, p. 119).

In 1975, the second phase of the Follow-Up Studies of World War II and Korean War Prisoners, entitled Morbidity, Disability and Maladjustments, was issued (Beebe, 1975). Its author, Gilbert W. Beebe, observed at the outset that studies of the long-term effect of catastrophic stress are “difficult to make and the frequently multidimensional character of such stress severely limits inferences about the etiologic role of specific components, such as malnutrition, social isolation, sensory deprivation, physical punishment, compulsory reeducation, and the like.” He further noted that recent studies of survivors of World War II German concentration camps provided useful information on a persistent defective state marked by “severe permanent psychiatric residuals and by nonspecific somatic symptoms” (Beebe, 1975, pp. 400-401). The researchers who conducted those studies believed that

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

[S]tarvation caused permanent damage that went largely unrecognized in the early years after release and came to be appreciated only as rehabilitation programs proved ineffective for so many of those who had suffered most severely.

(Beebe, 1975, p. 401)


Beebe’s findings were that morbidity, some types of maladjustment, and disability were elevated in POWs relative to controls especially for Pacific theater veterans. The “most remarkable and long-lasting differentials” were seen in hospitalization for psychoneurosis and for psychosis (schizophrenia). Data obtained supported a finding that many Pacific theater and Korean War POWs had “permanent psychologic impairments.” But he also noted that European theater POWs did not go “unscathed” with respect to psychoneurosis hospitalization (Beebe, 1975, p. 421).

Conditions causing excess morbidity among World War II Pacific theater POWs were nutritional disorders, neurologic problems, gastrointestinal, genitourinary, and bone diseases. Finally, the study also found higher rates of cardiovascular diseases for which atherosclerotic disease was a major contributor which warranted further scrutiny (Beebe, 1975, pp. 421-422).

Following enactment of Public Law 91-376, several bills were introduced that would have added additional chronic diseases to the presumptive list and/or extended the period of time from separation from service that they first be manifested. None were enacted. The Veterans Health Care Amendments of 1979 (Public Law 96-22. 96th Cong., 1st Sess.) did, however, authorize outpatient dental care benefits for all former POWs detained for 6 months or more. Although the law did not base dental benefit entitlement upon an explicit presumption—nor authorize disability compensation benefits—the Senate committee report accompanying the measure stated that many POWs developed dental conditions as a result of prolonged nutritional deprivation suffered while captive. It justified the provision in order “to provide that VA has full authority to meet the dental needs of former prisoners of war resulting from the conditions of their captivity.” The requirement that the former POW have been confined for 6 months or more was added because such “nutritional deficiencies result from a prolonged state of deprivation” and it was consistent with the 6-month requirement that was attached to other POW-related diseases (VA, 1993, p. 50).

In the Veterans Disability Compensation and Survivor Benefits Act of 1978 (Public Law 95-479. 95th Cong., 2d Sess.), Congress included provisions requiring VA to carry out a “comprehensive study on the compensation awarded to, and the health care needs of” former POWs. The results of the study were to include such administrative and legislative recommendations as “may be necessary to assure that former prisoners of war receive compensation and health-care benefits for all disabilities which may reasonably be attributed to their internment” (VA, 1993, p. 58; emphasis added).

The study had four requirements. First, VA was to analyze the adequacy of repatriation procedures and medical exams and the resultant medical records of POWs. Second, the study was to set forth the “types and severity of disabilities particularly prevalent” among former POWs “in various theaters of operations at various times” (emphasis added). Third, VA was charged with analyzing procedures used in determining health-care benefits and in adjudicating disability compensation claims, including an “analysis of the current use of statutory and regulatory provisions specifically relating to former prisoners of war.” Finally, VA was directed to survey and analyze “all of the medical literature on health-related problems of former prisoners of war” (Veterans Disability Compensation and Survivor Benefits Act of 1978. Public Law 95-479. 95th Cong., 2d Sess.).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The legislative history indicates that the study was prompted by questions about the adequacy of repatriation examinations and by concerns that health conditions that may have appeared minor at the time were becoming progressively more debilitating. It also cited Beebe’s 1975 follow-up study for the proposition that POWs had excess morbidity, and many were suffering from what was termed a “POW syndrome” (VA, 1993, p. 51).

The 1980s

At the beginning of the decade, the third phase of the follow-up studies, Mortality to January 1, 1976, authored by Robert Keehn, was issued (Keehn, 1980). That study continued to find increased risks of mortality among World War II Pacific theater and Korean War POWs, though the excess diminished over time. For Pacific theater veterans the principal cause for the mortality increase was tuberculosis and trauma, while for former Korean War POWs it was trauma. As to trauma, the report found that for Pacific theater POWs, suicide was responsible for two-thirds and accidents one-third of the excess deaths.

Given a consistent finding of “persistent psychologic impairment” and reported problems of adjustment to civilian life, the report concluded that

The finding of increased mortality due to trauma, both accidental (including cases of masked suicide) and suicide in former prisoners of war is not surprising. Increased feelings of frustration, anger, and tension lead to impatience and impulsive actions which are likely to contribute to both the risk and severity of accidental and self-inflicted injury.

(Keehn, 1980, p. 209)


Keehn also found excess deaths due to cirrhosis of the liver for all World War II and Korean War POWs appearing in the 10th year of follow-up. The author said that it was unclear if the increased deaths resulted from “poor nutrition as a prisoner, from the variety of diseases (viral, parasitic) experienced during captivity, or from increased alcohol consumption or a combination of these” (Keehn, 1980, p. 210). He added that

Although the role of diet in the etiology of alcoholic cirrhosis has not been clearly defined, it seems likely that a poor diet during captivity might predispose the liver to alcohol damage.

(Keehn, 1980, p. 210)


Finally, the study did not find excess POW deaths from malignant neoplasms or chronic and degenerative diseases, concluding that there was “no evidence to support the hypothesis that the stresses of captivity have accelerated degenerative changes in former prisoners” (Keehn, 1980, p. 210).

Thereafter, also in 1980, VA delivered its Study of Former Prisoners of War (VA, 1980) to Congress as required by Public Law 95-479 (Veterans Disability Compensation and Survivor Benefits Act of 1978. Public Law 95-479. 95th Cong., 2d Sess.). The principal finding which VA described as “essential to the entire study” was that

[T]he POW experience—characterized by starvation, diet, poor quality or nonexistent medical care, “death marches,” executions and torture—has historically been an extremely harsh and brutal experience.

(VA, 1980, p. 161)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Concerning the initial question about the quality of POW repatriation procedures, exams, and resultant medical records, the study found that the prescribed procedures were generally well designed and reflected “state-of-the-art medical knowledge and technique at the time they were designed,” which “if followed would have provided former POWs with adequate records” (VA, 1980, p. 71). Examination of VA claims folders, however, found that less than 20 percent of European theater POWs who filed disability claims had evidence of a repatriation examination. For Pacific theater and Korean War veterans, records of repatriation examinations were located in 60 percent and 85 percent of their files, respectively. In those cases where records were located, VA reviewer physicians judged 67 percent of World War II and 85 percent of Korean War POW repatriation exams as “providing a good or adequate basis for evaluating physical or psychiatric conditions.” The limitations were that over half of the examinations contained either no medical history or poor history of health status prior to capture. And, significantly, about one-third of the examinations had “inadequate evaluations of the POWs mental status and psychiatric conditions” (VA, 1980, p. 72).

Addressing the question of the types and severity of disabilities particularly prevalent among POWs in various theaters at various times, VA relied heavily on the NRC-funded follow-up reports previously described (Beebe, 1975; Keehn, 1980; Nefzger, 1970), and on VA mortality and morbidity data. The study found that Pacific theater former POWs were the most severely disabled group followed closely by Korean War POWs. Although European POWs had lower rates of disability, those rates still exceeded that of other World War II veterans. Studies were still continuing on Vietnam era veterans and hence unavailable at that time (VA, 1980).

Although the findings provided support for the contention that an Asian internment environment was harsher, the study cautioned that any attempt to apply it to individual cases would “be mistaken as conditions varied from camp to camp, thus resulting in a wide spectrum of disability even within the same theater.” The report specifically noted, contrary to popular opinion, that with the exception of some facilities where aviators were confined, most German stalags where American foot soldiers were held “did not abide by even minimum standards for POW treatment set forth in the Geneva Convention” (VA, 1980, p. 31).

The “most remarkable finding” of the study according to VA was that anxiety neurosis was “the most prevalent disability among former POWS from time of repatriation to the present.” Anxiety neurosis accounted for 12.7 percent of all service-connected conditions of former POWs, which was three times the rate of all veterans receiving compensation (VA, 1980, p. 95). VA found that

The significance of this disability relative to veteran controls remains regardless of the length of internment. This is especially apparent among former European theater POWs in which those POWs interned less as well as more than 6 months exhibit significantly higher rates of anxiety neurosis compared to other service-connected World War II veterans.

(VA, 1980, p. 95)


The study also confirmed that “systemic and malnutrition related diseases—malaria, beriberi, pellagra—are prevalent among former POWs, especially those interned in Asia.” Among former World War II Pacific theater and Korean War POWs there were also statistically significant service-connected disabilities with respect to eye diseases, respiratory diseases, and gastrointestinal diseases. Pacific theater former POWs also had statistically significant rates for genitourinary, psychoneurological, and cardiovascular diseases (VA, 1980, pp. 95-96).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The study’s third requirement was to analyze procedures used to determine eligibility for benefits with a particular emphasis on the statutory and regulatory provisions unique to POWs. VA concluded that

Former POWs generally have received special consideration in keeping with statutory and procedural provisions in terms of medical evaluations and disability compensation. Limitations in knowledge as to the long-term effects of the stresses and deprivations experienced by prisoners of war is a major obstacle for decision makers.

(VA, 1980, p. 128)


VA also reported that over the years it had changed its approach to the adjudication of POW claims by gradually developing flexibility in such areas as substantiation of claims in the absence of medical records for periods of internment and in the presumption of service incurrence for certain disabilities. This changed approach reflected the “evolution of the law” and the degree of flexibility roughly coincided with “advancements in medical knowledge” concerning the serious effects of imprisonment on health (VA, 1980, p. 121).

To support its assertion that POWs were accorded special consideration, VA compared their compensation rates with that of other veterans. Although less than 10 percent of war veterans were on the disability compensation rolls, 43.6 percent of former POWs were being compensated. Korean War POWs ranked highest with 59 percent followed by 50.6 percent of Pacific theater POWs and 42.2 percent of European theater POWs (VA, 1980, pp. 78-93).

Some 22.2 percent of non-POW veterans on the compensation roles had severe disabilities (rated 50 percent or more disabled) compared to 48.8 percent of Pacific theater, 20.1 percent European theater, and 34.7 percent Korean War POWs. Veterans on the disability compensation rolls who are rated at 60 percent or more are entitled to be paid at the total disability rate if they are determined to be individually unemployable. Of all veterans on the rolls, 5.3 percent were rated unemployable compared to 22 percent of World War II Pacific theater, 5.3 percent European theater, and 8.9 percent Korean War POWs (VA, 1980, pp. 78-93).

On the question of lack of uniformity raised by VSOs in deciding POW claims among regional offices, VA’s response was that an inquiry had been initiated by the General Accounting Office (GAO) in 1974-1975. A sample of cases had been selected and examined at four regional offices. GAO found “occasional variances in disability percentages … but no differences as to granting service connection” (VA, 1980, p. 127). Following this initial review, VA reported that GAO decided against conducting a formal study.

The literature review of health problems of former POWs, the fourth requirement of the study, included eyewitness accounts of disabilities during captivity, epidemiologic follow-up studies, analysis of concentration camp populations, and discussions of former POW family and social problems. VA said that its review showed that the higher rate of health problems experienced were “related to malnutrition, torture, climatic exposure, and other deprivations of internment.” The epidemiologic follow-up studies indicated that “residuals of these and other disabilities have persisted until the present time” (VA, 1980, p. 154).

Particularly noteworthy, according to VA, was that the psychological problems of former POWs, especially those of World War II, “closely resemble[d]” those of concentration camp survivors of the same period. Known as K-Z syndrome—and discussed in Beebe’s 1975 report—symptoms included

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

General anxiety and nervousness, “startle” reaction, insomnia and nightmares, phobias, psychosomatic complaints, memory lapses, moodiness, inferiority complex, obsession with the past, depression, apathy, and “survivor guilt.”

(VA, 1980, p. 154)


The psychological literature on K-Z syndrome and other forms of psychic stress revealed a “significantly higher amount of family and social maladjustment as evidenced by inadequate functioning in father/parent roles, and higher rates of unemployment and disability compensation” (VA, 1980, p. 154).

VA concluded its literature review by observing that many important questions were left unanswered including whether deaths due to trauma and cirrhosis were directly related to the POW experience, and if the arteriosclerosis experienced by former POWs directly related to the stress of internment (VA, 1980).

Two major legislative recommendations were contained in the study submitted to Congress. First, VA recommended that the law be amended to authorize eligibility for VA health care to former POWs for any disease or neuropsychiatric disability. VA observed that NRC/VA studies showed former POWs generally had higher mortality and morbidity rates and that this was reflected in their higher rates of service-connected disabilities. Yet, despite the special consideration given to POW claims, their adjudication was complicated by the frequent absence of medical information at the time of repatriation and by the fact that “medical science cannot, at this time, conclusively determine on an individual basis the origins of some disabilities particularly prevalent among former POWs.” Authorizing comprehensive VA inpatient and outpatient medical care for any disease or neuropsychiatric condition “would remove access barriers to VA medical care for those former POWs currently classified in a lower than 50 percent service-connected priority category” (VA, 1980, pp. 163-164).

The second legislative recommendation was to modify the existing statutory presumption of service connection for psychosis. VA proposed to “eliminate the requirement that psychoses suffered by POWs must become manifest within 2 years following service separation before the rebuttable presumption of service connection arises.” VA said that its literature review indicated that psychosis related to the POW experience “frequently appears years after service, not just immediately after separation,” citing NRC/VA follow-up studies published between 1946 and 1980 in support thereof (VA, 1980, p. 164).

VA also reported that it was undertaking several administrative actions as a result of its study. First, forthcoming guidelines on “posttraumatic stress neurosis” would have “explicit reference to former POWs as well as other combat veterans.” (Posttraumatic stress neurosis was a term scheduled on October 1, 1980, to become part of VA’s official diagnostic classification system to describe such anxiety neurosis.) The guidelines would “specifically be used to diagnose, treat, and rate former POWs with anxiety neurosis or similar neurotic disorders.” VA said this change was justified because former POWs had experienced a wide range of psychological problems, and “anxiety neurosis” had been the most prevalent disability of former POWs, according to NRC epidemiologic studies and VA compensation data. VA added that an analysis of anxiety neurosis with length of internment disclosed that it remained a “statistically significant service-connected disability among former POWs regardless of the amount of time in prison camp” (VA, 1980, pp. 165-166; emphasis added).

VA also announced that it would adopt a standardized protocol for disability compensation examinations for all former POWs similar to that developed by the military for former Vietnam

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

POWs. Additional research into POW problems was promised as was the formation of a POW advisory committee (VA, 1980).

Congressional hearings followed the receipt of VA’s study of POWs in both 1980 and 1981. The VSOs concurred with the VA recommendations but criticized them for being inadequate to meet the needs of the POWs and “totally unresponsive to the facts presented” in the report. The VSOs recommended expanding the use of presumptions for POW disability compensation claims (VA, 1993, p. 52).

Legislation was reported in June 1981 and enacted into law on August 14 as the Former Prisoner of War Benefits Act of 1981 (Public Law 97-37. 97th Cong., 1st Sess.). The existing presumptive provisions for POWs were amended to expand the definition of those considered to be prisoners of war and to modify the preconditions necessary for any presumption to attach to a former POW’s disability claim. First, the requirement that POWs had suffered dietary deficiencies, forced labor, or inhumane treatment in violation of the Geneva conventions was removed. Second, the minimum period of confinement was reduced from 6 months to 30 days. Third, the requirement that psychosis manifests itself within 2 years of separation of service in order to qualify for a service-connection presumption was deleted as recommended by VA. Finally, an additional presumption termed “any of the anxiety states” was added to the statute.

The legislative history reveals that removing Geneva Convention violations was intended to “relieve to a certain extent the burden on the former POW to submit evidence … that certain disabilities are service connected” (VA, 1993, p. 54). Reducing confinement time from 6 months to 30 days was justified on the grounds that

Though dietary deficiencies are clearly a function of time and malnourishment, medical evidence reveals that a person can suffer from malnutrition in less than 6 months.

(VA, 1993, p. 54)


The act (Former Prisoner of War Benefits Act of 1981. Public Law 97-37. 97th Cong., 1st Sess.) adopted the VA recommendation for unlimited access to health care for POWs. It also statutorily established an Advisory Committee on Former Prisoners of War which, among other things, was directed to submit an annual report to the VA Secretary containing an assessment of the needs of former POWs and any administrative or legislative recommendations the committee considered to be appropriate. The VA Secretary, in turn, was required to furnish his comments and intentions concerning these recommendations in an annual report to Congress (VA, 1993).

An additional presumption for mental disorders in POWs was added in the mid-1980s, the Veterans’ Compensation and Program Improvements Amendments of 1984 (Public Law 98-223. 98th Cong., 2d Sess.). This legislation added “dysthymic disorder” to the list of disabilities developing anytime after a POW’s separation from service for which a presumption of service connection would attach.

Senator Alan K. Simpson, chairman of the Senate Committee on Veterans Affairs, termed the inclusion a clarification of the original intent of the Former Prisoner of War Benefits Act of 1981. Speaking on the floor of the Senate during consideration of the measure he said that

The complexity of anxiety states, anxiety neuroses, posttraumatic stress disorder, and dysthymic disorders and their associated and sometimes interrelated diagnoses

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

inadvertently resulted in a lack of clarity regarding the granting of service connection for depression.

(VA, 1993, p. 55)


Further explanation was contained in the committee’s report accompanying the measure which recites that at the time the Senate reached agreement with the House on the 1981 act:

[The Senate] was not aware that there would be cases in which former POWS suffering from nonpsychotic depressive disorders would not be diagnosed as suffering from posttraumatic stress disorder and therefore not adjudged under VA guidelines to be service-connected disabled…. The committee intends that this addition would correct the inadvertent oversight in the original legislation and establish a presumption for a mental disorder which is linked in scientific literature to the POW experience.

(VA, 1993, pp. 55-56)


Two years later, the Veterans Benefits Improvement and Health Care Authorization Act of 1986 (Public Law 99-576. 99th Cong., 2d Sess.) added presumptions for “posttraumatic osteoarthritis” and “[o]rganic residuals of frostbite, if the VA Secretary determines that the veteran was detained or interned in climatic conditions consistent with the occurrence of frostbite.” Report language accompanying the legislation said that Congress expected VA “to give great weight to the veteran’s description of the circumstances of frostbite injury and to accept that description if it is possible that these circumstances occurred and the veteran suffers from residuals of frostbite” (U.S. Congress, Senate, Committee on Veterans’ Affairs, 1986, p. 30).

As for traumatic arthritis, the report observed that there was “disagreement as to the adequacy of medical science to distinguish between arthritis resulting from earlier trauma and arthritis which is the result of other causes or which normally occurs during the aging process” (U.S. Congress, Senate, Committee on Veterans’ Affairs, 1986, p. 30). As reported to the Senate floor, the bill simply directed VA to provide a report of the effectiveness of procedures for evaluating such disability claims. During floor consideration, however, the Senate added the specific presumption of posttraumatic osteoarthritis based on unidentified “additional information” received by the committee after it had filed its report.

In a letter to the Chairman of the Senate Committee on Veterans’ Affairs, VA administrator Thomas Turnage questioned the need for these presumptions, saying that if a POW had developed either condition it normally would have “manifested itself and required treatment upon repatriation or shortly thereafter.” But VA did not oppose the provisions, acknowledging that the presumption could serve as a safety net for those veterans whose repatriation processes were insufficient and for those “veterans who had a tendency to initially put aside traumatic experiences associated with captivity” (U.S. Congress, Senate, Committee on Veterans’ Affairs, 1986, pp. 128-129).

Three additional POW presumptions were added with the enactment of the Veterans Benefits and Services Act of 1988 (Public Law 100-322. 100th Cong., 2d Sess.). They were peripheral neuropathy (except where directly related to infectious causes), irritable bowel syndrome, and peptic ulcer disease. Senator Cranston, who sponsored the presumptions, cited a 1986 VA study of former Pacific and European theater POWs as well as a 1985 Australian study which found higher incidence of duodenal ulcers among POWs than controls (VA, 1993, p. 58). The Senate report accompanying the legislation (U.S. Congress, Senate, Committee on Veterans’ Affairs, 1987) found that peripheral neuropathy is causally related to exposure to cold tempera-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

tures, exhausting physical activity, and vitamin deficiency resulting from extreme malnutrition. The report also found that stress and malnutrition were probable risk factors for irritable bowel syndrome and peptic and duodenal ulcers.

VA again took the position that the conditions proposed to be added would ordinarily have become manifest and required treatment upon repatriation or shortly thereafter entitling the veterans to direct service connection without resort to a presumption. VA also maintained that irritable bowel syndrome was not a good candidate for a presumption because it was “a functional disorder of unknown etiology and pathogenesis” (VA, 1993, p. 59).

Finally, the measure as enacted reduced the minimum confinement time to qualify for dental health-care benefits to 30 days.

The 1990s

Although there were no new POW presumptions during this decade, it was an active period for studies dealing with the health effects of POW internment.

Several studies published in 1991 examined chronic depression among former POWs. The first, by Engdahl et al. (1991), found that long-term chronic depressive symptomatology persisted over 40 years and was elevated among POWs of all theaters when compared to control groups. The authors found that age, education, medical symptoms during captivity, and level of social support were related to later levels of adjustment. A second longitudinal study published the same year by Page et al. (1991) elaborated on previous research to show that not only was depressive symptomatology highly elevated in World War II and Korean War POWs, but it was elevated to the point where these populations closely resembled a clinical population of recovering depressives. Two major conclusions from this longitudinal study were that treatment during captivity is statistically linked with depressive symptoms, and that differences in these symptoms were attributable to captivity-related treatment, even when age at capture and education level were considered.

Other variables were considered as well. Conducting a 40-year follow-up of U.S. World War II and Korean War former POWs, Engdahl and Page (1991) measured captivity trauma variables and individual protective variables (age, education, medical symptoms during captivity, social support) to current depressive symptoms. Although depressive symptoms persisted more than 40 years with the knowledge that PTSD and generalized anxiety disorders are known to occur with elevated frequency in POW populations, the degrees of individual protective variables were related to levels of adjustment. This study made a case for the need to examine former POWs that adjusted well in order to understand both the role of specific protective variables and adjustment and resiliency following trauma.

Page’s (1991) work continued with respect to the validity and reliability of some of the work cited above. Despite the heavy reliance on survey data, a noticeable shortage of reports on the effects of nonresponse bias on the measurement of depression existed. Longitudinal data presented opportunities for different types of nonresponse bias, but could also be useful in modeling for bias because of previously collected data. Page found that a predictive model shows nonresponse bias on the reporting of depressive symptoms among former World War II and Korean POWs to be small.

In 1992, the IOM produced a report entitled The Health of Former Prisoners of War (IOM, 1992). This longitudinal study, which focused on morbidity, was initiated in 1986 and built on the earlier work of Cohen and Cooper (1954), Nefzger (1970), Beebe (1975), and Keehn (1980). Veterans were invited to a medical center to undergo the VA protocol exam, which included a

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

comprehensive physical and psychiatric examination. A face-to-face psychiatric interview and a battery of psychological tests were also administered. A caution was provided at the outset of the report that due to low response rates there could be “no confidence … that the group of respondents accurately reflects the composition of all former POWs” (IOM, 1992, p. 4). Nevertheless, the study author noted that the results were presented as descriptive data for two reasons. First, the data, which constituted the largest national collection of POW examinations ever gathered and analyzed, contained a number of findings worthy of note. Some of the findings confirmed earlier studies while other findings were suggestive and served the purpose of generating more definitive research studies.

The second reason to publish the data was not a scientific one, but rather a recognition that

[T]he examination data will have uses beyond the scientific one—for example, in providing material for discussion of military service-connected disabilities among former POWs. Despite the fact that sound inferences about the group of all former POWs cannot be drawn from the exam data in this report, policymakers who must deal with such issues should be able to review this descriptive information…. Not to report the findings of the examinations would surely raise more questions than the report, with its careful documentation of the study’s limitations, would raise. The results of this examination study are thus discussed below, and we urge a maximum of reasonable caution in their interpretation.

(IOM, 1992, p. 5; emphasis added)


The report brought attention to the high prevalence of a number of medical conditions for POWs as compared to controls, especially with regard to psychiatric illness. Prevalence rates for over 20 different medical conditions were discussed, with key results including the following:

  • Percentage of weight loss was associated with higher prevalence of intermittent claudication and arterial vascular disease and lower prevalence of osteoarthritis.

  • Pacific POWs had higher prevalence of PTSD, ulcer, schizophrenia, and generalized anxiety than European and Korean POWs.

  • Vitamin A deficiency and ulcer prevalence was higher for all groups.

  • Edema (also significantly associated with higher prevalence of ischemic heart disease and peripheral nerve disease) and vitamin B1 deficiency was higher for all groups.

  • Visual symptoms were associated with higher prevalence rates of cerebrovascular disease, ulcers, asthma, and PTSD.

  • Neurologic symptoms were associated with beriberi.

  • Korean conflict POWs showed higher prevalence rates for schizophrenia.

  • Vitamin A deficiency helped to explain the prevalence of asthma and cerebrovascular disease in POWs with visual symptoms.

(IOM, 1992, pp. 6-11)


The report also said that many of the organ-specific findings were familiar and that the appreciably increased prevalence of depressive disorders, PTSD, and generalized anxiety were not unexpected. Similar findings regarding peripheral nerve disease, ulcer, and gastroenteritis were also not surprising (IOM, 1992).

The report did state that the noteworthy association between current peripheral nerve disease and earlier edema “suggests … there may be persistent neurologic effects decades after the original nutritional disease may have been successfully treated and acute symptoms have

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

abated.” It added that the increased prevalence of schizophrenia among Korean War theater POWs and an “appreciable correlation with weight loss in this group offers further material for speculation” (IOM, 1992, p. 121).

Perhaps the most striking finding was the association between ischemic heart disease and previous reporting of localized edema. It was observed that there was much interest in heart disease among former POWs, but cautioned that the “lack of a clear biologic mechanism linking nutritional deprivation and subsequent chronic heart disease” together with the study’s low response rate required that one “remain somewhat skeptical of this finding of association” (IOM, 1992, p. 122; emphasis added). Nevertheless, the report stated that

Localized edema is a noteworthy risk factor for only two current medical conditions in these POW examinations—peripheral nerve disease and ischemic heart disease—both of which are acutely related to thiamin deficiency…. The specificity of association between localized edema and the only two medical conditions with well-established acute relationships to thiamin deficiency suggests that the association between earlier nutritional deprivation in prison camp and chronic ischemic heart disease is not an artifact.

(IOM, 1992, p. 122)


To better understand the characteristics that affect POWs’ reintegration into civilian life, Engdahl et al. published a report in 1993 that investigated long-term responses to captivity trauma in former POWs. The authors reported that symptoms at 20 years following release were related to those at 40 years following release. Many factors known to affect POWs’ long-term adjustment were not included in the study (e.g., combat exposure, postwar social support). This was due in part to their statistical infrequency or skewed nature (i.e., family history of mental illness, marital status at capture, military rank at capture). Trauma response was found to be determined by an interaction of characteristics of the individual and characteristics of the trauma, not primarily one over the other. The authors suggested that trauma response, from an evolutionary standpoint, may be better understood as adaptive due to its persistent nature (Engdahl et al., 1993).

Research on POWs then turned to more specific disease outcomes in an attempt to ascertain if a relationship existed between POW status and increased mortality or morbidity. A 1994 study by Page and Ostfeld examined why a specific marker of malnutrition (lower limb edema related to vitamin B deficiency) in former POWs of World War II and the Korean War was associated with a three-fold increase in subsequent death attributed to ischemic heart disease. No medical basis for the link was found, but the confirmed findings emphasized the need for further research into links between severe malnutrition and subsequent chronic disease, both for former POWs and for other severely malnourished populations (Page and Ostfeld, 1994).

Brass and Page (1996) examined health records of World War II POWs to see if severe or chronic stress increases risk of cerebrovascular disease. Compared to a control group, there appeared to be an association between stroke and being a former POW. There was no difference, however, in the prevalence of hypertension or diabetes. The results suggested that further research was needed to better define the risk factors for stroke and further understand the chronic effects of stress.

PTSD continued to gain great attention. Page et al. (1997) examined the literature on PTSD in POWs and compared lifetime PTSD prevalence among POWs and control subjects. After a follow-up period of 40 years, differences in prevalence rates existed between POWs and control subjects for depressive disorders as well as generalized anxiety. All groups of POWs shared

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

nearly the same lifetime and current PTSD rates. Among World War II POWs, however, roughly half of the POWs who once suffered from PTSD were not currently diagnosed with that condition, suggesting the possible presence of chronic, stable PTSD in the other half of the POWs evaluated. However, the authors stated that this may be explained by the symptoms causing less stress on the first group of individuals. The authors suggested that those with higher distress levels should be evaluated for secondary symptoms of PTSD, such as depression. They concluded that

Sensitivity toward older war veterans is vital. An awareness that their PTSD may have gone unnoticed by other health-care professionals for decades should encourage direct clinical inquiries about possible PTSD symptoms. We strongly recommend a structured interview…. PTSD symptoms have been all too common, yet undiagnosed among older war veterans, especially POWs.

(Page et al., 1997, p. 157; emphasis added)

Recent Developments

Work on PTSD continued into the new millennium when World War II and the Korean War POW interviews were examined for two separate index measures at two points in time—1965 and 1990. Results from Gold et al. (2000) supported previous research highlighting the severe psychological consequences of POW status 40-50 years following captivity. Trauma severity during captivity was found to be the best predictor of current PTSD symptomatology.

Other condition-specific research also continued. Page et al. (2000) examined data from 50 years of follow-up for World War II former POWs in Pacific and European theaters to see whether location or POW status could be linked to death attributed to melanoma. Results showed that higher sun exposure in young adulthood (to the extent that POW status related to higher sun exposure) was associated with higher risk of melanoma mortality, with Pacific POWs at highest risk. A 50-year mortality follow-up by Page and Tanner, also in 2000, reported that Pacific theater POWs had twice the rate of death due to Parkinson’s disease and roughly one-sixth the rate of death from motoneuron disease compared to the control group. There was no difference, however, in the death rate due to Parkinson’s or other motoneuron disease in European theater POWs.

Although the first 30 years after repatriation showed a survival advantage of World War II and Korean War former POWs for heart disease and stroke mortality, examination of death data at 50 years showed a significantly higher risk of heart disease deaths and slightly higher stroke mortality in POWs (aged 75 years and older) when compared to control groups. Page and Brass (2001) suggested that circulatory diseases from serious acute malnutrition and stresses from imprisonment may not appear until after many decades.

A report that attracted much attention was a 2000 study of cirrhosis mortality among former World War II and Korean War POWs by Page and Miller. The authors conducted a 50-year follow-up study of World War II and Korean War former POWs to compare increased cirrhosis mortality found after a previous 30-year follow-up period with that of controls. World War II POWs were found to have a 32 percent higher risk of cirrhosis mortality compared to controls, while Korean conflict POWs had the same risk of cirrhosis mortality as controls. Lifetime prevalence rates for chronic liver disease, jaundice, helminthiasis, and nutritional deficiency were higher in Pacific and Korean POWs. But, the authors observed that

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

[T]here were no remarkable differences in levels of self-reported alcohol consumption among our POWs and controls, nor were their self-reported levels of consumption markedly different from those in the general population. The fact that cirrhosis mortality was not apparently associated with alcohol consumption may have implications for American POWs seeking disability compensation for cirrhosis.

(Page and Miller, 2000, p. 783)


The implications suggested by the authors soon became manifest. In April 2001, VA officials had prepared internal estimates of the budget impact of adding cirrhosis of the liver as a presumption of service connection for former POWs. Annual benefit costs were estimated to be $45,828 in 2002 and $624,106 through 2006 for an estimated caseload of “no more than 70 over 5 years” (VA, 2001).

On February 10, 2003, VA issued a proposed rule to add cirrhosis of the liver to the list of diseases for which entitlement to service connection would be presumed for former prisoners of war (VA, 2003). The proposed rule explained that its intended effect was to make it easier for former POWs to obtain compensation for cirrhosis “based on scientific and medical research showing a significantly higher risk of death from cirrhosis in former World War II POWs than in the general population” (VA, 2003, p. 1). Before discussing the rationale for adding cirrhosis of the liver as a presumptive service-connected condition, the notice observed that under 38 CFR § 3.307(d) (Rebuttal of Service Incurrence of Aggravation) such presumptions may be rebutted by “competent evidence,” that is, “affirmative evidence that the disease was not incurred in service based on sound medical reasoning and consideration of all evidence of record” (VA, 2003, p. 2).

In support of the presumption, the proposed rule (VA, 2003) cited the Page and Miller (2000) study that found the risk of cirrhosis mortality for Pacific and European theater former POWs was 1.5 times the rate for the control group. It also cited an Australian study that found World War II POW deaths from cirrhosis at twice the rate as expected. Observing that “it appears that alcohol consumption does not provide an explanation for the higher mortality rates identified in POWs,” the rule said that “[t]he VA Secretary believes that the research cited above constitutes sound scientific evidence supporting the conclusion that an association exists between cirrhosis and POW status” (VA, 2003, p. 3; emphasis added).

Finally, the rule said that inasmuch as World War II POWs “comprise 93 percent” of all living ex-POWs, the “VA Secretary has therefore determined that it is appropriate to add cirrhosis of the liver … for which VA presumes service connection in all former POWs interned or detained for at least 30 days” (VA, 2003, p. 3). The proposed rule received no comments and became final on July 18, 2003.

Later that same year, the Veterans Benefits Act of 2003 (Public Law 108-183. 108th Cong., 1st Sess.) was signed into law. The measure contained two POW-specific provisions. First, it codified VA’s administrative decision concerning cirrhosis of the liver. Second, it removed the 30-day minimum confinement requirement for 5 of the 16 POW presumptive conditions. Those 5, for which no minimum confinement was required, were

(a) Psychosis; (b) Any of the anxiety states; (c) Dysthymic disorder (or depressive neurosis); (d) Organic residuals of frostbite, if the VA Secretary determines that the veteran was detained or interned in climate conditions consistent with the occurrences of frostbite; and (e) Posttraumatic osteoarthritis.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

In justifying the change, a Senate committee report on a similar bill observed that POWs were often treated brutally and, even if treated humanely, often suffered extreme mental anguish. Thus, the “30-day minimum internment requirement for purposes of presumptive service connection may be too restrictive for certain conditions” (U.S. Congress, Senate, Committee on Veterans’ Affairs, 2003, p. 10). In this connection, it should be noted that there was considerable public attention paid to Service members who were held as POWs for short periods of time shortly following the inception of Operation Iraqi Freedom in March 2003.

Perhaps the most significant development concerning POW presumptions occurred on October 7, 2004, when VA issued an Interim Final Rule entitled Presumptions of Service Connection for Disease Associated with Service Involving Detention or Internment as a Prisoner of War (VA, 2004a). The rule established guidelines for establishing presumptions of service connection for diseases associated with service involving detention or internment as a prisoner of war. VA justified the rules as necessary because POW claims present unique medical issues and because factors including the lack of contemporaneous medical records during periods of captivity and the “relatively small body of available medical information” present obstacles to substantiating claims for service-connected benefits based on POW service.

The guidelines were intended to help VA ensure that the claims are “decided fairly, consistently, and based on all available medical information concerning the diseases associated with detention or internment as a prisoner of war” (VA, 2004a, p. 60083). Finally, in utilizing the new guidelines, VA’s interim final rule also established presumptions of service connection for atherosclerotic and hypertensive heart disease and for stroke disease arising in former prisoners of war.

The new guidelines found at 38 CFR 1.18 were as follows:

§ 1.18—Guidelines for establishing presumptions of service connection for former prisoners of war.

  1. Purpose. The Secretary of Veterans Affairs will establish presumptions of service connection for former prisoners of war when necessary to prevent denials of benefits in significant numbers of meritorious claims.

  2. Standard. The Secretary may establish a presumption of service connection for a disease when the Secretary finds that there is at least limited/suggestive evidence that an increased risk of such disease is associated with service involving detention or internment as a prisoner of war and an association between such detention or internment and the disease is biologically plausible.

    1. Definition. The phrase “limited/suggestive evidence” refers to evidence of a sound scientific or medical nature that is reasonably suggestive of an association between prisoner-of-war experience and the disease, even though the evidence may be limited because matters such as chance, bias, and confounding could not be ruled out with confidence or because the relatively small size of the affected population restricts the data available for study.

    2. Examples. “Limited/suggestive evidence” may be found where one high-quality study detects a statistically significant association between the prisoner-of-war experience and disease, even though other studies may be inconclusive. It also may be satisfied where several smaller studies detect an association that is consistent in magnitude and direction. These examples are not exhaustive.

  1. Duration of Detention or Internment. In establishing a presumption of service connection under paragraph (b) of this section, the Secretary may, based on sound scien-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

tific or medical evidence, specify a minimum duration of detention or internment necessary for application of the presumption.

  1. Association. The requirement in paragraph (b) of this section that an increased risk of disease be “associated” with prisoner-of-war service may be satisfied by evidence that demonstrates either a statistical association or a causal association.

  2. Evidence. In making determinations under paragraph (b) of this section, the Secretary will consider, to the extent feasible:

    1. Evidence regarding the increased incidence of disease in former prisoners of war;

    2. Evidence regarding the health effects of circumstances or hardships similar to those experienced by prisoners of war (such as malnutrition, torture, physical abuse, or psychological stress);

    3. Evidence regarding the duration of exposure to circumstances or hardships experienced by prisoners of war that is associated with particular health effects; and

    4. Any other sound scientific or medical evidence the Secretary considers relevant.

  1. Evaluation of studies. In evaluating any study for the purposes of this section, the Secretary will consider:

    1. The degree to which the study’s findings are statistically significant;

    2. The degree to which any conclusions drawn from the study data have withstood peer review;

    3. Whether the methodology used to obtain the data can be replicated;

    4. The degree to which the data may be affected by chance, bias, or confounding factors; and

    5. The degree to which the data may be relevant to the experience of prisoners of war in view of similarities or differences in the circumstances of the study population.

  1. Contracts for Scientific Review and Analysis. To assist in making determinations under this section, the Secretary may contract with an appropriate expert body to review and summarize the scientific evidence, and assess the strength thereof, concerning the association between detention or internment as a prisoner of war and the occurrence of any disease, or for any other purpose relevant to the Secretary’s determinations.

The interim final rule contained an extended discussion and explication of these guidelines. It first noted that statutory and regulatory standards currently existed to guide VA in identifying diseases associated with exposure to herbicide agents, hazards of service in the Gulf War, and ionizing radiation, and that it would be helpful to establish standards to guide VA in identifying diseases associated with service involving detention or internment as a POW. The POW guidelines were characterized as substantially similar to the existing guidelines noted above, with “minor differences necessary to reflect considerations unique to former POWs” (VA, 2004a, p. 60084).

The rule noted that evidentiary presumptions were to serve a number of purposes, including the promotion of efficient resolution of service-connection issues by codifying medical findings and principles that otherwise may not be familiar to VA adjudicators. Presumptions promote “fair and consistent decision making by establishing simple adjudicatory rules” as well as assisting claimants who might face substantial difficulties “due to the complexity of the factual issues, the lack of contemporaneous medical records during service, or other circumstances” (VA, 2004a, p. 60084; emphasis added).

Acknowledging that POW experiences have varied with time, place, and other factors, the rule stated that “certain hardships are so prevalent across the spectrum of POW experience as to support the presumption that POWs as a group have incurred similar health risks.” Evidentiary

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

presumptions were also strongly justified by the absence of “contemporaneous personnel and health records to document events, injuries, or diseases during periods of captivity.” Moreover, presumptions simplify and expedite the claims adjudication process, a “particularly significant consideration for former POWs, more than 90 percent of whom served in World War II and are now, on average, over 80 years old” (VA, 2004a, p. 60084).

VA acknowledged that determining whether health effects may be associated with POW experience is “not a simple task.” The effects of the POW experience have been less extensively studied, because “there generally are not comparable civilian populations, and the number of former POWs available for study is comparatively small.” Accordingly, VA announced that it intended to establish presumptions of service connection when the medical evidence reasonably establishes an “association between the POW experience and particular diseases” (VA, 2004a, p. 60084; emphasis added).

Perhaps the most significant feature of the new POW guidelines was the explicit decision to establish POW presumptions utilizing the “limited/suggestive evidence of an association” classification. VA said this was essentially the same standard employed by IOM in reports it prepared for VA (IOM, 1994, 1996, 1999, 2000, 2001, 2003, 2005) analyzing the health effects of exposure to herbicide agents. “Limited/suggestive” evidence is one of four classifications employed by IOM in its Agent Orange reports and was a standard that had been utilized by VA in establishing presumptions of service connection. VA characterized the “limited/suggestive” evidence standard as a “useful analytical framework for assessing scientific evidence and determining whether a presumption of service connection may be warranted” (VA, 2004a, p. 60085).

IOM’s use of the term “limited/suggestive evidence of an association” was explained as referring to circumstances in which evidence is suggestive of an association but is limited because matters of chance, bias, and confounding cannot be ruled out with confidence (IOM, 1994, 1996, 1999, 2001, 2003, 2005). VA said its definition for POWs “adds that the evidence may be limited because the relatively small size of the affected population may restrict the data available for study.” This addition was significant, VA said, given the circumstances of the POW experience and the fact that “the population of surviving former POWs, most of whom served in World War II, is declining rapidly” (VA, 2004a, p. 60085). In the rule,VA added

Although we intend that any presumptions VA establishes will be based on sound scientific and medical evidence, we believe … that fairness to former POWs requires that VA fully evaluate the available data and not accord undue significance to the fact that such data are comparatively limited by the small size of the affected population.

(VA, 2004a, p. 60085)


Finally, the rule stated that

The requirement that the association be biologically “plausible” does not require proof of a casual relationship … it requires only a determination that there is a possible biologic mechanism, consistent with sound scientific evidence, by which the suspected precipitating event (POW experience) could lead to the health outcome. IOM routinely applies the concept of biologic plausibility in its reviews of the literature concerning the health effects of herbicide exposure and hazards of Gulf War Service….

(VA, 2004a, p. 60085)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Having established the guidelines in the October 7, 2004, rule, VA then applied them in establishing new presumptions of service connection for (1) “stroke and its complications,” and (2) “atherosclerotic heart disease or hypertensive vascular disease (including hypertensive heart disease) and their complications (including myocardial infarction, congestive heart failure, arrhythmia” (VA, 2004a, p. 60090).

With respect to the first presumption, VA admitted that there were very few studies investigating the possible relationship between POW experience and stroke. It referenced a 1996 Brass and Page study that found a seven-fold increase in the incidence of stroke among the POWs as compared to the control group. VA noted that the strength of those findings was limited by the small size of the study population. It also referenced a 2001 Brass and Page study that found a statistically significant increase in death from stroke among veterans who had experienced visual symptoms, such as night blindness, during their captivity. Because the presence of visual symptoms during captivity may be associated with vitamin A deficiency, this finding was regarded as consistent with the earlier study in suggesting an association between malnutrition during POW captivity and subsequent stroke (VA, 2004a).

On the recommendation of the Expert Panel on Strokes in Former Prisoners of War, VA’s Environmental Epidemiology Service conducted a study in 2003 using medical and death data from VA and the Health Care Financing Administration (HCFA) records (Kang and Bullman, 2003; as referenced in VA, 2004a). That study, which had not been published at the time the rule was issued, found that POWs had a significantly higher incidence of PTSD than the controls and further that POWs with PTSD had a higher incidence of stroke than POWs without PTSD. Although the study did not find a significantly increased risk of stroke among POWs as compared to non-POWs, the evidence for an association between PTSD and stroke among POWs was deemed consistent with findings stated in the 1996 Brass and Page study.

Observing that several studies had provided evidence suggesting an association between stress and stroke, the rule reiterated that the VA Secretary would consider evidence concerning the effects of circumstances or hardships similar to those experienced by POWs, including stress, in assessing the evidence for establishing presumptions of service connection.

Accordingly, based on the evidence discussed above, the VA Secretary determined that a presumption of service connection was warranted for stroke among former prisoners of war. The Brass and Page (1996) and Page and Brass (2001) POW studies both found an increased risk of stroke among former POWs. Although there was an absence of other directly corroborating studies, the lack of additional data was due in part to the small size of the POW population available for study and the limited number of studies generally undertaken in this field. Under those circumstances, VA concluded the lack of corroborating data did not imply the absence of an association (VA, 2004a).

Because “VA consider[ed] stress and malnutrition to be among the hardships ordinarily associated with POW experience” and because “evidence suggest[ed] that the risk of stroke increases with the severity of those hardships” the VA Secretary found that

[T]he available evidence is suggestive of an association between POW experience and stroke because sound scientific studies provide evidence of an association that is consistent in magnitude and direction, even though it is limited in some respects by the small size of the affected population and the correspondingly limited data available for study.

(VA, 2004a, p. 60086)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The VA Secretary further determined that an association between stroke and POW experience was biologically plausible, and for those reasons VA was establishing a presumption of service connection for stroke in former POWs.

Turning to the second presumption that was administratively added in 2004, VA again acknowledged that there were relatively few studies addressing the association between POW experience and heart disease (VA, 2004a). A series of older studies did not find consistent evidence of an association, as summarized in Page and Ostfeld’s 1994 study titled Malnutrition and Subsequent Ischemic Heart Disease in Former Prisoners of War of World War II and the Korean Conflict.

VA stated that more recent studies had yielded “intriguing findings concerning the association between heart disease and POW experience.” The 1994 study by Page and Ostfeld found a statistically significant increase in deaths due to ischemic heart disease among former POWs who experienced edema in their lower limbs during captivity. As previously discussed, the authors theorized that the findings might suggest an association between malnutrition during captivity and subsequent ischemic heart disease. Current VA regulations provided for presumptive service connection of ischemic heart disease in former POWs who experienced localized edema during captivity (VA, 2004a).

The 2001 study by Page and Brass found a trend of increased excess risk of heart disease with advanced age, with a statistically significant increased risk for former POWs aged 75 years or older. The authors suggested that the findings might indicate that the sequelae of serious, acute malnutrition may not appear until after many decades.

The 2003 VA study (Kang and Bullman, 2003; as referenced in VA, 2004a) analyzed records of inpatient and outpatient treatment from VA and HCFA to determine whether POWs had an increased incidence of certain diseases in comparison to the non-POW controls. The study “detected small increases in the incidence of hypertension and myocardial infarction among some, but not all, of the subpopulations examined, and not all of the findings were statistically significant. However, the study did find a statistically significant increased incidence of hypertension and chronic heart disease among World War II veterans with PTSD” (VA, 2004a, p. 60087). The conclusion that PTSD may be associated with cardiovascular disorders was also supported by a 1997 study (Boscarino, 1997) finding that Vietnam veterans diagnosed with PTSD had a significantly increased risk of circulatory disease many years after service.

Based on the evidence discussed above, the VA Secretary determined that a presumption of service connection was warranted for atherosclerotic heart disease and hypertensive vascular disease among former POWs. As in the instance of stroke, the VA Secretary concluded that the evidence suggesting an association between heart disease and specific hardships of POW experience—malnutrition and stress—was significant. Notwithstanding limited data, the VA Secretary concluded that sound scientific studies provided “limited/suggestive” evidence of an association between POW experience and heart disease, and that it was “biologically plausible” (VA, 2004a).

VA conceded that not all of the studies cited investigated the same range of heart diseases, and thus they did not clearly resolve the question of which types of heart disease may be associated with POW experience. For purposes of the presumption, however, all cardiovascular diseases were included by VA that were deemed consistent, in terms of biologic plausibility, with those findings because the diseases were potentially capable of being caused by the stress or malnutrition hardships of POW service. The presumption did not extend to diseases that arise from viral or bacterial causes, because the evidence concerning biologic plausibility did not support a finding that such heart diseases were associated with POW experience (VA, 2004a).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

With respect to certain types of atherosclerotic heart disease or hypertensive vascular disease that were covered by these presumptions, VA acknowledged little available evidence upon which to rule in or rule out the possibility that the condition is capable of being caused by the hardships of POW service. In those cases, VA chose “to resolve the doubt in favor of veterans and include the condition within the scope of the presumption,” saying that

Although the necessity of inclusion of some conditions may be uncertain from a purely scientific perspective, VA has decided as a policy matter to resolve this issue in favor of veterans because there is a reasonable basis for doing so. Presumptions of service connection for former POWs can be rebutted as provided in 38 U.S.C. 1113(a) and 38 CFR 3.307(d). Accordingly, if evidence in a case supports a finding that a particular presumptive condition was not actually caused by a veteran’s POW experience, VA may consider the presumption to be rebutted.

(VA, 2004a, p. 60087; emphasis added)


Finally, no minimum confinement time was required for either of the new presumptions:

Because the evidence indicates that heart disease and stroke potentially may be associated either with malnutrition during prolonged captivity or with stress due to circumstances such as torture or abuse, which may occur during even brief periods of captivity, we do not believe a minimum period of detention or internment is warranted for these presumptions.

(VA, 2004a, p. 60088)


The VA Secretary said that the rule was issued as an interim final rule without providing an opportunity for prior comment because he believed the rule was unlikely to generate any adverse public comment, “inasmuch as it confers a benefit on a deserving class of veterans based on sound scientific evidence.” Moreover, it was “impracticable to delay” inasmuch as

… the class of veterans affected by this rule is elderly and rapidly dwindling. Delay in implementing these rules would have a significant adverse effect and frustrate the beneficial purpose of this rule in view of the high mortality rate among the POW population and the fact that the majority of former POWs are at an age where their medical and financial needs are likely to be at their greatest.

(VA, 2004a, p. 60088)


Only one written comment was received, and it primarily sought to have the presumptions further expanded. The interim rule, without changes, became final on June 28, 2005.

Given the significance of VA guidelines for establishing presumptions of service connection for former POWs, the Committee sought material from VA concerning any information that may have been considered by VA in developing the guidelines and the two presumptions that was not referenced or discussed in the Federal Register notice. It also sought to ascertain what role, if any, the Advisory Committee on Former Prisoners of War had in the development of the rule. VA chose not to respond and stated in a letter to the Committee chair that statements made in the “context of robust deliberations” might be “misconstrued or misrepresented” and that to do so would “inhibit free discussion in future deliberations” (Dunne, 2006).

VA estimated that the new presumptions would have a first-year cost of $26 million increasing to $33 million in the fourth year and then gradually declining to $21 million in the

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

tenth year due to increasing POW mortality. The 5- and 10-year cumulative costs were $152.3 and $279.5 million, respectively (VA, 2004b, p. 4).

No additional POW presumptions have been added since 2005 either administratively or statutorily.

Lessons Learned

Americans and their elected representatives have long been concerned with the welfare of those who protected, defended, and sacrificed for their country. The extensive system of veteran benefits and their liberalized rules for qualification is a manifestation of this concern.

This concern is intensified if those veterans seeking assistance are viewed as having been subjected to extraordinary stresses and sacrifices. POWs are such a group. As VA declared in its 1980 study, the POW experience was an “extremely harsh and brutal experience” that was “characterized by starvation diet, poor quality or nonexistent medical care, ‘death marches,’ executions, and torture” (VA, 1980, p. 4).

In this context, the creation of certain presumptions with respect to disabilities claimed to be connected with a veteran’s experience as a POW is a natural reflection of a long-established concern for their welfare. These presumptions simplify adjudication in what otherwise would be difficult cases to obtain evidence and resolve complex issues, thus relieving the burden on both the veteran and VA. Presumptions have been particularly helpful in assessing POW claims where information about individual conditions of internment and complete medical records were frequently unavailable.

Finally, presumptions have enabled greater consistency in decision making, the absence of which generated much discontent by veterans who strongly communicated their concerns to their elected representatives. Indeed, it might reasonably be inferred that the subsequent modification of certain presumptions by reducing or eliminating preconditions for their applicability has been a further policy response to concerns registered by those former POWs who were not previously covered by presumptions.

These policy considerations were obviously important factors that were weighed by Congress as it considered available epidemiologic and other scientific findings derived from various POW studies. These studies (Beebe, 1975; Brass and Page, 1996; Cohen and Cooper, 1954; IOM, 1992; Keehn, 1980; Nefzger, 1970; Page and Brass, 2001; Page et al., 1991, 1997; VA, 1980) dating back to the early 1950s reveal a slow accretion of data and increased medical knowledge, particularly about malnutrition, stress, and the psychological effects of the POW experience.

At the same time, available evidence for certain presumptions sometimes came from only a single study, and the data were often limited by the small size of the affected number of POWs. Consequently, authors of most studies acknowledged the tentative and inconclusive nature of their findings and urged caution at drawing unwarranted inferences. As one study author stated, the mixture and interdependence of various factors of the POW experience and the variation of their relative intensities “from time to time, from place to place, and from man to man,” has “limit[ed] the scope and specificity of the inferences that may be drawn statistically” (Nefzger, 1970, p. 124).

Given the suggestive but scientifically uncertain results of many of the studies, it is not surprising that policy makers frequently decided to create service-connected presumptions when faced with the pressing claims of genuinely sick and disabled former POWs. As one author observed of the 1992 study of POWs, the descriptive data obtained had “uses beyond the scien-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

tific,” specifically in the discussion of military service-connected disabilities. The author added, “[d]espite the fact that sound inferences about the group of all former POWs cannot be drawn from the exam data in this report, policymakers who must deal with such issues should be able to review this descriptive information” (IOM, 1992, p. 5).

It should also be noted that VA in its 2004 guidelines for POW presumptions, while expressing an intent to base its determinations on “sound scientific and medical evidence,” has adopted a standard of “limited/suggestive evidence” for former POWs and added the additional caveat that “fairness to former POWs requires VA to fully evaluate the available data and not accord undue significance to the fact that such data are comparatively limited by the small size of the affected population” (VA, 2004a, p. 60085). Given the forgoing, it is apparent that although various POW studies have often provided useful scientific information to Congress and the executive branch, there were other policy-relevant factors that had equal or greater weight in the presumptive decisions established for POWs. Over time, the requirement that a presumptive decision be based on sound scientific and medical evidence has increasingly been overshadowed by these other considerations.

References

Bard, M. G. 1994. Forgotten victims: The abandonment of Americans in Hitler’s camps. Boulder, CO: Westview Press.

Beebe, G. W. 1975. Follow-up studies of World War II and Korean War prisoners. II. Morbidity, disability, and maladjustments. American Journal of Epidemiology 101(5):400-422.

Boscarino, J. A. 1997. Diseases among men 20 years after exposure to severe stress: Implications for clinical research and medical care. Psychosomatic Medicine 59(6):605-614.

Brass, L. M., and W. F. Page. 1996. Stroke in former prisoners of war. Journal of Stroke and Cerebrovascular Diseases 6(2):72-78.

Cohen, B. M., and M. Z. Cooper. 1954. A follow-up study of World War II prisoners of war. Washington, DC: Government Printing Office.

Dunne, P. W. 2006 (December). Department of Veterans Affairs responses: Questions following from the Veteran’s Disability Benefits Commission’s charge. Letter to Jonathan M. Samet.

Engdahl, B. E., and W. F. Page. 1991. Psychological effects of military captivity. In Epidemiology in military and veteran populations. Proceedings of the Second Biennial Conference. Washington, DC: National Academy Press. Pp. 49-66.

Engdahl, B. E., W. F. Page, and T. W. Miller. 1991. Age, education, maltreatment, and social support as predictors of chronic depression in former prisoners of war. Social Psychiatry and Psychiatric Epidemiology 26(2):63-67.

Engdahl, B. E., A. R. Harkness, R. E. Eberly, W. F. Page, and J. Bielinski. 1993. Structural models of captivity trauma, resilience, and trauma response among former prisoners of war 20 to 40 years after release. Social Psychiatry and Psychiatric Epidemiology 28(3):109-115.

Gold, P. B., B. E. Engdahl, R. E. Eberly, R. J. Blake, W. F. Page, and B. C. Frueh. 2000. Trauma exposure, resilience, social support, and PTSD construct validity among former prisoners of war. Social Psychiatry and Psychiatric Epidemiology 35(1):36-42.

HEW (U.S. Department of Health Education and Welfare). 1956. Effects of malnutrition and other hardships on the mortality and morbidity of former United States prisoners of war and civilian internees of World War II: An appraisal of current information. H.R. Doc. No. 296. 84th Cong., 2nd Sess.

IOM (Institute of Medicine). 1992. The health of former prisoners of war. Washington, DC: National Academy Press.

IOM. 1994. Veterans and Agent Orange: Health effects of herbicides used in Vietnam. Washington, DC: National Academy Press.

IOM. 1996. Veterans and Agent Orange: Update 1996. Washington, DC: National Academy Press.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

IOM. 1999. Veterans and Agent Orange: Update 1998. Washington, DC: National Academy Press.

IOM. 2000. Veterans and Agent Orange: Herbicide/dioxin exposure and type 2 diabetes. Washington, DC: National Academy Press.

IOM. 2001. Veterans and Agent Orange: Update 2000. Washington, DC: National Academy Press.

IOM. 2003. Veterans and Agent Orange: Update 2002. Washington, DC: The National Academies Press.

IOM. 2005. Veterans and Agent Orange: Update 2004. Washington, DC: The National Academies Press.

Kang, H. K., and T. A. Bullman. 2003 (unpublished). Ten year mortality and morbidity follow-up of former World War II and Korean War prisoners of war.

Keehn, R. J. 1980. Follow-up studies of World War II and Korean Conflict prisoners. III. Mortality to 1 January 1976. American Journal of Epidemiology 111(2):194-211.

Nefzger, M. D. 1970. Follow-up studies of World War II and Korean war prisoners. I. Study plan and mortality findings. American Journal of Epidemiology 91(2):123-138.

Page, W. F. 1991. Using longitudinal data to estimate non-response bias. Social Psychiatry and Psychiatric Epidemiology 26:127-131.

Page, W. F., and L. M. Brass. 2001. Long-term heart disease and stroke mortality among former American prisoners of war of World War II and the Korean Conflict: Results of a 50-year follow-up. Military Medicine 166(9):803-808.

Page, W. F., and R. N. Miller. 2000. Cirrhosis mortality among former American prisoners of war of World War II and the Korean Conflict: Results of a 50-year follow-up. Military Medicine 165(10):781-785.

Page, W. F., and A. M. Ostfeld. 1994. Malnutrition and subsequent ischemic heart disease in former prisoners of World War II and the Korean Conflict. Journal of Clinical Epidemiology 47(12):1437-1441.

Page, W. F., and C. M. Tanner. 2000. Parkinson’s disease and motor-neuron disease in former prisoners of war (Letter). Lancet 355(9206):843.

Page, W. F., B. E. Engdahl, and R. E. Eberly. 1991. Prevalence and correlates of depressive symptoms among former prisoners of war. Journal of Nervous and Mental Disease 179(11):670-677.

Page, W. F., B. E. Engdahl, R. E. Eberly, C. S. Fullerton, and R. J. Ursano. 1997. Persistence of PTSD in former prisoners of war. In Posttraumatic stress disorder: Acute and long-term responses to trauma and disaster, edited by C. S. Fullerton and R. J. Ursano. Washington, DC: American Psychiatric Press. Pp. 147-158.

Page, W. F., D. Whiteman, and M. Murphy. 2000. A comparison of melanoma mortality among WWII veterans of the Pacific and European theaters. Annals of Epidemiology 10(3):192-195.

President’s Commission on Veterans’ Pensions. 1956. A report to the president: Veterans benefits in the United States, Parts I and II: Findings and recommendations. Washington, DC: Government Printing Office.

U.S. Congress, Senate, Committee on Finance. 1970. Increase in veterans disability compensation. Senate Report Number 91-784. 91st Cong., 2nd Sess. April 23.

U.S. Congress, Senate, Committee on Veterans’ Affairs. 1986. Omnibus Veterans’ Benefits Improvement and Health Care Authorization Act of 1986. Senate Report Number 99-444. 99th Cong., 2nd Sess. September 16.

U.S. Congress, Senate, Committee on Veterans’ Affairs. 1987. Omnibus Veterans’ Benefits and Services Act of 1987. Senate Report Number 100-215. 100th Cong., 1st Sess. November 6.

U.S. Congress, Senate, Committee on Veterans’ Affairs. 2003. Veterans’ Benefits Enhancements Act. Senate Report Number 108-169. 108th Cong., 1st Sess. October 21.

VA (Department of Veterans Affairs). 1980. Study of former prisoners of war. Washington, DC: Government Printing Office.

VA. 1993. Analysis of presumptions of service connection. Washington, DC: Department of Veterans Affairs.

VA. 2001. Costing of regulation RIN 2900-AK83—Presumption of service connection for cirrhosis of the liver in former prisoners of war.

VA. 2003. Presumption of service connection for cirrhosis of the liver in former prisoners of war. Proposed Rule. February 10, 2003. Federal Register 68(27):6679.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

VA. 2004a. Presumptions of service connection for disease associated with service involving detention or internment as a prisoner of war. Interim Final Rule. October 7, 2004. Federal Register 69(194):60083-60090.

VA. 2004b. Additional presumptions for former POWs: Methodology (cost estimates). RIN 2900-AM09. 108th Cong., 2nd Sess.

VA. 2005. Former American prisoners of war. Washington, DC: Department of Veterans Affairs.

VA. 2006. Fact sheet: VA benefits for former prisoners of war. http://www1.va.gov/opa/fact/powmia.asp (accessed March 23, 2007).

CASE STUDY 4:
AMPUTEES AND CARDIOVASCULAR DISEASE PRESUMPTION

This case study examines the 1979 presumption of service connection for cardiovascular disease (CVD) that develops in veterans with certain types of service-related amputations. By exploring the scientific basis for this decision, as well as the implications of this decision for the VA, this analysis seeks to illustrate the challenges and implications of establishing a service connection for a common chronic condition with multiple contributing causal risk factors.

Background on Cardiovascular Disease

This presumption establishes service connection for CVD in veterans with certain types of lower extremity amputations. The literature on which this presumption is based examines the broad category of CVD, with specific attention to atherosclerotic (or ischemic) heart disease, the most common of the cardiovascular conditions. Two features of CVD in general and atherosclerotic disease in particular are important for evaluating the scientific evidence and events surrounding this presumption. First, CVDs are common in the general population, and their prevalence increases with advancing age (Figure I-3).

The high prevalence of CVD among the general adult population may pose challenges for determining whether, among veterans, CVD can be attributed to military service. The high prevalence may also translate into a large constituency advocating for a presumption, as well as determine the cost implications of such a presumption for VA. Second, there are multiple established risk factors for CVD, particularly atherosclerotic disease, that are common in the general population. Many of these are lifestyle factors, including smoking, obesity, and physical inactivity. Determining the relationship between risk factors that develop during military service and the additional risk factors that may develop after service is challenging, as is estimating the incremental risk of CVD attributable to military service.

The Need for a Presumption

Veterans with amputations resulting from trauma sustained during military service automatically receive compensation for this service-related condition and may also receive compensation for amputation-related complications. The presumption service connecting CVD among veterans with service-connected amputations (38 CFR § 3.310[b], Cardiovascular disease) was created after a comprehensive review of the prior literature and analysis of data from World War II veterans suggested a link might exist between certain types of amputation and CVD. In contrast with many presumptions that exist because of an evidence gap in exposure assessment (e.g., presumptions related to Agent Orange), this presumption was created because of gaps in the evidence linking amputations to CVD.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

FIGURE I-3 Prevalence of cardiovascular disease by age and sex.

SOURCE: Thom et al., 2006, p. 91.

Scientific and Legislative History of the Presumption for Cardiovascular Disease Among Amputees

The following is a brief review of the scientific and legislative events of relevance to this presumption.

Prior to 1976

In a study conducted by MFUA/NRC, Hrubec and Ryder (1979) reviewed the scientific literature published prior to 1976 on the relationship between amputation and mortality from CVD, the findings of which appeared to be mixed. The evidence for higher cardiovascular mortality rates stemmed primarily from one large Finnish study (Bakalim, 1969) of 4,738 World War II amputees that reported 63.1 percent greater cardiovascular death rates among amputees than among the general population (Hrubec and Ryder, 1979, p. 31). In their review, Hrubec and Ryder noted two major limitations of this study. The first limitation was that mortality among amputees participating in the study was compared with that of the general population, rather than veterans who had also sustained combat-related injuries. The second limitation was that the time period of comparison for the study group included the acute hospitalization for the amputation, a period in which risk of mortality may be higher.

Hrubec and Ryder noted several other studies conducted prior to 1976 that found no increased mortality among amputees. The Advisory Committee on Cardiovascular Disorders and Mortality Rate in Amputees (1954), for example, compared incidence of cardiovascular disorders among 27,000 British amputees from the 1914 war with that of both the general population and veterans with wounds to the lower extremities (not necessarily amputations). They concluded that limb amputation does not appear to “initiate or aggravate cardiovascular disorders to any significant extent” (Hrubec and Ryder, 1979, pp. 30-31).

In an effort to clarify the possible association between amputations and cardiovascular disorders, the Veterans Disability Compensation and Survivor Benefits Act of 1976 (Public Law 94-433. 94th Cong., 2d Sess.) mandated that, in addition to reviewing the existing literature on this topic, “an analysis of statistically valid samples of disability claims of veterans having service-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

connected extremity amputation matched by age, sex and war period with nonamputee veterans” be conducted.

The 1979 Medical Follow-Up Agency Report on Service-Connected Traumatic Limb Amputations and Subsequent Mortality from Cardiovascular Diseases

In 1979 Hrubec and Ryder delivered their report conducted by MFUA/NRC as a part of its contract with VA. MFUA had access to summary information from all Army hospitalizations from 1944 and 1945 and used this information to identify three comparison groups based on the type of injury—3,890 Service members with proximal limb amputations (at or above the knee or elbow), 2,918 with distal amputations (loss of part of foot or hand), and 3,890 with disfigurement (disfigurement other than head, face, or skull, and disfiguring scars). The authors compared the observed death rates in each of these three groups to the expected rates based on age and time-specific rates for U.S. males in the general population. Veterans in the three comparison groups were followed from January 1946 to April 1977, and their mortality from a variety of different causes was evaluated.

During more than 30 years of follow-up, 922 proximal amputees died; 714.1 deaths would have been expected based on the general U.S. male death rate (Hrubec and Ryder, 1979, p. 47). Compared with distal amputees and those with disfigurement, proximal amputees had a higher risk of all-cause mortality and mortality related to diabetes and CVD, particularly atherosclerotic (ischemic) heart disease. Mortality from CVD was similar between proximal amputees and the general population during the period immediately following amputation, but increased dramatically among the proximal amputees over later time periods. This greater than expected cardiovascular mortality among the proximal amputees was not affected by their age at amputation. The same pattern of increasing cardiovascular mortality was not observed among those with distal amputations or disfigurement; they continued to have the same or lower than expected death rates. In matched pair comparisons with disfigured veterans, the highest risk of CVD was observed among those with bilateral lower extremity amputations of any degree and unilateral amputations at the knee or above (Hrubec and Ryder, 1979).

Hrubec and Ryder noted that strict time constraints imposed by Congress prevented them from conducting detailed individual analyses to explore the potential mediators of the increased risk of cardiovascular mortality among proximal amputees. Their discussion on this topic included a brief review of the existing literature on other cardiovascular risk factors, and they proposed the following potential explanations for the association: (1) CVD may be a risk factor for amputation, as Service members with CVD at the time of injury may have been more likely to undergo amputation (versus procedures to salvage the limb) because of concerns for peri- and postoperative risk, (2) amputation may itself be a direct risk factor for CVD, and (3) amputation may be a risk factor for other factors (e.g., physiologic factors such as hypertension, as well as lifestyle factors such as inactivity, psychosocial stress, and smoking) known to place individuals at greater risk for CVD.

The authors explored wound infection as a potential mediator and found no increased risk of CVD by the presence of infection at the time of amputation. They also commented that smoking was unlikely to mediate the increased risk since the rates of lung cancer mortality are not higher among proximal amputees (although it is intriguing that increased risk of buccal cavity and pharynx cancer among proximal amputees is noted in this study but not commented on further).

Hrubec and Ryder concluded from their analysis that proximal traumatic amputation of the lower limbs was a risk for subsequent cardiovascular mortality and commented that, based on

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

their review of the existing risk factor literature, “the most likely factors of importance (for mediating this increased risk) include increased sedentary lifestyle among amputees and increased emotional stress caused by amputation” (Hrubec and Ryder, 1979, p. 43).

The results of this MFUA study led to 38 CFR § 3.310(b) (Cardiovascular disease) granting a service connection “for ischemic heart disease or other cardiovascular disease developing in a veteran who has a service-connected amputation of one lower extremity at or above the knee or service-connected amputations of both lower extremities at or above the ankles.”

The Scientific Literature Since the 1979 Presumption

The literature on CVD and amputation since the 1979 Hrubec and Ryder study is limited and has focused on exploring the mechanisms whereby amputation might lead to cardiovascular morbidity and mortality. A small study (Rose et al., 1986) of 31 Vietnam veterans comparing veterans with proximal lower extremity amputations to those with upper extremity amputations noted higher blood pressure, higher body mass index, and impaired glucose metabolism among the lower extremity amputees. Another study (Modan et al., 1998) followed 201 veterans of the Israeli army with proximal amputations for over 24 years and compared their risk factor profile and mortality with 1,832 controls (matched based on age and ethnic origin) from the general population. The authors found an increase in the risk of cardiovascular mortality among traumatic lower limb amputees. Surviving amputees did not have higher rates of obesity, physical inactivity, or hypertension compared with controls from the general population, but were more likely to have hyperinsulinemia and be in a hypercoagulable state. The authors suggested that hypercoagulability and hyperinsulinemia (independent of body mass index or inactivity) may be a consequence of amputation and may be a mechanism leading to higher rates of CVD.

Studies of cardiovascular risk in amputees have had limited ability to explore the mechanism of disease and the nature of the increased cardiovascular risk because (1) baseline data on risk factors and comorbid conditions at the time of amputation are either not available or not analyzed, and (2) longitudinal follow-up studies exploring the interim development of cardiovascular risk factors among amputees who died or developed CVD do not exist.

The presumption of service connection for CVD among amputees has not been reexamined since it was put into place in 1979. No studies have been commissioned to provide any further information regarding association of CVD with amputation in a more contemporary cohort.

Costs to VA Associated with This Presumption

No cost estimates were available for this analysis. However, the long-term cost implications of this legislation are likely to continue to rise as veterans of contemporary conflicts in Iraq and Afghanistan have experienced higher rates of amputations than those of veterans of previous wars (Bilmes, 2007).

Lessons Learned

This case study reveals several important lessons that are relevant when considering strategies for improving the current system of presumptions. These lessons center around the type of evidence necessary to put a presumption in place, specifically what level of evidence should be required for a presumption, how the scientific base of evidence is updated based on new studies, how to evaluate evidence regarding exposures, outcomes, and potential mediators, and what types of evidence might make the scientific basis for a presumption more robust.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The scientific basis for this presumption was a single retrospective study of World War II veterans conducted by MFUA (Hrubec and Ryder, 1979). Although this study was large and apparently scientifically sound, confirmation of these findings with additional studies was warranted. Particularly since this presumption was put in place via legislative action (and not the more easily malleable administrative action on the part of VA), the highest standard of scientific evidence should have been employed, including replicating these findings in additional cohorts and exploring potential mechanisms for the observed association.

Understanding the mechanisms whereby an exposure leads to an outcome has important scientific implications for establishing a causal relationship and certainly has clinical relevance. However, understanding the mediators of an association between exposure and outcomes should not have bearing on policy considerations inherent in presumptions. Whether amputation leads to CVD via sedentary lifestyle or derangements in the coagulation cascade are irrelevant from a policy perspective. Both pathways are ultimately initiated by exposure that occurred during military service (i.e., amputation). Only knowledge of exposure (that a veteran was amputated during military service) and the association between exposure and outcome (that amputations lead to higher rates of CVD) are necessary for a valid presumption; knowing which of the multiple, possible pathways might mediate this association is not required.

Cardiovascular mortality rates for men have decreased dramatically over the last 30 years (Figure I-4) because of improved prevention and more aggressive treatment of modifiable risk factors for CVD. Given the decreasing mortality rates because of improved treatment, it is reasonable to question whether the association between amputations and cardiovascular mortality observed in the original MFUA study would be observed among contemporary amputees. Unfortunately, standard procedures are not in place for incorporating new studies of amputations and CVD. The ability to incorporate new evidence is additionally limited because this presumption was issued by Congress and not administratively by VA.

Certain populations of veterans, such as amputees, may be at high risk for complications as a result of military service and in particular need of compensation for these complications. Certain exposures, such as amputations, are particularly challenging to study except in cohort studies of exposed Service members and veterans. Both of these factors argue strongly for ongoing surveillance of high-risk Service member populations by the Department of Defense (DoD) and veteran

FIGURE I-4 Trends in cardiovascular mortality rates from 1979 to 2003.

SOURCE: Thom et al., 2006, p. 95.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

populations by VA. Such ongoing surveillance could have provided important confirmatory evidence to support the basis for this presumption, as well as a natural mechanism to update the scientific evidence with contemporary exposed Service members and veterans.

References

Advisory Committee on Cardiovascular Disorders and Mortality Rate in Amputees. 1954. Ultimate conclusions. London, England: Ministry of Pensions.

Bakalim, G. 1969. Causes of death in a series of 4738 Finnish war amputees. Artificial Limbs 13(1): 27-36.

Bilmes, L. 2007. Soldiers returning from Iraq and Afghanistan: The long-term costs of providing veterans medical care and disability benefits. RWP07-001. Paper prepared for the Allied Social Sciences Association Meeting, Chicago, IL, January 2007. http://www.epsusa.org/events/aea2007papers/bilmes.htm (accessed April 24, 2007).

Hrubec, Z., and R. A. Ryder. 1979. Report to the Veterans’ Administration Department of Medicine and Surgery on service-connected traumatic limb amputations and subsequent mortality from cardiovascular disease and other causes of death. Bulletin of Prosthetic Research 16(2):29-53.

Modan, M., E. Peles, H. Halkin, H. Nitzan, M. Azaria, S. Gitel, D. Dolfin, and B. Modan. 1998. Increased cardiovascular disease mortality rates in traumatic lower limb amputees. American Journal of Cardiology 82(10):1242-1247.

Rose, H. G., R. S. Yalow, P. Schweitzer, and E. Schwartz. 1986. Insulin as a potential factor influencing blood pressure in amputees. Hypertension 8(9):793-800.

Thom, T., N. Haase, W. Rosamond, V. J. Howard, J. Rumsfeld, T. Manolio, Z.-J. Zheng, K. Flegal, C. O’Donnell, S. Kittner, D. Lloyd-Jones, D. C. Goff, Jr., Y. Hong, R. Adams, G. Friday, K. Furie, P. Gorelick, B. Kissela, J. Marler, J. Meigs, V. Roger, S. Sidney, P. Sorlie, J. Steinberger, S. Wasserthiel-Smoller, M. Wilson, and P. Wolf. 2006. Heart disease and stroke statistics—2006 update: A report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 113(6):e85-e151. http://circ.ahajournals.org/cgi/content/short/113/6/e85#SEC3 (accessed March 9, 2007).

CASE STUDY 5:
RADIATION PRESUMPTIONS

This case study examines decisions related to radiation made over several decades for various groups of individuals involved in different military efforts. The case study describes the scientific basis and uncertainties for assessing exposure and risk to ionizing radiation.

The Scientific Basis of Risk Assessment for Ionizing Radiation

From the late 1940s to the early 1960s, about 200,000 American servicemen and women—along with civilians and groups from other countries—participated in above-ground tests of nuclear weapons (IOM, 2000, p. 1). U.S. testing was conducted primarily at the Nevada Test Site and the Marshall Islands in the Pacific Ocean. Many of the people at the test sites may have received substantial doses of ionizing radiation, and, as might be expected in any comparable population, many have developed cancer and other chronic diseases over the ensuing years. These “atomic veterans” are understandably concerned that their diseases could be service connected and have been quite vocal in seeking compensation. To provide compensation, it must be determined how likely it is that any particular cancer among the atomic veterans might be service related.

There is no doubt that radiation can cause cancer. Evidence in support of this claim comes from a vast literature of human epidemiologic studies, experimental animal studies, and basic

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

radiobiology research on cell cultures (ACHRE, 1995; Caldwell et al., 1980, 1983; Dalager et al., 2000; Darby et al., 1988, 1991, 1993; Doll et al., 1998; IOM, 1995, 2000; Johnson et al., 1997; Muirhead and Kendall, 2003; Muirhead et al., 2003, 2004; Pearce et al., 1997; Raman et al., 1987; Robinette et al., 1985; Watanabe et al., 1995). All major national and international advisory committees concerned with radiation risk assessment agree on this point (IARC, 2000; NRC, 2006; UNSCEAR, 2000). There is thus a prima facie case for causation for any individual with cancer and a history of radiation exposure. However, the magnitude of the risk and hence the probability of a causal connection in any individual is highly variable, depending upon such factors as the type of cancer, the radiation dose, age at exposure, time interval between exposure and cancer, and individual characteristics that determine response to radiation. Risk estimates are also inherently uncertain, both the estimates of population dose-response relationships and estimates based on an individual’s estimated dose and other modifying factors. The rarer the cancer, the more unstable the risk estimates will be, based on epidemiologic data, unless data for rarer sites are combined with other similar cancers (thereby introducing additional uncertainty about the appropriateness of that combination). Thus, it is certainly not the case that every cancer in a radiation-exposed individual is caused by that exposure, and therein lies the difficulty in establishing presumptions of causation or other criteria for compensation.

Although this case study will be focused on cancer, certain other endpoints have been definitely linked to ionizing radiation—notably severe mental retardation in those exposed during pregnancy and cataracts—and evidence for certain other diseases, such as CVD following high doses and nonmalignant tumors, is still evolving (NRC, 2006). There is also the potential for genetic effects that would be transmissible from one generation to the next; although genetic transmission has been demonstrated in animal experiments, there is so far no convincing evidence in humans, and the risk is generally thought to be low (NRC, 2006).

The mechanism by which radiation causes cancer is the physical interaction of a quantum of energy (e.g., an X ray or a particle such as a neutron or alpha particle) with a cell, leading to mutation of its DNA. Although such mutations can take several forms, the most serious come from double-strand breaks. There are several biological pathways for repair of such lesions, but some of these processes are error prone, and if not repaired or repaired incorrectly, a mutation can be fixed into the affected cell and passed on to its daughter cells. Depending upon the specific part of the chromosome affected, a mutation can lead to a chain of subsequent events and the growth of a clone of mutated daughter cells that can ultimately develop into a fully malignant tumor. Thus, radiation most frequently acts as one of the initiating steps in this complex process, although it is also possible that it can promote some of the later steps in an already initiated precancerous lesion. The probability of such an initial mutational event happening thus depends primarily on the probability of a damaging interaction between energy and DNA; the probability of such interaction depends on dose. The resulting cancer risk, however, can be modified by numerous other factors involved in the subsequent stages of malignant transformation or by the sensitivity of the target tissue (NRC, 2006).

The nature of this dose-response relationship and its modification by other factors has been extensively studied, both observationally in humans and in various experimental systems. Although experimental studies can be very useful for exploring the basic mechanisms of radiation carcinogenesis, quantitative risk estimates in humans have been derived from epidemiologic studies that have data on exposure. Numerous radiation-exposed populations have been studied, including the atomic bomb survivors, patients treated with diagnostic and therapeutic radiation, occupationally exposed nuclear workers and uranium miners, and groups exposed environmen-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

tally to emissions from nuclear accidents, nuclear weapons production and testing, or other sources. An overall conclusion that can be reached based on this body of evidence is that cancer rates roughly increase linearly with dose, with no evidence of an absolutely “safe” dose (a threshold below which there is no increase in risk). The slope of this dose-response relationship, which varies with cancer type, type of radiation, age, latency, and other factors is basically what determines an individual’s risk given their exposure, and hence the probability of causation (PC) if that person develops cancer. Various expert groups such as the NAS (NRC, 2006), the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR, 2000), and the International Agency for Research on Cancer (IARC, 2000) have compiled the full body of evidence on dose-response relationships and performed extensive joint analyses of the raw data from the major studies in order to develop these risk estimates and evaluate their uncertainties.

Veterans’ Studies

Among the sources of human risk estimates uniquely relevant to compensation policy for veterans are direct studies of radiation-exposed military personnel, specifically the atomic veterans. Several hundred thousand such veterans in total, mostly Americans but also substantial numbers of British veterans and other nationalities, have been included in several epidemiologic studies (ACHRE, 1995; Caldwell et al., 1980, 1983; Dalager et al., 2000; Darby et al., 1988, 1991, 1993; IOM, 1995, 2000; Johnson et al., 1997; Muirhead and Kendall, 2003; Muirhead et al., 2003, 2004; Pearce et al., 1997; Raman et al., 1987; Robinette et al., 1985; Watanabe et al., 1995). One of the first of these tests involved about 40,000 participants at the first hydrogen bomb tests at Bikini in 1946 (Operation CROSSROADS), later studied by the MFUA of the IOM (Johnson et al., 1997). A summary of these various studies is provided in the Appendix found at the end of this case study. Unfortunately, results from these studies are somewhat inconsistent, largely because most of these veterans’ doses were relatively low, and hence any elevation of cancer risks would be expected to be close to the limits of detection even in large epidemiologic studies.

Uncertainties in doses and the potential for confounding further add to the difficulty of establishing definitive dose-response relationships from such studies. (For example, the majority of participants in the CROSSROADS study had no dose estimates available.) The epidemiologic studies of radiation-exposed veterans are therefore more useful for establishing the plausibility of a population-level causal connection between veterans’ radiation exposures and subsequent cancer risks rather than for estimating these risks directly. Indeed, all of the point estimates found in Table I-1 and even most of the upper 95 percent confidence limits are substantially lower than the relative risk of 2, which translates to a 50 percent probability of causation for exposed veterans. Given these uncertainties of the data on veterans, a negative or inconsistent finding cannot be taken as definitive evidence against a causal connection, in the face of the wealth of positive evidence from other epidemiologic studies. But a positive finding could help bolster the evidence from other sources and can suggest specific cancer types that might be plausibly connected to radiation in the absence of compelling evidence against such an association.

For comparison, a graphical presentation of the relative risks from the atomic bomb survivors (Preston et al., 2003) is reproduced in Figure I-5. The specific cancers listed are those with sufficient numbers of cases available to support separate analysis and that show a clear radiation effect. The category “other solid cancers” may be quite heterogeneous in terms of radiosensitivity, but in general, convincing evidence to show that any specific cancer in this group is not radiosensitive would be elusive.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

TABLE I-1 Summary of Standardized Mortality Rates (95% Confidence Limits) from U.S. Veterans’ Studies of Cancer Conducted by the Institute of Medicine (IOM) in Participants in the CROSSROADS and Five Series of Nuclear Weapons Testing Exercises

Study Population

All Deaths

Cancer Mortality

Leukemia Mortalitya

CROSSROADSb

 

 

 

All Navy

1.05 (1.02-1.07)

1.01 (0.96-1.07)

1.02 (0.75-1.39)

Boarded target ships

1.06 (1.01-1.10)

1.03 (0.94-1.12)

1.01 (0.61-1.66)

Did not board target ships

1.04 (1.02-1.07)

1.01 (0.95-1.07)

1.02 (0.74-1.42)

Five Series Studyc

 

 

 

All participants vs. all non-participants

1.00 (0.98-1.02)

1.02 (0.98-1.06)

1.14 (0.90-1.44)

Land only

0.96 (0.93-0.99)

1.00 (0.95-1.06)

1.49 (1.04-2.13)

Sea only

1.03 (1.00-1.06)

1.04 (0.98-1.10)

0.92 (0.67-1.27)

a Excluding chronic lymphocytic leukemia (CLL).

b IOM, 1996, p. 65.

c IOM, 2002, pp. 63, 71.

FIGURE I-5 Estimates of excess relative risk per Sv for site-specific cancers among the atomic bomb survivors.

SOURCE: Preston et al., 2003.

The National Institutes of Health Radioepidemiologic Tables and Interactive Radiation Epidemiology Program

Because the risks vary by such factors as dose, type of radiation, age, gender, and latency, it is not particularly useful to settle upon a single summary number for risk. Instead, some algorithm is needed to predict an individual’s risk or related PC given the specifics of his or her situa-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

tion. In 1985, a provision of the Orphan Drug Act (Orphan Drug Act of 1983. Public Law 97-414. 97th Cong., 2d Sess.) directed the National Institutes of Health (NIH) to assemble a set of “radioepidemiologic tables,” which were to provide estimates of the PC for 12 types of cancer as a function of dose, age, gender, latency, and other factors (NIH, 1985, p. 65). Some examples of the calculations provided in the resulting report (NIH, 1985) are illustrated in Table I-2. The report also listed 14 specific cancer sites for which there was either no evidence of radiosensitivity or risk estimates were too uncertain to support PC calculations (NIH, 1985, p. 262). The estimates in the NIH report were derived mainly from risk estimates by earlier expert committees, principally the Biological Effects of Ionizing Radiation (BEIR) III report (NRC, 1980), and the methodology was carefully evaluated by a separate expert committee convened by the NAS (Lagakos and Mosteller, 1986; NRC, 1984). A summary of their conclusions is discussed in Chapter 9 of this report.

The act that created the 1985 radioepidemiologic tables (Orphan Drug Act of 1983. Public Law 97-414. 97th Cong., 2d Sess.) also included a provision calling for the tables to be periodically updated, but more than 15 years passed before this occurred. At that point, a new committee created a computer program, the Interactive Radiation Epidemiology Program (IREP), which could be used by individuals or compensation bodies to compute individualized PC estimates, given the specifics (dose, age, cancer type, etc.) of their experience. In addition to making the calculator accessible interactively on the Internet (https://www.niosh-irep.com/irep%5Fniosh/), an important advance was the inclusion of uncertainty estimates for each PC, based both on uncertainties in the epidemiologic data and the individual’s history (see the last two columns of Table I-2 for the previous illustrative examples). The interpretation of these uncertainties will be discussed later in this case study.

History of Radiation Compensation Criteria for Veterans

Congress has enacted three major pieces of legislation addressing compensation standards relating to the exposure of various populations to ionizing radiation from government programs, as reviewed in detail in Chapter 4 and earlier NAS reports (NRC, 2003, 2005). The first of these, the Veterans’ Dioxin and Radiation Exposure Compensation Standards Act of 1984 (Public Law 98-542. 98th Cong., 2d Sess.), did not establish any presumptions, but directed VA to establish a

TABLE I-2 Illustrative Calculations of the Probability of Causation (PC) Provided in the Original Radioepidemiological Tables and the Revised Interactive Radiation Epidemiology Program

Cancer Site

Gender

Age(s) at Exposure

Dose (Gy)

RR

PC

PC (1985)

PC (IREP)

95% CI (IREP)

Breast

Female

25

45

0.1 (X-ray)

1.03

3%

9%

4-26%

Bone

Male

15

20

0.7 (alpha)

15.2

93%

51%

8-92%

Thyroid

Male

17

25

0.3 (X-ray)

12.5

92%

66%

29-90%

Lung

Male (non-smoker)

28

55

0.4 (gamma)

1.23

19%

20%

7-41%

Acute leukemia

Female

20-35

44

0.14 (total)

1.10

9%

16%

4-36%

SOURCE: IREP (https://www.niosh-irep.com/irep%5Fniosh/); NIH, 1985.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

system for compensating veterans that eventually formed the basis for nonpresumptive compensation (Claims Based on Exposure to Ionizing Radiation. 2006. 38 CFR § 3.311), requiring evidence of sufficient exposure in the form of a quantitative dose reconstruction. These regulations permit claims for 24 “radiogenic diseases,” mainly specific cancers, plus certain cataracts and nonmalignant thyroid disease, as well as “any other cancer” not specifically enumerated.

In response to “concerns about the possible health effects of exposure to radiation” (Podgor, 2007, p. 522), Congress later enacted the Radiation-Exposed Veterans Compensation Act of 1988 (REVCA) (Public Law 100-321. 100th Cong., 2d Sess.) which established presumptions for 13 cancers. Congress later codified these presumptions (Disease Subject to Presumptive Service Connection. 2006. 38 CFR § 3.309) which includes 21 cancer types. The scientific basis for this selection appears to have been based at least in part on a report of the Science Panel of the Committee on Interagency Radiation Research and Policy Coordination (CIRRPC, 1988). This report explicitly avoids addressing any policy implications of their conclusions, but mentions a concurrent policy panel that intended to conduct such a review (CIRRPC, 1988, pp. iii-iv); no such report has been located by the Committee, and VA has not responded to queries from this Committee about how the criteria provided in REVCA (Public Law 100-321. 100th Cong., 2d Sess.) were developed. The list first promulgated in REVCA (Public Law 100-321. 100th Cong., 2d Sess.) has been amended several times subsequently in response to developing scientific evidence, but arguably the list did not fully reflect the available scientific evidence at the time. It excluded some clearly radiation-related cancers (e.g., lung, ovarian, and prostate) while including others (e.g., pharynx, small intestine, and bile ducts) with only tenuous scientific evidence to support an association of risk with radiation exposure. Nevertheless, recognizing the general carcinogenic potential of ionizing radiation, some of these rarer cancers, for which site-specific risk has not been convincingly demonstrated, could nevertheless be linked to radiation, as are other more common cancers.

The Radiation Exposure Compensation Act of 1990 (Public Law 101-426. 101st Cong., 2d Sess.) did not directly relate to veterans, but instead focused on two other exposed populations, the Colorado plateau miners and other uranium miners and residents downwind of the Nevada Test Site. The act is relevant to consider here because of how some presumptions that were established for these groups differ from those for veterans. Uranium miners are exposed to radon gas and its radioactive decay products (“radon daughters” or “radon progeny”), which deliver substantial doses to the lung epithelium, leading to lung cancer. The miners could claim compensation only for lung cancer, by criteria that involved their dose and smoking history. Because few if any veterans would have experienced similar types of exposure, this comparison is not particularly relevant here. The downwinders, on the other hand, received qualitatively similar exposures to those of many of the atomic veterans, albeit usually at generally low doses. The Radiation Exposure Compensation Act of 1990 did not require any individualized dose estimates for the downwinders, but rather based presumptions on county of residence and cancer type. As one might expect, the cancers for which these presumptions were established were similar to those of the Radiation-Exposed Veterans Compensation Act of 1988 (Public Law 100-321. 100th Cong., 2d Sess.).

The adjustment for smoking for uranium miners who developed lung cancer deserves further comment. There were relatively few studies of the interaction between smoking and ionizing radiation effects on lung cancer risk, but some of the earlier studies indicated a joint effect that was either approximately additive (or the atomic bomb survivors [Prentice et al., 1983]) or somewhat less than multiplicative (for underground uranium miners [Lubin and Steindorf, 1995]). On this

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

basis, the relative risk (RR) per unit of radiation would indeed be lower for smokers than for nonsmokers and the corresponding “doubling dose”—the dose required to attain an RR of 2—would be higher. More extensive reanalyses of all the miner data, however, subsequently indicated that a multiplicative model, or one intermediate between additive and multiplicative, was more appropriate (NRC, 1988). Under the multiplicative model, the RR for radiation would be the same for smokers and nonsmokers, and there would be no need to take smoking into account in deciding about compensation, even though the absolute risk is much higher in smokers. The reason for this is that both the background risk and the excess risk due to radiation are raised by the same factor, so the proportion of the risk attributable to radiation is the same. In addition, of course, there is the practical difficulty of establishing a smoking history in the face of an obvious incentive for a claimant to deny it.

Many of the subsequent amendments to these three acts, detailed in the footnotes to Table I-3 under the column heading “Other,” were in response to pressure from veterans groups seeking parity with the provisions of the Radiation Exposure Compensation Act of 1990.

Recommendations of the Advisory Committee on Human Radiation Experiments (1995) and the Human Radiation Interagency Working Group

In 1994, President Clinton established the Advisory Committee on Human Radiation Experiments (ACHRE), primarily to consider compensation policy for participants of certain medical experiments on human subjects with ionizing radiation and for people exposed to certain intentional releases of radiation into the environment for research purposes (ACHRE, 1995). However, the advisory committee interpreted its charge to encompass a number of other radiation-exposed groups, including the atomic veterans and the uranium miners. The advisory committee made a number of recommendations regarding compensation policy for these groups, noting the failure to update the radioepidemiological tables and calling for a review of the presumptions for the atomic veterans (ACHRE, 1995, ch. 18). These recommendations were in turn considered by the Radiation Exposure Compensation Act Committee (a committee established by the Human Radiation Interagency Working Group) which concluded that “the latest epidemiologic data suggest that the current exposure criteria do not accurately distinguish among lung cancer cases on a ‘more likely than not’ basis, and, therefore, do not effectively implement congressional intent” (RECAC, 1996, executive summary). They recommended revised criteria based either on dose or duration of exposure.

Dealing with Uncertainties

Uncertainties arise at both the population and individual level. At the population level, there can be uncertainty about whether a causal relationship between radiation and cancer has been established at all, as discussed in Chapter 9. If there is convincing evidence of a causal relationship, the estimate of risk attributable to exposure will be subject to statistical variation, since these estimates are based on a finite sample. The risk for particular categories of individuals (e.g., categories defined by dose, age, and time since exposure) will inevitably be even more uncertain than the average risk, since such estimates will be based on smaller sample sizes than for the overall risk. Epidemiologic risk estimates can also be distorted by selection bias, uncontrolled confounding, exposure measurement error, or a host of other potential biases. At the individual level, there are additional uncertainties about the specifics of an individual’s history (as discussed in Chapters 7 and 9). These might include the estimate of an individual’s dose (or

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

TABLE I-3 Selected Cancers and Diseases for Which Presumptive Service Connection Has Been Established Under Various Acts of Congress or Regulations

Disease

Veterans Dioxin and Radiation Exposure Compensation Standards Act of 1984

Radiation-Exposed Veterans Compensation Act of 1988

Radiation Exposure Compensation Act of 1990

Other

BEIR VII

Leukemia (excluding CLL)

x

x

x

 

x

Cancer:

 

 

 

 

 

Thyroid

x

x

x

 

x

Urinary bladder

 

 

 

f

x

Kidney

 

 

 

k

 

Lung

x

 

 

h, j

x

Bronchiolo-alveolar

 

 

 

g

 

Bone

x

 

 

h, j

 

Female breast

x

x

x

 

x

Male breast

 

 

 

f

 

Pharynx

 

x

x

 

 

Esophagus

 

x

x

 

 

Stomach

 

x

x

 

x

Colon

 

 

 

f, h, j

x

Small intestine

 

x

x

 

 

Pancreas

 

x

x

 

 

Bile ducts

 

x

x

 

 

Gall bladder

 

x

x

 

 

Skin

x

 

 

 

 

Liver

x

x

x

 

x

Salivary

 

 

 

f

 

Ovarian

 

 

 

b, f, h, j

x

Brain and CNS

 

 

 

c, f, h, j

 

Rectum

 

 

 

d

 

Prostate

 

 

 

e

x

Uterus

 

 

 

 

x

Any other cancer

 

 

 

e

x

Multiple myeloma

 

x

x

 

 

Lymphomas (excl. Hodgkin’s)

 

x

x

d

 

Nonmalignant thyroid nodules

 

 

 

a

 

Cataracts (posterior subcapsulary)

 

 

 

a

 

Parathyroid adenoma

 

 

 

b, k

 

Polycythemia vera

x

 

 

i

 

SOURCES: aVA, 1989; bVA, 1993; cVA, 1994; dVA, 1995; eVA, 1998; fRadiation Exposure Compensation Act Amendments of 2000 (Public Law 106-245. 106th Cong., 2d Sess.); gVA, 2000; hVA, 2002b; iVA, 2002a; jVeterans Benefits Improvement Act of 2004 (Public Law 108-454. 108th Cong., 2d Sess.); kClaims Based on Exposure to Ionizing Radiation. 2006. 38 CFR § 3.311.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

even, in the case of a presumptive causation case, whether the individual met the minimum exposure criterion to qualify for the class at all), as well as other specific criteria (e.g., latency). IREP, described above, makes an attempt to incorporate both types of uncertainty into a “credibility interval” for each PC estimate it computes.

Other Considerations

Costs

The only information on the costs of existing radiation presumptions available to the Committee came from a single memo prepared by the Office of Resource Management concerning the proposal to add bronchiolo-aveolar carcinoma of the lung to the list of presumptions. This document estimated a caseload of 120 veterans and 338 survivors during fiscal year 2000, at a cost of $5.9 million, cumulative cost through 2004 of $33.3 million, with administrative costs of $308,000 (VA, 1999, p. 1).

Sensitivity and Specificity

As there is no unique marker for the cause of any given cancer and because cancers can be caused by any of a number of different factors (separately or acting in combination), any compensation scheme must necessarily lead to both false positives (compensation of cases not caused by radiation) and false negatives (failure to compensate a case that in fact was caused by radiation). Current VA policy is to err on the side of false positives. Thus, the aim of the current policy is to achieve high sensitivity, necessarily at the expense of low specificity. Given the probabilistic nature of radiation carcinogenesis, no compensation policy can hope to be both highly sensitive and highly specific.

Secrecy

Arguably a unique feature of much of the radiation story concerns the national security elements that pervaded the entire Cold War culture in which nuclear weapons testing exercises were conducted. This secrecy element has been addressed in detail in the ACHRE report (1995) and Chapter 11 of this report. Although much of the information about the veterans’ participation in nuclear weapons testing has now been declassified, access to records and even in some instances the simple fact of a veteran’s participation in particular exercises has undeniably been a barrier to establishing a credible claim in the past.

Lessons Learned

This case study illustrates a situation quite different from many of those in other case study chapters; for radiation, there is an abundant literature of solid science—some relating to dose-response relationships in general populations, some specific to veteran—supon which to base compensation policy. Despite this, there remain numerous uncertainties, particularly with respect to estimation of an individual’s exposures and with respect to the risk for specific rare cancers. These uncertainties in large part are responsible for the shift in emphasis from individual PC-based criteria for compensation in the 1984 Act (Veterans’ Dioxin and Radiation Exposure Compensation Standards Act of 1984. Public Law 98-542. 98th Cong., 2d Sess.) to the estab-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

lishment of presumptions in the 1988 act (Radiation-Exposed Veterans Compensation Act of 1988. Public Law 100-321. 100th Cong., 2d Sess.). This evolution reflects the growing recognition of the difficulty of establishing criteria for individual compensation in the face of such uncertainties and the need for presumptions.

In general, studies of the effects of radiation in veteran populations have been of limited utility, in part because of the relatively small number of excess cancers expected in the available cohorts and, importantly, because of the failure to track individual exposures systematically or accurately. For this reason, these direct observations in the population of greatest relevance should not negate the wealth of more informative data from other populations. Nevertheless, the availability of informative epidemiologic data from other populations has made it possible to construct quantitative models to guide compensation policy for radiation-exposed veterans, which has often not been possible for other exposures. These studies also illustrate the difficulty of dealing with other causal risk factors for cancer, such as smoking. Yet even in that case, this case study has demonstrated that it is possible to develop scientifically based policies using strong epidemiologic evidence about their joint effects.

Ultimately, much of the force behind the movement for compensation for the atomic veterans derived from a sense of outrage that their government deliberately exposed them to radiation with at least some knowledge of risks involved at the time (Podgor, 2007). Furthermore, these risks were often denied by government officials, both at the time of exposure when they were not properly informed and later when diseases were manifest and their attempts at redress were rebuffed. On this basis, the veterans feel their claims for compensation are enhanced by the culpability of the government (ACHRE, 1995).

Appendix: Epidemiologic Studies of Radiation-Exposed Veterans

The first study of cancer in the atomic veterans was published by investigators from the CDC in 1980 (Caldwell et al., 1980, p. 1577), describing nine cases of all types of leukemia among the 3,224 participants of nuclear test explosion “Smoky” conducted in 1957, compared with 3.5 expected, a statistically significant excess. A subsequent report (Caldwell et al., 1983, p. 622) raised the toll to 10 leukemias compared with 4.0 expected, but found no statistically significant excess of any other cancers. The mean dose for the entire cohort was 0.5 cGy (ranging from 0-3.0 cGy for the leukemia cases, with a mean of 1.0 cGy) (Caldwell et al., 1980, p. 1577). The above-ground test conducted at the Nevada Test Site, “Smoky,” produced substantial fallout exposures to Utah downwinders in addition to the onsite military participants. However, the original investigators declared that the measured doses were inadequate to account for all of the observed leukemia excess (Caldwell et al., 1983).

Largely in response to the CDC study, NAS undertook a more extensive epidemiologic study of 46,186 participants in five series of tests at the Nevada Test Site (Robinette et al., 1985, as referenced in IOM, 2000, p. 29). This report confirmed the previous excess of leukemia among Smoky participants, but found no excesses among the other participants or any consistent patterns for other cancers. This study was later found to be flawed by the inclusion of 4,500 individuals who had not participated, the exclusion of 15,000 others who had, and inadequacies in the dosimetry (Gelband, 1992, as referenced in IOM, 2000, p. 29). A separate IOM committee concluded that the film badge dosimetry was unsuitable for epidemiologic purposes and recommended that dose reconstruction methods be used instead to impute dose estimates to study participants (IOM, 1995, p. 14). This formed the basis for a revised report on this cohort, now including 70,000 participants and 65,000 comparable nonparticipants (IOM, 2000, p. 1). This

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

report found no significant differences between the two groups for overall or site-specific cancer mortality, although a nonsignificant 14 percent increase in leukemia was noted.

A cohort of 40,000 participants at the first hydrogen bomb tests at Bikini in 1946 (Operation CROSSROADS) was compared with a similar cohort of nonparticipants in a study by IOM’s MFUA (Johnson et al., 1997). A significant 5 percent increase in all cause mortality (P < .001) was seen, but there were no significant excesses for all cancers (1.4 percent) or for leukemia (2 percent). The authors concluded that this pattern did not support a radiation hypothesis.

Another study was conducted on 8,554 Navy participants in a series of 35 tests conducted at the Pacific Proving Grounds in the Marshall Islands in 1958 (Operation HARDTACK) compared with 14,625 Navy veterans who did not participate in any tests (Watanabe et al., 1995, pp. 524-525). Overall, the exposed cohort received an average of 0.4 cGy and experienced a nonsignificant 14 percent increase in total cancer (95% confidence interval [CI], −2 to 33%) and a significant 47 percent increase in digestive cancers (6-104%). Among the 1,064 participants who received over 1 cGy, a significant 42 percent excess (95% CI, 4-96%) of total cancer was seen relative to nonparticipants, but no single site was noteworthy, and there were no significant dose-response trends. In comparison with general population rates, the excess of total cancer was only 12 percent (−1 to 27 percent) and 24 percent for digestive cancers (−4 to 57 percent) (Watanabe et al., 1995, pp. 524-525).

A subsequent analysis (Dalager et al., 2000) expanded the high-dose cohort to include 1,010 participants who received at least 5 cGy (374 Navy participants at the Pacific Proving Grounds, 636 from other services, mainly from the Nevada Test Site). In this group, the overall mortality was increased by 22 percent (RR = 1.22; 95% CI 1.04-1.44%); mortality from all lymphopoietic cancers increased by 272 percent (RR = 3.72; 95% CI 1.28-10.83); and there was a nonsignificant 41 percent increase in respiratory cancers (RR = 1.41; 95% CI 0.91-2.18), but there was no excess of digestive cancers or all other cancers. The comparison of all high-dose participants with Navy-only controls found that the numbers of cancers in the 374 Navy participants was inadequate to establish significant excesses, except for lymphopoietic cancers and all causes combined (Dalager et al., 2000).

In addition to U.S. military participants, veterans from other countries have also been studied, and it is reasonable to expect that their experience would be broadly comparable. A small Canadian study (Raman et al., 1987) of 954 participants and twice as many controls found no differences, but another small cohort of 528 New Zealand participants compared to 1,504 controls (Pearce et al., 1997, p. 139) found a significant 5.6-fold excess mortality from leukemia (n = 4) and 3.8-fold excess of deaths from hematologic malignancies (n = 8) (Pearce et al., 1997, p. 142).

A much larger study of British participants of nuclear tests conducted in Australia (Darby et al., 1988, p. 335) found higher rates of leukemia and multiple myeloma than in matched controls (28 compared with 6). Compared with national rates, however, it appeared that the excess was driven more by unusually low rates in controls than by high rates in cases. Various updates of the UK studies have been reported (Darby et al., 1991, 1993; Muirhead and Kendall, 2003; Muirhead et al., 2003, 2004), which have generally confirmed the original findings, except for eliminating the excess of multiple myeloma. The most recent publication (Muirhead et al., 2004, p. 227) shows an 83 percent excess (15-193 percent) of all leukemia excluding CLL, rising to a 199 percent excess (26-641 percent) for the period 2 to 25 years after exposure. As before, however, the rate relative to the general population was only slightly elevated in the exposed cohort (RR of 1.06 [n = 40] overall or of 1.23 [n = 18] 2 to 25 years after exposure), compared with

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

considerably lower rates in the control cohort (RR of 0.58 [n = 23] overall or relative risk of 0.36 [n = 6] 2 to 25 years after exposure) (Muirhead et al., 2004, p. 227).

This literature up to 1995 was reviewed by ACHRE (1995, ch. 10), who attempted to put these results in perspective by noting that among the roughly 220,000 U.S. participants, current risk estimates would predict about 100 excess lifetime cancers, a small number in comparison with the 48,000 that would be expected from natural causes, an excess whose magnitude is not only uncertain, but would be virtually undetectable by even the best designed epidemiologic study. They also estimated that among the 1,200 veterans who received more than 5 cGy, about 8 excess cancers would be expected. Nevertheless, the committee was highly critical of the government’s culture of secrecy that made it very difficult for veterans to learn about their experiences, their failure to protect participants against any but acute effects from high-dose exposures, and their failure to maintain adequate records about either the individuals’ participation or their policy-making process.

References

ACHRE (Advisory Committee on Human Radiation Experiments). 1995. The human radiation experiments: Final report. http://www.hss.energy.gov/HealthSafety/ohre/roadmap/achre/report.html (accessed June 24, 2007).

Caldwell, G. G., D. B. Kelley, and C. W. Heath, Jr. 1980. Leukemia among participants in military maneuvers at a nuclear bomb test. A preliminary report. Journal of the American Medical Association 244(14):1575-1578.

Caldwell, G. G., D. Kelley, M. Zack, H. Falk, and C. W. Heath, Jr. 1983. Mortality and cancer frequency among military nuclear test (Smoky) participants, 1957 through 1979. Journal of the American Medical Association 250(5):620-624.

CIRRPC (Committee on Interagency Radiation Research and Policy Coordination). 1988. Use of probability of causation by the Veterans Administration in the adjudication of claims of injury due to exposure to ionizing radiation. Science Panel Report Number 6. Washington, DC: Office of Science and Technology Policy. http://www.cdc.gov/niosh/ocas/pdfs/irep/cirrpc.pdf (accessed March 25, 2007).

Dalager, N. A., H. K. Kang, and C. M. Mahan. 2000. Cancer mortality among the highest exposed U.S. atmospheric nuclear test participants. Journal of Occupational and Environmental Medicine 42(8):798-805.

Darby, S. C., G. M. Kendall, T. P. Fell, J. A. O’Hagan, C. R. Muirhead, J. R. Ennis, A. M. Ball, J. A. Dennis, and R. Doll. 1988. A summary of mortality and incidence of cancer in men from the United Kingdom who participated in the United Kingdom’s atmospheric nuclear weapon tests and experimental programmes. British Medical Journal 296(6618):332-338.

Darby, S. C., J. A. O’Hagan, G. M. Kendall, R. Doll, T. P. Fell, and C. R. Muirhead. 1991. Completeness of follow up in a cohort study of mortality using the United Kingdom National Health Service Central Registers and records held by the Department of Social Security. Journal of Epidemiology and Community Health 45(1):65-70.

Darby, S. C., G. M. Kendall, T. P. Fell, R. Doll, A. A. Goodill, A. J. Conquest, D. A. Jackson, and R. G. Haylock. 1993. Further follow up of mortality and incidence of cancer in men from the United Kingdom who participated in the United Kingdom’s atmospheric nuclear weapon tests and experimental programmes. British Medical Journal 307(6918):1530-1535.

Doll, R., C. Sharp, C. Muirhead, and S. Darby. 1998. Study of UK men who had participated in the UK nuclear weapons test programme. Journal of Radiological Protection 18(3):209-210.

Gelband, H. 1992. A discussion of questions about the 1985 NAS report “Mortality of nuclear weapons test participants.” Washington, DC: Office of Technology Assessment.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

IARC (International Agency for Research on Cancer). 2000. Ionizing radiation, part 1: X- and gamma (g)-radiation, and neutrons. Lyon, France: International Agency for Research on Cancer.

IOM (Institute of Medicine). 1995. A review of the dosimetry data available in the Nuclear Test Personnel Review (NTPR) Program. Washington, DC: National Academy Press.

IOM. 1996. Mortality of veteran participants in the CROSSROADS nuclear test. Washington, DC: National Academy Press.

IOM. 2000. The Five Series Study: Mortality of military participants in U.S. nuclear weapons tests. Washington, DC: National Academy Press.

IOM. 2002. The Five Series Study: Mortality of military participants in U.S. nuclear weapons tests. Washington, DC: National Academy Press.

Johnson, J. C., S. Thaul, W. F. Page, and H. Crawford. 1997. Mortality of veteran participants in the CROSSROADS nuclear test. Health Physics 73(1):187-189.

Lagakos, S., and F. Mosteller. 1986. Assigned shares in compensation for radiation-related cancers. Risk Analysis 6(3):345-357.

Lubin, J. H., and K. Steindorf. 1995. Cigarette use and the estimation of lung cancer attributable to radon in the United States. Radiation Research 141(1):79-85.

Muirhead, C. R., and G. M. Kendall. 2003. UK nuclear-test veterans. Lancet 362(9380):331-332.

Muirhead, C. R., D. Bingham, R. G. Haylock, J. A. O’Hagan, A. A. Goodill, G. L. Berridge, M. A. English, N. Hunter, and G. M. Kendall. 2003. Follow up of mortality and incidence of cancer 1952-98 in men from the UK who participated in the UK’s atmospheric nuclear weapon tests and experimental programmes. Occupational and Environmental Medicine 60(3):165-172.

Muirhead, C. R., G. M. Kendall, S. C. Darby, R. Doll, R. G. Haylock, J. A. O’Hagan, G. L. Berridge, M. A. Phillipson, and N. Hunter. 2004. Epidemiological studies of UK test veterans: II. Mortality and cancer incidence. Journal of Radiological Protection 24(3):219-241.

NIH (National Institutes of Health). 1985. Report of the National Institutes of Health ad hoc working group to develop radioepidemiologic tables. NIH Publication No. 85-2748. Washington, DC: U.S. Department of Health and Human Services.

NRC (National Research Council). 1980. The effects on populations of exposure to low levels of ionizing radiation. Washington, DC: National Academy Press.

NRC. 1984. Assigned share for radiation as a cause of cancer: Review of radioepidemiological tables assigning probability of causation. Washington, DC: National Academy Press.

NRC. 1988. Health risks of radon and other internally deposited alpha-emitters: BEIR IV. Washington, DC: National Academy Press.

NRC. 2003. A review of the dose reconstruction program of the Defense Threat Reduction Agency. Washington, DC: The National Academies Press.

NRC. 2005. Assessment of the scientific information for the radiation exposure screening and education program. Washington, DC: The National Academies Press.

NRC. 2006. Health risks from exposure to low levels of ionizing radiation: BEIR VII - Phase 2. Washington, DC: The National Academies Press.

Pearce, N., R. Winkelmann, J. Kennedy, S. Lewis, G. Purdie, T. Slater, I. Prior, and J. Fraser. 1997. Further follow-up of New Zealand participants in United Kingdom atmospheric nuclear weapons tests in the Pacific. Cancer Causes and Control 8(2):139-145.

Podgor, M. F. 2007. The inability of World War II atomic veterans to obtain disability benefits: Time is running out on our chance to fix the system. Elder Law Journal 13(2):519-552.

Prentice, R. L., Y. Yoshimoto, and M. W. Mason. 1983. Relationship of cigarette smoking and radiation exposure to cancer mortality in Hiroshima and Nagasaki. Journal of the National Cancer Institute 70(4):611-622.

Preston, D. L., Y. Shimizu, D. A. Pierce, A. Suyama, and K. Mabuchi. 2003. Studies of mortality of atomic bomb survivors. Report 13: Solid cancer and noncancer disease mortality: 1950-1997. Radiation Research 160(4):381-407.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Raman, S., C. S. Dulberg, R. A. Spasoff, and T. Scott. 1987. Mortality among Canadian military personnel exposed to low-dose radiation. Canadian Medical Association Journal 136(10):1051-1056.

RECAC (Radiation Exposure Compensation Act Committee). 1996. Final report of the Radiation Exposure Compensation Act Committee. Washington, DC: Human Radiation Interagency Working Group. http://www.hss.energy.gov/HealthSafety/ohre/roadmap/uranium/index.html (accessed June 24, 2007).

Robinette, C. D., S. Jablon, and T. L. Preston. 1985. Studies of participants in nuclear tests. Washington, DC: National Academy Press.

UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation). 2000. UNSCEAR 2000 report: Sources and effects of ionizing radiation. Vol. 2. New York: United Nations. http://www.unscear.org/unscear/en/publications/2000_2.html (accessed June 22, 2007).

VA (Department of Veterans Affairs). 1989. Claims based on exposure to ionizing radiation. Final regulations. Federal Register 54(200):42802-42803.

VA. 1993. Claims based on exposure to ionizing radiation. Final rule. Federal Register 58(7):16358-16359.

VA. 1994. Claims based on exposure to ionizing radiation. Final rule. Federal Register 59(171):45975.

VA. 1995. Claims based on exposure to ionizing radiation (lymphomas other than Hodgkin’s disease and cancer of the rectum). Final rule. Federal Register 60(198):53276-53277.

VA. 1998. Claims based on exposure to ionizing radiation (prostate cancer and any other cancer). Final rule. Federal Register 63(185):50993-50995.

VA. 1999. H.R. 690, 106th Congress: Methodology (Cost Estimate). Washington, DC: Office of Resource Management.

VA. 2000. The Veterans Millennium Health Care and Benefits Act. Final rule. Federal Register 65(136):43699-43700.

VA. 2002a. Claims based on exposure to ionizing radiation. Final rule. Federal Register 67(31):6870-6871.

VA. 2002b. Diseases specific to radiation-exposed veterans. Final rule. Federal Register 67(17):3612-3616.

Watanabe, K. K., H. K. Kang, and N. A. Dalager. 1995. Cancer mortality risk among military participants of a 1958 atmospheric nuclear weapons test. American Journal of Public Health 85(4):523-527.

CASE STUDY 6:
MUSTARD GAS AND LEWISITE PRESUMPTIONS

This case study examines the 1992 and 1994 decisions by the VA to establish presumptive service connection for health outcomes related to mustard gas and lewisite exposures among World War II veterans. It outlines several issues critical to protecting and treating those in military service: secrecy and classification of exposures, the latency period to adverse health outcome following exposures, lack of medical records, the involvement of “volunteers” in chemical experiments, the classification of scientific evidence and the use of scientific evidence, and the impact of policy considerations in making compensation decisions.

Background

In 1992, VA recognized that some World War II veterans had been exposed to mustard gas during laboratory experiments that involved full-body, field, or chamber tests of protective equipment. A presumption was proposed that covered several diseases associated with exposure to mustard gas. VA justified this presumptive service connection based on several factors: (1) the studies were classified; (2) participants were directed not to discuss their participation in the studies; (3) their medical records were sparse; and (4) no long-term follow-up was conducted or provided for the participants (VA, 1992a). After public comment on this proposed regulation, VA issued a final rule in July 1992 (VA, 1992b). In early 1994, VA revised the presumption

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

based on a study conducted by the IOM on the health effects of exposure to mustard gas and lewisite and issued a proposed rule (VA, 1994a). The second presumptive service-connection rule was issued in August 1994 and amended the original proposal by (1) adding more diseases to the original list of diseases; (2) adding the compound lewisite to the rule; (3) adding veterans who might have been exposed during World War I and in studies after the end of World War II; and (4) clarifying the extent of exposure. Again the presumptive decision was based on the same factors stated in the original proposed rule (VA, 1994b). In August 2006, the VA Under Secretary for Health issued an information letter intended to inform health-care providers about the diseases resulting from exposure to mustard gas, lewisite, and some 250 different agents used in studies conducted at Edgewood-Aberdeen during the period of 1955 to 1975 (VA, 2006a). However, this letter specifically states that VA does not presumptively recognize any long-term health effects from agents other than mustard agents and lewisite. Any health effects from exposure to the 250 chemical agents used during chemical warfare testing studies are to be addressed on an individual basis (VA, 2006a).

Although chemical warfare has its roots in antiquity, World War I was the beginning of modern day chemical agent usage. The German army released tens of thousands of pounds of chlorine gas in April 1915 near Ypres, Belgium, as an offensive weapon (Joy, 1997). In July 1917 the blistering agent sulfur mustard, called mustard gas, was released for the first time in the same general area (Joy, 1997). This agent caused the most casualties during World War I, but not the most fatalities. Some 400,000 troops were injured by sulfur mustard during World War I (Gilchrist, 1928, as referenced in IOM, 1993, p. 9). This began the trend of concern regarding mustard gas and other chemical agents. Further concern was forthcoming with the December 1943 Bari Harbor, Italy, incident and the initiation and continued laboratory testing begun during World War II (Alexander, 1947; Harris and Paxman, 1982; Reminick, 2001; Tucker, 2006). The alleged use of mustard agents offensively by other countries since World War II has been documented: Egypt against Yemen in the 1960s (Shoham, 1998) and Iraq against Iran in the 1980s (Balali-Mood et al., 2005; IOM, 1993; Sidell et al., 1997; Tucker, 2006).

Although there was no chemical warfare during World War II, the lingering memories of World War I chemical usage by the German army remained foremost in the minds of the U.S. military. Chemical and biological warfare agent use and testing against the Chinese by the notorious Japanese Unit 731, commanded by General Shiro Ishii, spurred the United States to develop protective clothing and respirators, better termed gas masks (Harris, 1994). Sulfur mustard use was a major concern (IOM, 1993). Consequently, history reveals that the U.S. government began preparing for chemical agent usage prior to the attack on Pearl Harbor by the Japanese in December 1941. Under President Franklin D. Roosevelt, war-related research units were established and placed under the control of the White House Office of Scientific Research and Development (IOM, 1993). Two groups were established to investigate different aspects of the mustard agents and lewisite. The Committee on Medical Research studied protective ointments and alternative treatments through the NRC’s Committee on Treatment of Gas Casualties. The second group, the National Defense Research Committee, studied protective clothing and gas masks through different military units. Both these groups were involved in secret testing programs involving mustard agents and lewisite. Because of a lack of animal models of human injury, researchers in these programs decided in 1942 to use human subjects (IOM 1993). In total these two testing programs involved approximately 60,000 military personnel as human experimental subjects deemed by the government to be “volunteers” (IOM, 1993, pp. vii, 1). Some 4,000 volunteers participated in gas chamber or field exercises (IOM, 1993, p. 1). The gas chamber tests

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

involved high concentrations of mustard agents or lewisite. The field exercises were conducted over heavily contaminated soils. Exposure concentrations ranged from mild, such as a drop of agent on the arm in patch tests, to severe, repeated gas chamber trials with and without protective clothing. All human experimental subjects were sworn to secrecy regarding their involvement in these tests (IOM, 1993).

However, concerns regarding health consequences of sulfur mustard and lewisite exposures were not entertained by VA until the early 1990s. Concern by VA at this time was brought about by some of the former human subjects as they attempted to obtain medical help for diseases thought to be associated with exposure to mustard gas or lewisite. VA was unable to resolve these cases for two reasons: (1) the absence of records or documentation regarding the individual’s participation in exposure testing, and (2) the uncertainty of health effects linked to the two agents of concern, sulfur mustard and lewisite (VA, 1992a,b). On January 15, 1992, VA announced guidelines for the handling and subsequent adjudication of these cases (VA, 1992a). These guidelines included a lessening of normal requirements for documenting an individual’s participation in exposure testing and awarding of compensation based upon disease presentation.

VA at this time was proposing to adjudicate compensation claims for disabilities or deaths resulting from the chronic effects of exposure to mustard gas while in service. Supplementary information provided in the proposed rule states that

Some Naval personnel were experimentally exposed to mustard gas during full-body, field or chamber tests of protective equipment and clothing conducted at the Naval Research Laboratory, located at Edgewood Arsenal, Washington, DC, between 1943 and 1945. Similar testing may have been conducted at other locations during World War II. These World War II tests were classified, participants were instructed not to discuss their involvement, and medical records associated with the tests are generally unavailable. No long-term follow-up examinations were conducted. For these reasons, some participants may not have filed claims with VA for disabilities resulting from mustard gas poisoning, or, if they did file claims, may have experienced difficulty in establishing entitlement to benefits.


VA believes that the special circumstances surrounding these World War II testing programs have placed veterans who participated in them at a disadvantage when attempting to establish entitlement to compensation for disability or death resulting from experimental exposure. The proposed rule specifies that, if exposure occurred under the described circumstances, disabilities or deaths resulting from certain diseases are to be recognized as connected to a veteran’s exposure in-service.


A review of the available medical literature by Veterans Health Administration (VHA) personnel indicates that the chronic, long-term effects of acute mustard gas poisoning may include laryngitis, bronchitis, emphysema, asthma, conjunctivitis, keratitis, and corneal opacities. Chronic forms of these conditions which developed subsequent to experimental exposure during World War II will be service-connected.

(VA, 1992a, pp. 1-2)


After public comment regarding the proposed rule, VA published the final rule with an effective date of July 31, 1992 (VA, 1992b). The proposed rule was modified to include World War I veterans who could claim residual disability from sulfur mustard exposure. These individuals would be able to show service medical record entries regarding acute sulfur mustard exposure

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

and that the long-term chronic effects from exposure have existed continually since the acute exposure episode.

The use of long-term chronic effects of exposure was retained in the final rule by the following VA rationale:

Veterans who were exposed to mustard gas during experimental tests of protective clothing and equipment during World War II, however, face a potentially insurmountable disadvantage when attempting to establish entitlement to compensation. Those tests were conducted behind a strictly enforced veil of secrecy, medical records associated with the tests are generally unavailable, and no long-term follow-up examinations were conducted. As a result, service medical records for individuals who participated in those tests may not show evidence of the acute effects of mustard gas exposure. Furthermore, it is likely that participants who developed chronic effects of exposure did not previously file compensation claims with VA solely because they had been instructed not to discuss their involvement in the tests. Physicians who may have treated these veterans for chronic effects more than 40 years ago have almost certainly retired from private practice, making it impossible for a veteran to establish that a chronic form of one of the specified disabilities has existed continually since exposure to mustard gas.

(VA, 1992b, p. 2)


The VA indicated in the final rule that it would not have a significant financial impact: (1) the annual effect on the economy would be less than $100 million; (2) there would be no major increase in costs or prices; and (3) it would not “have significant adverse effects on competition, employment, investment, productivity, innovation, or on the ability of United States-based enterprises to compete with foreign-based enterprises in domestic or export markets” (VA, 1992b, p. 2). Additionally, VA would contract with the NAS to address questions raised during the public comment period for which they did not have sufficient information to answer and issue a subsequent rule if necessary (VA, 1992b).

IOM Report Entitled Veterans at Risk: The Health Effects of Mustard Gas and Lewisite

As a result of public comment to the proposed rule, VA stated in the final rule that they would contract with NAS to conduct a review of the published literature worldwide, medical and scientific, to include languages other than English, covering the long-term health effects of mustard gas exposure. NAS, through IOM, established a committee to address this issue and to publish a report. The committee (Committee to Survey the Health Effects of Mustard Gas and Lewisite) included experts in the disciplines of toxicology, chemistry, epidemiology, psychology, oncology, dermatology, ophthalmology, occupational medicine, and environmental medicine. The task delivered to this committee was to “survey the medical and scientific literature on mustard agents and lewisite, assess the strength of association between exposure to these agents and the development of specific diseases, identify the gaps in the literature, and recommend strategies and approaches to deal with any gaps found” (IOM, 1993, p. vi).

IOM Report General Conclusions

The lack of follow-up health assessments of the human subjects in the WWII gas chamber and field tests severely diminished the amount and quality of information that could

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

be applied in the assessment of long-term health consequences of exposure to mustard agents and Lewisite.


The levels of exposure to mustard agents or Lewisite experienced by the human subjects may have been much higher than inferred in the summaries of the gas chamber and field tests.


The committee was additionally dismayed that there were no epidemiological studies done of mustard agent-exposed, U.S. chemical weapons production workers, war gas handlers and trainers, or combat casualties from WWII.

(IOM, 1993, p. 3)

IOM Report Specific Findings

The data found generally fell into three categories of causal relationships: indicated, suggested, or insufficient evidence of a relationship. The committee emphasized that “no condition evaluated could be removed from consideration as a health consequence of exposure to these agents. Thus, for many diseases there remains significant doubt” (IOM, 1993, p. 4).

The evidence found indicated a causal relationship between exposure and the following health conditions:

  • Respiratory cancers

    • Nasopharyngeal

    • Laryngeal

    • Lung

  • Skin cancer

  • Pigmentation abnormalities of the skin

  • Chronic skin ulceration and scar formation

  • Leukemia (typically acute nonlymphocytic type, nitrogen mustard)

  • Chronic respiratory diseases (also Lewisite)

    • Asthma

    • Chronic bronchitis

    • Emphysema

    • Chronic obstructive pulmonary disease

    • Chronic laryngitis

  • Recurrent corneal ulcerative disease (includes corneal opacities; acute severe injuries to eye from Lewisite will also persist)

  • Delayed recurrent keratitis of the eye

  • Chronic conjunctivitis

  • Bone marrow depression and (resulting) immunosuppression (an acute effect that may result in greater susceptibility to serious infections with secondary permanent damage to vital organ systems)

  • Psychological disorders

    • Mood disorders

    • Anxiety disorders (including post-traumatic stress disorder [PTSD])

    • Other traumatic stress disorder responses (These may result from traumatic or stressful features of the exposure experience, not a toxic effect of the agents themselves.)

  • Sexual dysfunction (Scrotal and penile scarring may prevent or inhibit normal sexual performance or activity.)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The evidence found suggested a causal relationship between exposure and the following health conditions:

  • Leukemia (acute nonlymphocytic type, sulfur mustard)

  • Reproductive dysfunction (genotoxicity, mutagenicity, etc.; mustard agents)

There was insufficient evidence found to demonstrate a causal relationship between exposure and the following health conditions:

  • Gastrointestinal diseases

  • Hematologic diseases

  • Neurological diseases

  • Reproductive dysfunction (Lewisite)

  • Cardiovascular diseases (Except for those that may result from serious infections shortly following exposure—heart disease resulting from rheumatic fever, for example.)

(IOM, 1993, pp. 4-5)

IOM Report Recommendations

The committee recommends that the Department of Veterans Affairs (VA) institute a program to identify each human subject in the WWII testing programs (chamber and field tests, and to the degree possible, patch tests), so that these individuals can be notified of their exposures and the likely health risks associated with those exposures. Further, all subjects so identified, if still living, should be medically evaluated and followed by the VA as to their health status in the future. These individuals should also, if they request it, be treated by the VA for any exposure-related health problems discovered. Morbidity and mortality studies should be performed by the VA, comparing chamber, field, and patch test cohorts to appropriate control groups, in order to resolve some of the remaining questions about the health risks associated with exposure to these agents.

(IOM, 1993, p. 5)

The committee recommends that careful attention be paid by health care providers to the special problems and concerns of the affected veterans and their families. This attention may include the convening of a special task force of experts in stress disorders and risk perception to aid the VA, further than this committee is able, in the establishment of comprehensive guidelines for handling of these cases.

(IOM, 1993, pp. 6-7)

The committee additionally recommends that the Department of Defense (DoD) should use all means at its disposal, including public channels, to identify former chemical warfare production workers (military or civilian) and individuals exposed to mustard agents or Lewisite from gas handling, training, the Bari Harbor disaster, or other circumstances. Records of former military personnel could be turned over to the VA for notification, inclusion in morbidity and mortality studies, and health status evaluation. Records of the civilian personnel should be used by the DoD to advise former workers as to their health risks and options for seeking appropriate compensation for any illnesses that resulted from their exposures.

(IOM, 1993, p. 7)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The committee recommends that the VA and DoD publicly announce and widely advertise that personnel exposed to mustard agents or Lewisite during their service are released from any oath of secrecy taken at the time. In addition, professional educational materials should be prepared by the VA or DoD, or both, and made available for physicians who may be treating affected individuals. These materials should incorporate the latest information regarding the long-term health effects of exposure to mustard agents and Lewisite.

(IOM, 1993, p. 8)


After review of the IOM report, VA issued a proposed rule on January 24, 1994 (VA, 1994a). The proposed amendment to VA adjudication regulations was based on the NAS study of the long-term health effects of exposure to mustard agents and lewisite. The NAS study, commissioned by VA, indicated that there is a relationship between mustard agent and lewisite exposure and subsequent development of certain medical conditions. The intended effect of the proposed rule was to expand compensation eligibility.

The NAS study (IOM, 1993) substantiated VA’s prior determination that a casual relationship existed between mustard gas exposure and the resultant disease development. Further, the original seven diseases associated with this exposure were confirmed. However, the NAS study revealed numerous other medical effects as indicated above. VA accepted the additional medical conditions indicated by the NAS and included them in the proposed rule. A few minor exceptions were noted. Mesothelioma was excluded from the lung cancer grouping because it has only been associated with asbestos exposure. The term skin cancer was considered to be too broad a designation, and VA chose to include only squamous cell carcinomas of the skin. Likewise, VA indicated that, in their opinion,

… there is no reason to establish presumptive service connection for “pigmentation abnormalities of the skin” because these abnormalities would be obvious from the time of the exposure to vesicant agents rather than occurring many years after exposure, as in the case of cancer. Also, because the usual places for mustard gas burns are areas of the body which are not visible, i.e., moist areas of the body such as the groin and axilla, rather than exposed areas as in the case of sunburn, most pigmentation abnormalities resulting from these burns would not be considered disabling, unless they interfered with the veteran’s ability to function…. Since compensation is only payable for a disability resulting from an injury suffered or disease contracted in line of duty or from aggravation of a preexisting injury or disease contracted in line of duty [Basic Entitlement. 2006. 38 USC 1110; Basic Entitlement. 2005. 38 USC 1121; Basic Entitlement. 2005. 38 USC 1131; Deaths Entitling Survivors to Dependency and Indemnity Compensation. 2005. 38 USC 1310.], and since exposure to vesicant agents does not cause a type of pigmentation abnormality which is disabling, we do not propose to include pigmentation abnormalities of the skin in the regulation. However, we propose to include scar formation in the regulation.

(VA, 1994a, p. 3)


Because the NAS study reported that data demonstrated a “suggestive” causal or “insufficient” causal relationship for certain health outcomes, VA stated the following in its proposed rule:

NAS found that the evidence was “suggestive” of a causal relationship between exposure to mustard gas and reproductive dysfunction (genotoxicity, mutagenicity, etc.) and expo-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

sure to sulfur mustard and leukemia. NAS found insufficient evidence of a causal relationship between exposure to mustard gas and gastrointestinal diseases, hematologic diseases, neurological diseases, cardiovascular diseases, and for reproductive dysfunction as a result of exposure to Lewisite. As NAS itself indicates, further study in these areas is necessary and in our judgment, the scientific and medical evidence on the whole does not support the establishment of presumptions for these conditions.

(VA, 1994a, p. 4)


The NAS report further did not address any condition of exposure other than whole-body exposure conditions. Therefore, VA did not include any veterans who were engaged in patch or drop testing of vesicants in the proposed rule. VA’s rationale was that since the NAS report did not address the issue of patch or drop testing, then the only conclusion that can be drawn is that NAS findings regarding specific diseases are linked only to full-body exposure (VA, 1994a).

Lastly, the proposed rule addressed lewisite as well as mustard gas. Lewisite exposure was not included in the final rule issued July 31, 1992:

The current regulation applies only to those veterans exposed while participating in secret tests of protective equipment during World War II; we propose to expand it to cover any verified full-body exposure during military service, which will allow veterans exposed to mustard gas under battlefield conditions in World War I, those present at the German air raid on the harbor of Bari, Italy, in World War II, and those engaged in manufacturing and handling vesicant agents during their military service to be eligible for consideration under this regulation.

(VA, 1994a, p. 4)


After public comment on the proposed rule, VA issued the final rule on August 18, 1994:

The regulation published on July 31, 1992, applied only to those veterans who experienced full-body exposure to mustard gas while participating in secret tests of protective equipment during World War II. This amendment expands that regulation to cover any full-body exposure to mustard gas or Lewisite during military service, and it now applies to veterans exposed under battlefield conditions in World War I, those present at the German air raid on the harbor of Bari, Italy, in World War II, those engaged in manufacturing and handling vesicant agents during their military service, etc. By expanding the number of conditions, vesicant agents, and veterans covered, this amendment clearly represents a significant liberalization of the previous criteria.

(VA, 1994b, p. 4)


VA again expressed a concern about the secrecy issue covering the government’s work on mustard gas and lewisite:

It is unquestionably beyond VA’s ability to modify historical events by regulation; however, we believe that this regulation is an appropriate government response to these issues. VA recognizes that because the tests were secret and no follow-up examinations were conducted, veterans who took part in them are at a disadvantage when attempting to

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

establish entitlement to compensation. This regulation addresses that situation by establishing a regulatory framework which recognizes that specific conditions are likely to result from exposure to vesicant agents and relieves veterans of the burden of submitting evidence to establish those associations in individual claims.

(VA, 1994b, p. 5)


As a result of the issuance of the final rule, 38 CFR § 3.316 (Claims Based on Chronic Effects of Exposure to Mustard Gas. 2006), currently reads as follows:

  1. Except as provided in paragraph (b) of this section, exposure to the specified vesicant agents during active military service under the circumstances described below together with the subsequent development of any of the indicated conditions is sufficient to establish service-connection for that condition:

    1. Full-body exposure to nitrogen or sulfur mustard during active military service together with the subsequent development of chronic conjunctivitis, keratitis, corneal opacities, scar formation, or the following cancers: Nasopharyngeal; laryngeal; lung (except mesothelioma); or squamous cell carcinoma of the skin.

    2. Full-body exposure to nitrogen or sulfur mustard or Lewisite during active military service together with the subsequent development of a chronic form of laryngitis, bronchitis, emphysema, asthma or chronic obstructive pulmonary disease.

    3. Full-body exposure to nitrogen mustard during active military service together with the subsequent development of acute nonlymphocytic leukemia. (b) Service connection will not be established under this section if the claimed condition is due to the veteran’s own willful misconduct (See Sec. 3.301[c]) or there is affirmative evidence that establishes a nonservice-related supervening condition or event as the cause of the claimed condition (See Sec. 3.303).

(Claims Based on Chronic Effects of Exposure to Mustard Gas. 2006. 38 CFR 3.316)

Analysis of Presumptive Decisions

The final rule published in 1992 (VA, 1992b) was too restrictive, both from a standpoint of the personnel covered by the rule and the health effects considered by the rule. VA addressed only World War II veterans who were experimentally exposed to mustard gas during full-body, field, or chamber tests of protective equipment. The seven health effects and the restricted personnel inclusion were based on a VA review process only. The presumptions used were based on the fact that the World War II tests were classified (minimally Secret), the “volunteers” were directed not to discuss their involvement with any of the studies, medical service records were not maintained or were not available because of the secret classification, and consequently no long-term medical examinations were conducted. Because of these stated reasons some of the human volunteers who experienced adverse health effects probably did not file for any VA assistance, and those who did file would have experienced great difficulty in establishing service connection.

The final rule published in August 1994 (VA, 1994b) greatly expanded the scope of effects and the population of workers covered. Based on the NAS report (IOM, 1993), the regulation was amended to cover any verified full-body exposure to mustard gas or lewisite during military service and included veterans from World War I exposed during battlefield conditions. This rule

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

also significantly increased the number of adverse health effects which would be covered as a result of full-body exposure to mustard gas or lewisite.

The significantly higher number of medical conditions to be covered and the increased number of Service members to be included for disability coverage indicated that VA was sensitive to veterans’ attempts to obtain VA assistance. Because of the secrecy classification applied to the veterans who worked on the projects, and subsequently to their medical service records, the usual methods of establishing eligibility were deemed to be too great a burden for the veteran. Consequently, the presumptions used were to the benefit of the veteran. VA was attempting to be extremely sensitive to the needs of the veteran in lieu of the strict classification assignment. A fine line must be addressed for national security measures and the health needs of the personnel assigned to work on these national security measures. As indicated in the IOM report, there is still hesitancy in declassifying many of the study results (IOM, 1993). The IOM committee addressing this issue was not given full cooperation in retrieving documents from these studies while the DoD stated that they were not declassified.

Costs could not be fully addressed in this case study because no financial information was received from VA. In the July 1992 rule (VA, 1992b) VA did indicate that there would be no significant financial impact on the amended regulation for compensation. The cost of the amendment would not have an annual effect on the economy of $100 million or more, it would not cause a major increase in costs or prices, and it would not have significant adverse effects on other economic parameters. The August 1994 amendment (VA, 1994b) also indicated no adversity in economic impact. However, VA has not presented IOM with any dollar values for either amendment period.

In view of the current classification scheme based upon national security concerns, those individuals working in classified studies remain potentially at risk. Only after a few individuals involved in the World War II studies attempted to obtain VA assistance was VA able to implement disability assistance. Failure to declassify study results up to 50 years after exposure puts our veterans at considerable risk of disease manifestation and progression without possible warranted screening and therapy. Although project classification may be necessary for DoD, a method should be developed and implemented to account for medical intervention in such cases. The absence of medical service records, no long-term medical or epidemiologic follow-up, and the enforced requirement for secrecy hampered veterans and VA from providing the necessary assistance to the veteran or the beneficiary. Future projects must include a means to ensure medical care is afforded to the participants if needed. A thorough communication path and cooperative working arrangements need to be explored and developed between DoD and VA to ensure that such a lack of medical attention does not occur in the future. If secrecy is mandatory to the level exhibited by the current mustard gas and lewisite rule, then future study participants will be hindered as much as past participants. This issue was examined in great detail in the report, Veterans at Risk, and recommendations were provided (IOM, 1993). The committee found that the secrecy coverage still existed some 50 years after the World War II programs were initiated. Even though some documents were declassified, many of them remained “restricted” and consequently are not releasable to the public. Additionally, not all facilities freely cooperated with IOM in releasing documents. Some did not release documents until the committee work had essentially been completed.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Classification and Secrecy Requirements

The oath of secrecy is still held inviolate by many of the veterans involved in the World War II mustard gas and lewisite studies. These military and civilian workers took an oath to keep all information involving these studies secret, and they have maintained that oath to this day. Even though this oath has prevented possibly thousands of individuals from obtaining necessary medical care, the oath has not been violated. They took a secrecy oath and legally and ethically then maintained their oath. However, the government that they are protecting has not reciprocated in protecting them. The lack of medical records and long-term medical care for past exposures to mustard gas and/or lewisite has not been provided to all military and civilian workers, ostensibly due to the secrecy clause. Classification is mandatory to preserve national security. Consequences of this same classification can be paramount as is evidenced in the mustard gas and lewisite exposure issue. An appropriate resolution between DoD and VA needs to be made to protect both the nation and the nation’s governmental workers.

There are three entities involved in the secrecy issue: DoD scientific and military organizations, VA disability/compensation programs, and the individual Service member. At times, DoD must undertake secret studies to obtain results that will be used to provide the best security possible for our country. To provide the maximum protection, these studies may require a classification level. Many of the studies being conducted today are appropriately under this mantle of classification. On the other hand, VA must provide assistance to the veterans who have provided a necessary service to the government and the country. VA cannot provide this assistance if the necessary information is not given to them to ensure that entitled individuals are tracked, medically treated, and compensated appropriately. Furthermore, the exposures and agents must be identified so that health consequences can be linked. The individual workers must be given the ability to seek VA assistance as the need arises. Using the mantle of secrecy to protect the country is a dire hindrance to the veteran in seeking the necessary medical assistance. Policy considerations may be involved in both the roles played by DoD and VA. Both agencies have requirements and responsibilities that must be met.

The IOM report Veterans at Risk (1993) provided three recommendations addressing this issue:

The committee additionally recommends that the Department of Defense (DoD) should use all means at its disposal, including public channels, to identify former chemical warfare production workers (military or civilian) and individuals exposed to mustard agents or Lewisite from gas handling, training, the Bari harbor disaster, or other circumstances. Records of former military personnel could be turned over to the VA for notification, inclusion in morbidity and mortality studies, and health status evaluation. Records of the civilian personnel should be used by the DoD to advise former workers as to their health risks and options for seeking appropriate compensation for any illnesses that resulted from their exposures.

(IOM, 1993, p. 7)


The IOM committee additionally recommended that the secrecy requirement for workers involved in the World War I and World War II mustard gas and lewisite studies be deleted so that the individuals may seek medical assistance without the threat of punishment for violating the secrecy oath:

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The committee recommends that the VA and DoD publicly announce and widely advertise that personnel exposed to mustard agents or Lewisite during their service are released from any oath of secrecy taken at the time. In addition, professional educational materials should be prepared by the DoD or VA, or both, and made available for physicians who may be treating affected individuals. These materials should incorporate the latest information regarding the long-term health effects of exposure to mustard agents and Lewisite.

(IOM, 1993, p. 8)


The VA final rule issued in August 1994 (VA, 1994b) addressed the disease and eligibility issues for veterans. DoD’s response to the IOM committee recommendations has not been fully ascertained.

VHA Under Secretary for Health’s Information Letters

Since the IOM report was issued in 1993, three information letters have been released by the VHA Under Secretary for Health as well as a veteran health initiative. The Under Secretary for Health’s information letter IL 10-2002-016 (VA, 2002) provided information to clinicians who might encounter and treat veterans involved in Project Shipboard Hazard and Defense. While this letter primarily addressed biological agents, there is a section addressing chemical agents as well. Reference is made to reports produced by IOM. The Under Secretary for Health’s information letter IL 10-2005-004 (VA, 2005) provided information to clinicians who might treat veterans who had been exposed to mustard or lewisite chemical warfare agents as part of human experimentation conducted by DoD through World War II. Extensive reference is made to the IOM report (1993). This letter indicates that because of the secrecy issue related to these studies, VA health-care providers were not aware of the history of these chemicals being used. Guidance was presented for the health-care providers to consider the diseases covered by the presumptive decision rule and emphasized that medical care should focus on the current health status of the veteran. Further, direction was given, emphasizing that there is no test available to confirm exposure to these compounds that occurred decades ago. The Under Secretary for Health’s information letter IL 10-2006-010 (VA, 2006a) again provided information to clinicians on this issue. This letter stated that the Veterans Benefits Administration released notification letters on June 30, 2006, to DoD-identified veterans who were test subjects in military experiments and advised the veteran of benefits they might be entitled to because of their participatory exposures. Again the diseases identified in the final rule for mustard and lewisite agent exposure as causative were discussed and references to websites for additional information were provided. Chemical Warfare Agent Experiments Among U.S. Service Members is a report that describes more chemicals evaluated at Edgewood-Aberdeen than the mustard and lewisite agents (VA, 2006b).

In October 2003, a veterans health initiative was released as an independent study course (VA, 2003). The veterans health initiative, titled Health Effects from Chemical, Biological and Radiological Weapons, contained a forwarding message from the VA Under Secretary for Health, Robert H. Roswell, M.D., stating “greater general awareness of the specialized health issues facing persons with [Chemical, Biological and Radiological] injuries is needed to assure therapeutically appropriate clinical processes” (VA, 2003, p. ii). Dr. Roswell emphasized that every clinician must complete the study course in order to better serve the health needs of the veteran.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Lessons Learned

A number of concerns are raised by this presumptive decision. Foremost is the issue of secrecy surrounding military and governmental studies involving warfare agents. Classifying warfare studies based on national security is necessary in many cases. However, this classification can lead to concerns about health that might not be resolved for decades. In the case of mustard and lewisite agents national security took precedence over identifying the long-term health consequences of exposure. Volunteers who participated in these studies maintained their secrecy oaths for decades even though health issues arose. When these issues were brought to the attention of VA a second problem arose: the lack of definitive medical records for these individuals. Medical records that did exist did not contain any information relating to potential mustard gas or lewisite exposures. Consequently, health-care providers were unaware of these health-related issues. Long-term medical follow-up therefore was not provided to the study participants. A third area of concern is that this classification precluded health-care providers from being aware that symptoms of mustard gas or lewisite exposure might be exhibited by patients. As a result, the health-care providers did not look for these occupationally related symptoms.

Several Under Secretary for Health information letters have been issued by VA to address this gap. As indicated in the presidential task force 2003 report (President’s Task Force to Improve Health Care Delivery for Our Nation’s Veterans, 2003) and a recent GAO report (GAO, 2006a), the health of our veterans must be considered for the entire time of an individual’s military service. The presidential task force specifically cited the lack of exposure data to a known environmental hazard as a root cause for not being able to determine compensatory issues for our veterans (GAO, 2006a,b). While DoD apparently is addressing the tracking and recording of the Service members’ movements to link with exposure data collected, both areas remain concerns for the veteran. Inadequate exposure data are collected, and the ability to vector these data to location and troop movement has limited attribution of disease to exposure agents for individual veterans. Exposure assessment is the key to disability benefits for veterans. A recent report (GAO, 2004) indicates that force health protection and surveillance policy are not as good as they could be, but are improving, especially with more current deployments. The final rule for mustard gas and lewisite (VA, 1994b), based upon the 1993 IOM report, outlined three categories of causal relationships for health consequences of exposure: indicated, suggested, and insufficient evidence of a relationship. VA acknowledged in the final rule that certain health consequences could be linked directly to mustard gas or lewisite exposure and that a second grouping of health consequences had a suggestive linkage based upon the 1993 IOM report. Although the IOM report recommended many diseases that could be associated with exposure to these agents, VA specifically eliminated several of the diseases as not being related to mustard gas or lewisite exposure. Consequently, a precedent was established by VA for causal health consequences associated with specific chemical agents as recommended by IOM. Lastly, this case study indicates an apparent lack of communication and coordination between DoD and VA regarding individual Service members and government workers involved in studies, chemical agents used in these studies, and any actual or potential exposure data for the individuals involved. This lack of coordination was emphatically pointed out in the IOM 1993 report Veterans at Risk.

References

Alexander, S. F. 1947. Medical report of the Bari Harbor mustard casualties. The Military Surgeon 101(1):2-17.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Balali-Mood, M., M. Hefazi, M. Mahmoudi, I. Jalali, D. Attaran, M. Maleki, M.-R. E. Razavi, G. Zare, M.-R. Jaafari, and A. Tabatabaee. 2005. Evaluation of delayed toxic effects of sulfur mustard poisoning in severely intoxicated Iranian veterans: A cross-sectional study. Journal of Medical, Chemical, Biological, and Radiological Defense 3:1-32.

GAO (Government Accountability Office). 2004. Report to Congressional requesters. Defense health care: Force health protection and surveillance policy compliance was mixed, but appears better for recent deployments. GAO-05-120. Washington, DC: Government Accountability Office. http://www.gao.gov/new.items/d05120.pdf (accessed March 19, 2007).

GAO. 2006a. Military disability evaluation: Ensuring consistent and timely outcomes for reserve and active duty members. Testimony before the Subcommittee on Military Personnel, Committee on Armed Services, House of Representatives. GAO-06-561T. Washington, DC: United States Government Accountability Office. http://www.gao.gov/new.items/d06561t.pdf (accessed February 28, 2007).

GAO. 2006b. VA and DOD health care: Efforts to provide seamless transition of care for OEF and OIF servicemembers and veterans. GAO-06-794R. Washington, DC: United States Government Accountability Office. http://www.gao.gov/htext/d06794r.html (accessed July 9, 2007).

Gilchrist, H. L. 1928. A comparative study of World War casualties from gas and other weapons. Washington, DC: U.S. Government Printing Office.

Harris, R., and J. Paxman. 1982. A higher form of killing: The secret story of chemical and biological warfare. New York: Hill and Wang.

Harris, S. H. 1994. Factories of death: Japanese biological warfare, 1932-45, and the American coverup. New York: Routledge.

IOM (Institution of Medicine). 1993. Veterans at risk: The health effects of mustard gas and lewisite. Washington, DC: National Academy Press.

Joy, R. J. T. 1997. Historical aspects of medical defense against chemical warfare. In Textbook of military medicine: Medical aspects of chemical and biological warfare, edited by F. R. Sidell, E. T. Takafuji, and D. R. Franz. Washington, DC: Surgeon General United States Army. Pp. 87-109. http://www.bordeninstitute.army.mil/published_volumes/chemBio/Ch3.pdf (accessed June 29, 2007).

President’s Task Force to Improve Health Care Delivery for our Nation’s Veterans. 2003. A brief guide to the final report. Arlington, VA: President’s Task Force to Improve Health Care Delivery for our Nation’s Veterans. http://veterans.house.gov/spotlight/ShortTaskForceReport.pdf (accessed July 11, 2007).

Reminick, G. 2001. Nightmare in Bari. The World War II Liberty Ship poison gas disaster and cover-up. El Cerrito, CA: Glencannon Press.

Shoham, D. 1998. Chemical and biological weapons in Egypt. The Nonproliferation Review (Spring-Summer):48-58. http://cns.miis.edu/pubs/npr/vol05/53/shoham53.pdf (accessed March 20, 2007).

Sidell, F. R., J. S. Urbanetti, W. J. Smith, and C. G. Hurst. 1997. Vesicants. In Textbook of military medicine: Medical aspects of chemical and biological warfare, edited by F. R. Sidell, E. T. Takafuji, and D. R. Franz. Washington, DC: Surgeon General United States Army. Pp. 197-228. http://www.bordeninstitute.army.mil/published_volumes/chemBio/Ch7.pdf (accessed July 2, 2007).

Tucker, J. B. 2006. War of nerves. Chemical warfare from World War I to Al-Qaeda. New York: Pantheon Books.

VA (Department of Veterans Affairs). 1992a. Claims based on chronic effects of exposure to mustard gas. Proposed rule. January 15, 1992. Federal Register 57(10):1699-1700.

VA. 1992b. Claims based on chronic effects of exposure to mustard gas. Final rule. Federal Register 57(148):33875-33877.

VA. 1994a. Claims based on chronic effects of exposure to vesicant agents. Proposed rule. Federal Register 59(15):3532-3534.

VA. 1994b. Claims based on chronic effects of exposure to mustard gas or lewisite. Final rule. Federal Register 59(159):42497-42500.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

VA. 2002. Possible occupational health exposures of veterans involved in Project SHAD tests. Veterans Health Administration and Under Secretary for Health information letter. Il-10-2002-016. Washington, DC: Department of Veterans Affairs. http://www1.va.gov/shad/docs/IL_10-2002-016.pdf (accessed February 28, 2007).

VA. 2003. Veterans Health Initiative: Health effects from chemical, biological, and radiological weapons. Washington, DC: Department of Veterans Affairs. http://www1.va.gov/vhi/docs/CBR_www.pdf (accessed March 19, 2007).

VA. 2005. Health effects among veterans exposed to mustard and lewisite chemical warfare agents. Under Secretary for Health information letter, IL 10-2005-004. http://www1.va.gov/environagents/docs/USHInfoLetterIL10-2005-004_March_14_2005.pdf (accessed July 2, 2007).

VA. 2006a. Potential health effects among veterans involved in military chemical warfare agent experiments conducted from 1955 to 1975. Under Secretary for Health information letter, Il 10-2006-010. Washington, DC: Department of Veterans Affairs. http://www1.va.gov/environagents/docs/USHInfoLetterIL10-2006-010.pdf (accessed February 28, 2007).

VA. 2006b. Chemical warfare agent experiments among U.S. service members. http://www.va.gov/EnvironAgents/docs/Fact_Sheet_Edgewoodaberdeen_Chemical_Agent_Experiments_Information_paper.pdf (accessed March 20, 2007).

CASE STUDY 7:
GULF WAR PRESUMPTIONS

Introduction

This case study examines the legislative and scientific history for Gulf War presumptions. This analysis illustrates the challenges and implications of establishing service-connected presumptions for health outcomes when exposure data are limited (or unavailable) but environmental agents present in theater may pose potential health concerns. The case study also addresses a syndrome defined by symptoms, rather than a specific illness linked to a specific etiological factor.

Background

Although the duration of combat engagement during the first Gulf War was measured only in days, it also included a protracted aerial campaign of many weeks and ultimately involved nearly 700,000 U.S. troops (IOM, 2006b, p. 11). As a result, while the first Gulf War’s time span may have been relatively brief, its scope and scale affected many U.S. military personnel. It also created a long list of health concerns in Gulf War veterans who believed they had been exposed to biological and chemical agents during Gulf War service that might have adversely affected their health. Veterans from the first Gulf War returned home with a constellation of symptoms that were initially termed Gulf War syndrome. However, the diversity of symptoms did not cluster into a specific group such that it could be defined as a syndrome and has been modified to the descriptions of “unexplained illnesses” or “undiagnosed illnesses.”

The Gulf War presumption process was heavily influenced by the Agent Orange presumption history. The Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.) represented the culmination of repeated efforts to create a comprehensive presumption decision-making process. This process proposed an integration of the VA policy response to complaints of veterans being made to Congress and to a scientific evidence review performed by the IOM. In many ways, the Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.) was the example

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

for the Gulf War legislation which provided the guidance for a review process examining agents of interest in light of Gulf War illnesses. The history of the government’s sluggish response to radiation and herbicide exposure concerns played a role in the establishment of Congress’ Gulf War presumptions for undiagnosed illnesses and chronic multisymptom illness. Following Congress’ experiences with establishing radiation and Agent Orange legislation in the 1980s and early 1990s, Congress did not want to wait to lend aid to the first Gulf War veterans.

A Review of the Scientific and Legislative Landmarks Related to Gulf War Presumptions

Almost immediately following the first Gulf War, veterans began to complain of numerous adverse health effects that they attributed to service in the Persian Gulf. It became evident that the Persian Gulf veterans’ health complaints differed significantly from those voiced by Vietnam veterans. Vietnam veterans ascribed many of their medical woes to one agent—Agent Orange—and claimed that the numerous well-defined illnesses from which they suffered, ranging from skin and liver illnesses to a variety of cancers, could be attributed to exposure to Agent Orange. Persian Gulf War veterans, on the other hand, appeared to be suffering from syndromes and symptoms rather than established illnesses. They were concerned about exposures to a host of toxic environmental agents that were present during the Persian Gulf War. VA claimed that many of the exposures experienced by Persian Gulf War veterans were no different from those experienced by civilians in the United States and for the great majority of veterans, the magnitude of the exposures was small and limited to a short period of time (Brown, 2005).

Congress responded by authorizing VA to provide health-care services on a priority basis (Priority VA Health Care for Persian Gulf Veterans. 1993. Public Law 103-210. 103d Cong., 1st Sess.). An effort to gather data about the nature of the illnesses and symptoms that veterans claimed to have, as well as to conduct research on environmental exposures that occurred during the war was started. In 1994, Congress passed the Veterans’ Benefits Improvements Act of 1994 (Public Law 103-446. 103d Cong., 2d Sess., codified as 38 USC § 1117) which authorized VA to compensate veterans for certain chronic disabilities and illnesses that cannot be attributed to any known clinical diagnosis. Originally, these adverse health effects had to manifest within 2 years of service in the Persian Gulf for a veteran to receive compensation. This presumptive interval has, however, been extended a number of times, most recently in December 2006, so that compensation is provided to the veteran if the disability manifests itself by December 31, 2011 (VA, 2006).

In 1998, Congress passed the Persian Gulf War Veterans Act of 1998 (Public Law 105-277, 105th Cong., 2d Sess., codified, in part, as Presumptions of Service Connection for Illnesses Associated with Service in the Persian Gulf during the Persian Gulf War. 2006. 38 USC § 1118) and the Veterans Programs Enhancement Act of 1998 (Public Law 105-368. 105th Cong., 2d Sess.). These acts, which used the Vietnam War Agent Orange legislation as a starting point, set up procedures for establishing presumptions of service connection for diseases incurred by Gulf War Veterans. As in the Agent Orange legislation, “a positive association” is required between an environmental exposure associated with service in the Persian Gulf War and “the occurrence of a diagnosed or undiagnosed illness in humans or animals” (Presumptions of Service Connection for Illnesses Associated with Service in the Persian Gulf During the Persian Gulf War. 2006. 38 USC § 1118[b][1][B][ii]). In addition, the Gulf War Act stated that an association “shall be considered to be positive … if the credible evidence for the association is equal to or outweighs the credible evidence against the association” (Presumptions of Service Connection for Illnesses

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Associated with Service in the Persian Gulf During the Persian Gulf War. 2006. 38 USC § 1118 [b][3]). The same scientific review process was established as for Agent Orange; an agreement was made with the NAS to set up a committee that was charged with answering the same three questions previously asked with regard to Agent Orange (Persian Gulf War Veterans Act of 1998. Public Law 105-277. 105th Cong., 2d Sess. § 1603). However, as discussed later in this case study, the charges for the Gulf War committees were not identical and addressed many different potential etiological agents of interest. As with the Agent Orange reports, each IOM committee was to provide a report to VA, and the VA Secretary was to determine whether a presumption should be made for specific health outcomes and then publish the decision(s) in the Federal Register. IOM has issued several volumes of Gulf War reports (IOM, 2000, 2003, 2005, 2006b, 2007). VA published a Federal Register notice (VA, 2001a) following receipt of IOM’s report for Gulf War Volume 1 (IOM, 2000). However, Federal Register notices and determination of presumptions for the Gulf War have yet to be published by VA following the receipt of the remaining IOM Gulf War reports.

In 2001, the VA Secretary “determined that there [was] no basis to establish a presumption of service connection for any disease based on service in the Persian Gulf during the Persian Gulf War” (VA, 2001b, p. 35702). This determination was based, in part, on review of the first Gulf War and Health report (IOM, 2000) which reviewed the evidence regarding associations between health outcomes and exposures experienced during the Persian Gulf War, including depleted uranium, pyridostigmine bromide (PB), sarin, and vaccines (anthrax, botulinum toxoid, and multiple vaccines) (IOM, 2000). The VA Secretary determined that “neither the acute and transient symptoms resulting from possible sarin exposure, nor any long-term health consequences associated with possible sarin exposure, warrant a presumption of service-connection” (VA, 2001b, p. 35708). For PB, the VA Secretary concluded that “these acute effects [of PB, as reported in IOM, 2000] were not in the nature of an illness within the contemplation of the governing statute” and that “such effects failed to meet the standards for establishment of presumptive service connection based on exposure to PB” (VA, 2001b, p. 35708). Regarding long-term adverse health effects based on exposure to PB, the VA Secretary determined that “the credible evidence against an association between long-term adverse health effects and PB outweighs the credible evidence for such an association” (VA, 2001b, p. 35708). A similar conclusion was reached for exposure to vaccines and transient acute local and systemic effects. The VA Secretary concluded that “such [acute] effects failed to meet the standards for establishment of presumptive service connection based on anthrax vaccination, botulinum toxoid vaccination or multiple vaccinations” (VA, 2001b, p. 35710).

The Presumption for “Undiagnosed Illnesses” in 1995

The initial presumptive legislation for the Gulf War was in response to numerous symptoms experienced by returning Gulf War veterans. The symptoms were not related to a single or even a few organ systems and were not easily explained by a unifying mechanism, such as being attributed to an infectious cause or the result of drinking tainted water, for example. The legislation arose primarily from the concerns expressed by veterans and without the interaction between VA and IOM that occurred previous to the passage of the Agent Orange legislation.

The list of “signs or symptoms which may be manifestations of undiagnosed illness or medically unexplained chronic multisymptom illness” that were considered as associated with undiagnosed illnesses during the Persian Gulf War and that were ultimately presumptively linked by Congress to service in the Persian Gulf via legislation in 1995 “include, but are not limited to

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
  • Fatigue

  • Signs or symptoms involving skin

  • Headache

  • Muscle pain

  • Joint pain

  • Neurologic signs or symptoms

  • Neuropsychological signs or symptoms

  • Signs or symptoms involving the respiratory system (upper or lower)

  • Sleep disturbances

  • Gastrointestinal signs or symptoms

  • Cardiovascular signs or symptoms

  • Abnormal weight loss

  • Menstrual disorders” (VA, 1995, pp. 6665-6666)

The Presumption of Chronic Multisymptom Illness in 2001

In 2001, chronic multisymptom illness was added by Congress to the unexplained illness provision for presumption of disability acquired during Gulf War service (Veterans Education and Benefits Expansion Act of 2001. Public Law 107-103. 107th Cong., 1st Sess.). Several additional health outcomes were included under this grouping. The list of symptom complexes that were considered as associated with chronic multisymptom illness included: chronic fatigue syndrome, fibromyalgia, and irritable bowel syndrome. Although some of these symptoms overlap with the list included in the undiagnosed illness provision, these symptoms were not specifically designated under one or the other grouping but may exist under either. This new terminology added confusion to an already nonspecific nomenclature for the symptoms or conditions qualifying for presumptive coverage of Gulf War veterans.

Additional Legislative Actions for Gulf War Veterans

Between the time of the initial Gulf War presumption legislation of 1995 (VA, 1995) and the addition of the chronic multisymptom illness provision in 2001 (Veterans Education and Benefits Expansion Act of 2001. Public Law 107-103. 107th Cong., 1st Sess.), there were several other legislative actions to address Gulf War veterans’ needs. These legislative actions did not enlarge the domain of diseases or symptoms being considered as presumptively linked to Gulf War service, but rather addressed administrative aspects of the existing legislation, such as the timeframe of the presumptive period. Extension of the presumptive time period has occurred several times. Originally, it was required that a disease manifest itself within 2 years following military service. At the close of 2001, the presumptive time period was extended to 5 years following military service (VA, 2001a). In December 2006, the presumptive time period was again extended so that compensation would be provided if the disability manifested by December 31, 2011 (VA, 2006). VA stated in this Federal Register notice that “this amendment is necessary to extend the presumptive period for qualifying chronic disabilities resulting from undiagnosed illnesses that must become manifest to a compensable degree in order that entitlement for compensation be established. The intended effect of this amendment is to provide consistency in VA adjudication policy and preserve certain rights afforded to Persian Gulf War veterans and ensure fairness for current and future Persian Gulf War veterans” (VA, 2006, p. 75670). VA also stated in this Federal

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Register notice that it had extended the presumptive period for undiagnosed illnesses from December 31, 2001, to December 31, 2006, “based upon ongoing research that would require review by the Secretary” (VA, 2006, p. 75670).

Exposures of Concern

A number of candidate toxicants were mentioned as potential causes for undiagnosed illnesses in the initial legislation of 1994-1995 (Veterans’ Benefits Improvements Act of 1994. Public Law 103-446. 103d Cong., 2d Sess. § 102, codified as 38 USC § 1117). Congress found that U.S. troops in the Gulf War were exposed to “fumes and smoke from military operations, oil well fires, diesel exhaust, paints, pesticides, depleted uranium, infectious agents, investigational drugs and vaccines, and indigenous diseases, and were also given multiple immunizations” (Veterans’ Benefits Improvements Act of 1994. Public Law 103-446. 103d Cong., 2d Sess. § 102). However, these nominated toxicants were not well linked to specific types of symptoms nor was a retrospective exposure assessment performed to estimate exposure intensities or likelihood of exposure potential. Instead several descriptive scenarios were suggested as plausibly leading to opportunities for exposure, at least to some Service members.

In the Persian Gulf War Veterans Act of 1998 (Public Law 105-277. 105th Cong., 2d Sess. § 1602), diseases related to a presumed exposure to “a biological, chemical, or other toxic agent, environmental or wartime hazard, or preventive medicine or vaccine” were added to the description of illnesses presumed to be service connected. This change in the legislative focus moved the presumption from one based upon an outcome (i.e., unexplained illnesses) to one based on an exposure agent (e.g., depleted uranium or PB). This important change resulted in adding potential adverse health outcomes to the presumptions list that might be linked to the action or toxicity of the now “presumed” agent of exposure.

This change in focus was likely influenced by maturation of hypotheses regarding environmental exposures in the Persian Gulf that had only been alluded to in the first presumptive legislation of 1995 (VA, 1995). In addition, the Presidential Advisory Committee on Gulf War Veterans’ Illnesses commissioned by President Clinton issued their report in December of 1996 regarding the nature and risk factors for Gulf War illness. While not specifically focused on presumptions of service connection, the report did raise inadequately characterized environmental exposures and poor troop tracking as areas in need of improvement to avoid future postconflict health concerns (Presidential Advisory Committee on Gulf War Veterans’ Illnesses, 1996). The controversies about such hazards had motivated VA to commission reports by IOM on possible health effects associated with exposures during the Persian Gulf War (IOM, 1999, 2000, 2003, 2005, 2006b, 2007).

IOM Gulf War Reports

IOM reports were initially requested by VA that focused on the group of toxicants that had emerged as possible candidates for causing unexplained illnesses (IOM, 2000, 2003, 2005). Review of the first grouping of toxicants began in late 1998, and a report was published in 2000 (IOM, 2000). This first report reviewed depleted uranium (a modestly radioactive heavy metal used in projectiles and armament), PB (a chemical antidote for cholinergic “nerve” agent weapons), sarin (a chemical nerve agent with cholinergic properties), and vaccines, including that for anthrax (IOM, 2000).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

A second Gulf War report by IOM was completed in 2003 that reviewed insecticides and solvents used by U.S. troops during the Gulf War. The committee was charged by VA “to assess the scientific evidence regarding long-term health effects associated with exposure to specific agents that were potentially present during the Gulf War” (IOM, 2003, p. 2).

The third major Gulf War review by IOM included fuels, combustion products, and propellants as its focus (IOM, 2005). Although driven by a toxicological literature review, it appears that this study was also commissioned by VA to think more broadly about exposures and to inform VA about illnesses (among veterans) that might not be fully appreciated. This committee was to look not only at specific hazards that may have been encountered in the Persian Gulf, but also asked to comment on “the increased risk of illness among people exposed to the putative agents during service in the Persian Gulf” (IOM, 2005, p. 2). This committee’s report was issued in 2005 and was also hoped to offer some prevention strategies for troops deployed to Iraq for the current conflict.

A fourth Gulf War review by IOM, which was commissioned by VA, did not focus on toxicants but rather was directed at determining what peer-reviewed medical literature, taken together, can tell us about the health status of Gulf War veterans (IOM, 2006b). In this instance, VA was interested in an integrated review and asked for the committee’s insight about deployment-related illnesses that might not be fully appreciated. Their report was issued in 2006.

A fifth Gulf War review by IOM was commissioned by VA in 2004 to examine the infectious agent exposures that may have contributed to illness in the deployed Gulf War cohort. The report of the committee was issued in 2007 (IOM, 2007) and identified nine pathogenic agents endemic to the Gulf War theater and capable of causing illnesses with long-term adverse health effects. These agents were listed regardless of whether or not their related illnesses were acutely diagnosed in Gulf War troops. Further refinement of this list would require that the likelihood of infection with a candidate agent during deployment be equal to or greater than the likelihood of infection when not deployed.

Interspersed among these major review activities by IOM were several smaller projects commissioned by VA that addressed more limited topics, such as updated literature reviews for sarin (IOM, 2004) or a review for a specific disease or complaint (IOM, 2006a,c). Two additional studies are in progress—one dealing with PTSD and one with deployment-related stress. Work continues at IOM to provide VA with the most up-to-date research and published studies related to the Gulf War.

A summary of the statements of task as well as the conclusions from each of the IOM Gulf War reports can be found in Appendix H.

The VA Response to IOM Gulf War Reports

Typically, VA responds to IOM reports with policy decisions that it publishes in the Federal Register. To date, only one Federal Register notice (VA, 2001a) has been published by VA following receipt of an IOM Gulf War report (IOM, 2000). Federal Register notices have not yet been published in response to IOM Gulf War reports, volumes 2-5 (IOM, 2003, 2005, 2006b, 2007). To date, presumptions have not been granted by VA for Persian Gulf veterans pursuant to 38 USC § 1118 (Compensation for Service-Connected Disability or Death. 2006. 38 USC § 1118), although the machinery for establishing presumptions is in place. In February 2006, the VA Secretary wrote to leaders of the House and Senate Veterans’ Affairs Committees advising them that the evidence currently available did not warrant the establishment of Persian Gulf presumptions, and that VA would publish notices of its decision in the Federal Register “explaining

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

the basis for that determination” (Nicholson, 2006a,b,c,d). As of the writing of this report, no statement appears in the Federal Register, although such statements are required to be filed within 60 days of the receipt of an IOM report (Persian Gulf War Veterans Act of 1998. Public Law 105-277, 105th Cong., 2d Sess. § 1602). Although the presumptive service-connection mechanism has not been a major pathway for service connection of Persian Gulf veterans, it has been reported that a comparatively large percentage of Persian Gulf veterans have been able to collect compensation through the direct service-connection route (as stated by Pamperin, 2006).

As of now, the Persian Gulf statutes would govern any presumptions dealing with current service in Iraq or any other area in the Persian Gulf. (As defined in 38 USC § 101, the term “Persian Gulf War” means “ the period beginning on August 2, 1990, and ending on the date thereafter prescribed by Presidential proclamation or by law” [Definitions. 2006. 38 USC § 101(33)]).

Cost Implications of Gulf War Presumptions

No cost estimates were made available by VA or published in the Federal Register for this analysis. However, the long-term cost implications of Gulf War legislation are likely to continue to rise as veterans of contemporary conflicts in the Persian Gulf continue to return from Iraq and Afghanistan. During the Committee’s first open session meeting, Mr. Tom Pamperin, a VA representative, stated that 3,259 veterans had been presumptively service-connected for undiagnosed illnesses and that these individuals had typically been rated at 10 percent (Pamperin, 2006). Mr. Pamperin stated that most of the Gulf War veterans who have secured service connection did so via the direct service-connection route (as stated during Pamperin, 2006). In VA’s Federal Register notice that extended the presumptive period for compensation of Gulf War veterans to 2011, VA stated that “VA continues to receive claims for qualifying chronic disabilities. In 2005 for example, VA received 2,241 new claims with diagnostic codes that would be affected by this final rule, and we continue to receive such claims during 2006” (VA, 2006, p. 75671).

Lessons Learned

This case study offers several lessons that are relevant when considering strategies for improving the current presumptive disability decision-making process. These lessons relate to the role of Congress in issuing Gulf War legislation and establishing presumptions, VA in responding to the receipt of IOM reports, and IOM in evaluating and presenting the body of evidence relating to potential Gulf War exposures. In addition, the case study makes clear that closer coordination and collaboration between the DoD and VA are needed with regard to health monitoring and tracking of military personnel and veterans as well as exposure assessment efforts.

The principal participants in the current framework of the presumptive disability decision-making process include veterans and VSOs, VA, Congress, and IOM. Although each of these entities is motivated to assure the well-being of the veteran, there is also a need to balance the granting of benefits against resource constraints. Comparative fairness needs to be maintained by Congress and VA in its handling of each group of veterans. The current presumptive framework for Gulf War illnesses appears to have been strongly driven by time pressures on Congress and VA to respond to the concerns of Gulf War veterans, and to do so more rapidly than took place with radiation and Agent Orange. When a presumption of service connection for a disease or health condition is legislated by Congress (e.g., unexplained illnesses) rather than through the accrual and evaluation of scientific evidence, there is the potential to diminish the credibility of a presumptive decision-making process that is evidence based. A misperception then may arise that the decision was evidence based, even though it was actually driven by other considerations.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Congress

In the politically charged and time-pressured context of responding to Gulf War veterans in the 1990s, it was clear that Congress believed that action needed to be taken quickly. However, the response was not grounded in specific diagnostic tests or in a validated pattern of symptoms that could be linked to exposures in the Gulf War. Rather, a presumption was made for the entity, undiagnosed illnesses, that was based largely around features of numerous individual cases. The affected veterans, many previously robust in their health, experienced inexplicable symptomatology and deterioration of health. Thus, a decision was made by Congress to endorse the veteran’s self-reported symptom complaints as sufficient evidence for documenting the occurrence of a Gulf War service-connected illness.

In establishing undiagnosed illnesses and chronic multisymptom illness, Congress defined medical conditions and health outcomes for the presumptive disability decision-making process for the first time.

There seems to have been little application of exposure assessment in determining eligibility for a service-connected Gulf War condition. This approach to handling affected veterans was largely authored outside of VA by Congress and then handed to VA to administer. The decision to not consider exposure history in determining eligibility for a presumption is similar to the exposure presumption made for Agent Orange, meaning if a Service member set foot in the Vietnam theater they were considered “exposed,” and thus any illnesses linked to that exposure can be presumed to be service connected. In the example of the Gulf War, Congress directed VA to apply an approach with high sensitivity (i.e., toward including all possible claims, but thereby risking a high false positive rate).

Congress used two legislative approaches for veterans of the Gulf War in the absence of exposure data or a unifying case definition of Gulf War illness. The 1995 decision (VA, 1995) to term a list of conditions as presumptive signs or symptoms clustered under the rubric of undiagnosed illnesses permitted medical care and other benefits to be provided to affected veterans by VA. The second approach departed from the initial symptom-based presumption model and in the 1998 Act (Persian Gulf War Veterans Act of 1998. Public Law 105-277. 105th Cong., 2d Sess. § 1602) mandated any additional conditions or symptoms as presumptive that could be linked to “a biological, chemical, or other toxic agent, environmental or wartime hazard, or preventive medicine or vaccine.” This decision added to the menu of potential illnesses that could be considered presumptively linked to service in the Gulf that had not yet been so deemed.

VA

The Gulf War example makes clear the critical need for improved collaboration and cooperation between DoD and VA with respect to health monitoring and tracking of military personnel and veterans as well as improved exposure assessment efforts. Attempts at retrospective reconstruction of troop exposures during military service have identified the gaps in exposure information available to inform decisions about veterans’ health. Indeed, lack of these data has been noted by many IOM committees and by stakeholders grappling with service-connected disability determinations (IOM, 2000, 2003, 2005). There are many barriers, both historical and contemporary, to eliminating such information gaps through research and data collection. Certainly, the most obvious is the potential conflict of mission in collecting exposure data during wartime deployments. On the other hand, since the Gulf War began and perhaps taking into consideration the Agent Orange experience, there has been increased understanding of the need for exposure

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

assessment as an element of force health protection and readiness (DoD, 1997, 2006). The Committee fully understands that exposure assessment and troop tracking are challenging in a fixed environment; the Committee also recognizes the considerable efforts that both DoD and VA will need to make to improve exposure assessment and troop tracking during the chaotic and immediately hazardous environment of a wartime deployment. Nevertheless, talented professionals with technical knowledge, trained in scientific disciplines including engineering, toxicology, industrial hygiene, and epidemiology, regularly perform such assessments in other complex work environments and can be more fully engaged in troop exposure assessment. Indeed, substantial effort is needed before deployment to anticipate and evaluate potential hazards and the fielding of monitoring equipment and control measures that will be needed. The plans of DoD to develop and link health and environmental surveillance data will be an important element in attaining a longitudinal database for such information on every Service member. Applied research in the areas of large surveillance database development and real-time exposure assessment for candidate toxicants is to be encouraged. The commitment of DoD and VA to work more closely in harmonizing the medical records of all Service members to allow a truly seamless transition from DoD to VA health care and follow-up is strongly endorsed by the Committee and should be well resourced by the leadership of both DoD and VA.

VA has now received several IOM reports; however, Federal Register notices have not been published to explain how VA has evaluated the IOM reports and if VA will or will not establish presumptive connections as a result of the IOM reports.

VA continues to extend the presumptive time period for undiagnosed illnesses, and will likely continue to do so because of the current conflict in Iraq.

VA had an apparently limited role in the initial approach to Gulf War presumptions; its role primarily consisted of implementing a policy assigned by Congress in the absence of scientific evidence, a contrast to VA’s actions in subsequent Gulf War-related presumptive decision-making actions. In latter examples, VA has asked IOM to examine adverse health effects rather than the existing “undiagnosed illness” or “chronic diseases” provisions established by Congress with relation to Gulf War service. These reviews have been the basis for subsequent decision-making by VA, which has not yet established any presumptions for the Gulf War.

The Committee was provided with an example of how VA was considering exposure potential to Gulf War veterans in a presentation made by a VA official. The VA official discussed VA’s consideration of an IOM review regarding leukemia and its links to benzene exposure. VA took the position that leukemia developing in a Gulf War veteran was not presumptively service connected. VA explained to the Committee that because benzene is widely encountered in the environment generally, such a presumption of exposure occurring exclusively or primarily in the Persian Gulf could not be made. VA also pointed out that the individual exposure opportunity of the veteran during military service, which could be examined for service-connected benefits on a case-by-case basis, was a reasonable remedy for the veteran for whom exposure during Gulf War service and leukemia development were linked. This would occur via direct service-connection methods.

IOM

The Committee agrees with the inclusion of a category for evidence sufficient to infer causality, “Sufficient Evidence of a Causal Relationship,” in IOM evaluations for the Gulf War and Health series.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

The lack of exposure information for Gulf War veterans was emphasized in the Gulf War and Health series by IOM (2000, 2003, 2005). The inclusion of a specific substance, such as benzene, on a list of candidate toxicants that a veteran may have encountered in the Persian Gulf was based on one of the following qualitative assessments—a listing of solvents potentially used or obtained locally or known constituents of a product used—and not necessarily on actual supply lists or industrial hygiene measurements. Indeed, for just this reason, the vast majority of evidence considered in the IOM Gulf War reports does not come from studies of veterans but of other exposed cohorts described in the published scientific literature (IOM, 2000, 2003, 2005). This lack of exposure and health information from studies on veterans points to the need for exposure estimation and surveillance for health risks.

References

Brown, M. 2005. The role of science in Department of Veterans Affairs disability compensation policies for environmental and occupational illnesses and injuries. Paper presented at the first meeting of the IOM’s Committee on the Evaluation of the Presumptive Disability Decision-Making Process for Veterans, Washington, DC.

DoD (Department of Defense). 1997. Implementation and application of joint medical surveillance for deployments. Instruction number 6490.3. Washington, DC: Department of Defense.

DoD. 2006. Deployment health. Instruction number 6490.03. Washington, DC: Department of Defense.

IOM (Institute of Medicine). 1999. Gulf War veterans: Measuring health. Washington, DC: National Academy Press.

IOM. 2000. Gulf War and health, volume 1: Depleted uranium, pyridostigmine bromide, sarin, vaccines. Washington, DC: National Academy Press.

IOM. 2003. Gulf War and health, volume 2: Insecticides and solvents. Washington, DC: The National Academies Press.

IOM. 2004. Gulf War and health: Updated literature review of sarin. Washington, DC: The National Academies Press.

IOM. 2005. Gulf War and health, volume 3: Fuels, combustion products, and propellants. Washington, DC: The National Academies Press.

IOM. 2006a. Amyotrophic lateral sclerosis in veterans: Review of the scientific literature. Washington, DC: The National Academies Press.

IOM. 2006b. Gulf War and health, volume 4: Health effects of serving in the Gulf War. Washington, DC: The National Academies Press.

IOM. 2006c. Posttraumatic stress disorder: Diagnosis and assessment. Washington, DC: The National Academies Press.

IOM. 2007. Gulf War and health, volume 5: Infectious diseases. Washington, DC: The National Academies Press.

Nicholson, R. J. 2006a. Letter to Larry E. Craig, Chairman, Committee on Veterans’ Affairs, United States Senate. February 24, 2006.

Nicholson, R. J. 2006b. Letter to Steve Buyer, Chairman, Committee on Veterans’ Affairs, U.S. House of Representatives. February 24, 2006.

Nicholson, R. J. 2006c. Letter to Congressman Lane Evans. February 24, 2006.

Nicholson, R. J. 2006d. Letter to Senator Daniel K. Akaka. February 24, 2006.

Pamperin, T. 2006. Presumptions in VA’s disability program—The effect. Paper presented at the first meeting of the IOM’s Committee on the Evaluation of the Presumptive Disability Decision-Making Process for Veterans, Washington, DC.

Presidential Advisory Committee on Gulf War Veterans’ Illnesses. 1996. Final report. Washington, DC: U.S. Government Printing Office.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

VA (Department of Veterans Affairs). 1995. Compensation for certain undiagnosed illnesses. Final rule. Federal Register 60(23):6660-6666.

VA. 2001a. Compensation for certain undiagnosed illnesses. Interim rule with requests for comments. Federal Register 62(82):23138-23139.

VA. 2001b. Illnesses not associated with service in the gulf during the Gulf War. Notice. Federal Register 66(130):35702-35710.

VA. 2006. Extension of the presumptive period for compensation for Gulf War veterans. Interim Final Rule. Federal Register 71(242):75669-75672.

CASE STUDY 8:
AGENT ORANGE AND PROSTATE CANCER PRESUMPTION

This case study examines the 1996 decision by the VA to establish presumptive service connection for prostate cancer based on herbicide exposure among Vietnam veterans. VA’s presumption followed IOM’s Agent Orange committee’s report which assigned a classification of “limited/suggestive evidence” for the association of exposure to Agent Orange and prostate cancer. This case study explores aspects of the development of a presumption for Vietnam veterans. It illustrates the challenges and implications of establishing a service connection for a common chronic condition when exposure data are unavailable and evidence of association is limited.

Background Information on Prostate Cancer

Prostate cancer is among the most common cancers in men. In fact, it is the second highest cause of cancer deaths in U.S. males. Overall prostate cancer age-adjusted death rates for all U.S. males are reasonably stable (27.9/100,000 in 2004) (see http://seer.cancer.gov/csr/1975_2004/results_merged/topic_mor_trends.pdf). There is a marked difference between Caucasians and African Americans—the 2004 death rate for African Americans is more than twice that of Caucasians (62.3 vs. 25.6 per 100,000). Prostate cancer is primarily a disease of older males; the comparable Surveillance, Epidemiology, and End Results data for annual death rates from prostate cancer for men over age 65 is 188.7 per 100,000 while the rate for men younger than 65 is 1.8 per 100,000 (see http://seer.cancer.gov/csr/1975_2004/results_merged/topic_annualrates.pdf). The steep increase in age incidence has led the medical community to the conclusion that if a man lives long enough, he will develop prostate cancer. The prevalence of prostate cancer in the United States is increasing, possibly reflecting earlier diagnosis and better treatment modalities.

The Need for a Presumption

The history of the Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.), establishing a presumption of service connection for diseases associated with herbicide exposure during the Vietnam conflict, as well as the procedure of adding additional diseases to this presumption, has been reviewed elsewhere in this report (see Chapter 2, Chapter 4, and Appendix D). The need for an exposure presumption for Agent Orange-associated diseases is based in part on the difficulty of establishing with certainty the degree of exposure to herbicides among Vietnam veterans (IOM, 1994). The Agent Orange Act eliminates the need for an individual Vietnam veteran to provide evidence of herbicide exposure; exposure is presumed for all Vietnam veterans.

Although the need for a presumption is based primarily on the difficulty of establishing exposure, presumptions also address gaps in the evidence for association. An examination of the

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

exact charge to IOM for determining the diseases that are linked to herbicides is instructive. For each disease, IOM has been asked to determine, to the extent that available data permit meaningful determinations,

  1. whether a statistical association with herbicide exposure exists, taking into account the strength of the scientific evidence and the appropriateness of the statistical and epidemiologic methods used to detect the association;

  2. the increased risk of the disease among those exposed to herbicides during Vietnam service; and

  3. whether there is a plausible biological mechanism or other evidence of a causal relationship between herbicide exposure and the disease.

Each of the IOM reports that examined herbicides and prostate cancer risk (IOM, 1994, 1996, 1999, 2001, 2003, 2005) focused almost exclusively on addressing the task described in charge 1, and each report included additional discussions of the more limited data to address charge 3. Each of the IOM reports address charge 2 but acknowledge the difficulty of providing risk estimates because of the lack of exposure data. IOM (2005) notes that

Although there are data to suggest an association between exposure to the chemicals of interest and prostate cancer, the lack of exposure information on Vietnam veterans precludes quantification of any possible increase in their risk.

(IOM, 2005, p. 282)


The presumption of service connection for prostate cancer in Vietnam veterans therefore serves more than just to address the gap in evidence of exposure. As the lack of exposure data makes quantification of the magnitude of the association between military service and prostate cancer difficult, this presumption also implicitly serves to address a gap in the evidence for association between herbicides and prostate cancer among Vietnam veterans.

Development of the Prostate Cancer Presumption

As a result of Public Law 102-4 (Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess.), IOM signed an agreement with VA to review and summarize the strength of the scientific evidence concerning the association between herbicide exposure during Vietnam service and diseases or conditions that may be associated with this exposure. The first report was issued in 1994 (IOM, 1994) and has been updated on a biennial basis since that time. This section reviews the scientific evidence contained in the IOM reports and the legislative events that resulted from the conclusions of these reports. It is not meant to be a comprehensive analysis of the entire body of scientific literature on this topic, but rather to highlight those events and data that inform the conclusions in the reports and changes in legislation.

Each of the IOM Agent Orange committees’ reviews of prostate cancer has categorized the evidence between Agent Orange and prostate cancer as “limited/suggestive evidence of an association.” The IOM Agent Orange committees’ definition of the category “limited/suggestive evidence of an association” includes an example that “at least one high-quality study shows a positive association, but the results of other studies are inconsistent” (IOM, 1994, 1996, 1999, 2001, 2003, 2005). Each of the IOM Agent Orange committees reviewing prostate cancer found at least one study of agricultural workers exposed to herbicides, in particular phenoxy herbicides, that

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

had statistically significant findings and was deemed to be of high quality. Each IOM report also cites other studies with some supporting evidence of an association between Agent Orange and prostate cancer, although the results of some of the studies are not statistically significant (IOM, 1994, 1996, 1999, 2001, 2003, 2005). The 1994 IOM report states that

One large well-done study in farmers showed an increased risk, and subanalyses in this study indicate that the increased risk is specifically associated with herbicide exposure (OR = 2.2, confidence interval [CI] 1.3-3.8; Morrison et al., 1993).

(IOM, 1994, p. 519)


The IOM Agent Orange committees have tended to rely largely on epidemiologic findings for the evidentiary classifications. For example, the Conclusions section of the 2003 report states that

Strength of Evidence from Epidemiologic Studies

On the basis of its evaluation of the epidemiologic evidence reviewed in this and previous Veterans and Agent Orange reports, the committee finds that there is limited or suggestive evidence of an association between exposure to at least one of the chemicals of interest (2,4-D, 2,4,5-T or its contaminant TCDD, picloram, or cacodylic acid) and prostate cancer. Although the associations are not large, a number of studies provide evidence suggestive of a small increase in morbidity or mortality from prostate cancer. The evidence regarding association is drawn from occupational studies in which subjects were exposed to a variety of pesticides, herbicides, and herbicide components and from studies of Vietnam veterans.

(IOM, 2003, p. 323)

The Extent of Association Reported by the IOM Agent Orange Committees and the Presumptive Decision for Prostate Cancer by VA

The Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.) specifies two categories of diseases that may be presumptively service connected. For one group of specified diseases, a presumption is to be given “unless there is affirmative evidence to establish that the veteran was not exposed to any such agent during that service” (Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess. § 2[a][3]). The act then specifies a second group: “each additional disease (if any) that the VA Secretary determines in regulations prescribed under this section warrants a presumption of service connection by reason of having positive association with exposure to an herbicide agent” (Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess. § 2[a][1][B]; emphasis added). Prostate cancer falls into the second group. The language in the act defines a positive association as one in which “the credible evidence of the association is equal to or outweighs the credible evidence against the association” (Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess., § 2[b][3]). Each of the IOM Agent Orange committees was charged with determining

  1. whether a statistical association with herbicide exposure exists, taking into account the strength of the scientific evidence and the appropriateness of the statistical and epidemiologic methods used to detect the association;

  2. the increased risk of the disease among those exposed to herbicides during Vietnam service; and

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
  1. whether there is a plausible biological mechanism or other evidence of a causal relationship between herbicide exposure and the disease.

In 2007, the Committee has no way to assess how the congressional language (“equal to or outweighs”) and the three points of the charge to the IOM Agent Orange committees were viewed by VA and the IOM Agent Orange committees. We could not specifically determine how these points were considered by the 1994 IOM Agent Orange committee, and the subsequent committees, in their deliberations and their design of the evidence categories chosen by the 1994 committee. We could also not judge the rationale for VA’s translation of the Agent Orange committee’s category of “limited/suggestive” and its evaluation of biological plausibility in response to charge 3 into a 1996 presumptive decision for prostate cancer based upon the congressional language of “equal to or outweighs.”

In 1994, after review of the first IOM Agent Orange report (IOM, 1994), the VA Secretary determined that “the credible evidence against an association between prostate cancer and herbicide exposure outweighs the credible evidence for such an association, and he has determined that a positive association does not exist” (VA, 1994, p. 342). At this time, no presumption of service connection based on exposure to herbicides used in the Republic of Vietnam during the Vietnam era was issued for prostate cancer. Two years later, after the release of the second IOM Agent Orange report (IOM, 1996), the VA Secretary determined that, although prostate cancer remained in the “limited/suggestive” category of evidence, there was “a positive association between herbicide exposure and prostate cancer” (VA, 1996c, p. 41369). VA’s determination was based, in part, on the review of several new occupational studies and veteran studies as reported in the 1996 IOM report (IOM, 1996) and presumptive service connection for prostate cancer was established (VA, 1996d).

Biological Plausibility

The relationship between prostate cancer and endocrine hormones has long been posited, as having the “endocrine-disruptive” effects of dioxins. This hormonal hypothesis implies toxicological mechanisms that might contribute to understanding the plausibility of a causal association. The lack of integration with epidemiological evidence of animal toxicology and dose issues related to biological plausibility to make an overall conclusion regarding the causal evidence for effect distinguishes the IOM approach from those of other national and international agencies considering the causal relationship between dioxins and cancer.

It is usual in assessing the evidence of a causal relationship between an environmental agent and cancer to evaluate its biological plausibility. For example, issues related to potential mechanisms of carcinogenesis are prominently discussed in reviews of possible adverse health consequences of dioxins by the International Agency for Research on Cancer (IARC, 1997), the National Toxicology Program (NTP, 2006), the Agency for Toxic Substances and Disease Registry (ATSDR, 1998), the U.S. Environmental Protection Agency (EPA, 2003), and the NAS (NRC, 2006a). The integration of basic mechanistic considerations, experimentation in laboratory animals, and findings in exposed humans are central to the weight-of-evidence approaches used by these organizations in considering the potential consequences of environmental agents. National Academy committees using this holistic process include those responsible for the various BEIR reports (NRC, 1972, 1979, 1980, 1988, 1990, 1998, 1999b, 2006b) and such documents as Arsenic in Drinking Water (NRC, 1999a).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

As previously stated, the congressional language of “[w]hether there exists a plausible biological mechanism or other evidence of a causal relationship between herbicide exposure and the disease” directs the review of the issue of biological plausibility for each disease evaluated (Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess. § 3[d][C]). The first IOM Agent Orange committee noted the paucity of data on chlorophenoxy herbicides and most of its emphasis was on dioxin in its discussions of biological plausibility. The first IOM Agent Orange committee summarized their review of the carcinogenicity data as follows:

TCDD has been shown to have a wide range of effects in laboratory animals on growth regulation, hormone systems, and other factors associated with the regulation of activities in normal cells. In addition, TCDD has been shown to cause cancer in laboratory animals at a variety of sites. If TCDD has similar effects on cell regulation in humans, it is plausible that it could have an effect on human cancer incidence.

(IOM, 1994, pp. 521-522)


This conclusion that TCDD is a plausible carcinogen was subsequently supported by IARC (IARC, 1997). It should be noted that the epidemiologic data used by the IOM Agent Orange committees to categorize prostate cancer as “limited/suggestive” was due to the evaluation of a study of agricultural workers using phenoxy herbicides contaminated with dioxins. It could be hypothesized that phenoxy herbicides, irrespective of contamination with dioxins, may cause prostate cancer. This hypothesis might explain why studies of industrial workers and the Seveso population which have been heavily exposed to dioxins, but not phenoxy herbicides, have not shown evidence that exposure to dioxins is causally related to prostate cancer. If this hypothesis were correct, then the paragraph quoted above from the 1994 report as well as below from the 1996 report would not provide a biological basis for this association.

The second IOM Agent Orange committee provided more in-depth information about dioxins, in general, as well as a rationale for not considering specific cancers as requested by Congress (IOM, 1996). Based on its review of an extensive toxicological database, the committee stated

Given these data, which establish biological plausibility for cancer in general but not for all specific sites, the committee chose not to summarize biologic plausibility for each cancer reviewed in this chapter. Toxicological data are provided only for a small number of cancer types that have specific, relevant experimental data.

(IOM, 1996, p. 176)


The most recent IOM Agent Orange committee provided a much more detailed review of the toxicological mechanisms underlying the effects of dioxins, including a brief discussion of potential mechanisms related to prostate cancer (IOM, 2005). However, the Committee could not determine how this information was generally considered when classifying the extent of an association between Agent Orange and prostate cancer—which was still based upon review of individual epidemiologic studies (IOM, 2005).

Dose Issues

A common epidemiologic approach to exploring whether there is a causal relationship between chemicals and a specific cancer has been to focus on cohorts with particularly high levels of exposure, usually in an occupational setting. There are a number of such cohorts that have

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

been heavily exposed to dioxins and related compounds. As a corollary, biological markers of effect (when available) can serve as useful indicators of high levels of exposure. For dioxins and related compounds, chloracne is a common health outcome in those most heavily exposed and has been used as an exposure surrogate.

Consideration of comparative doses among published studies does not appear to have played a major role in the deliberations of the IOM Agent Orange committees. For example, the findings of the studies were not arranged by likelihood of high-dose exposures, such as in IARC reviews. The IOM Agent Orange committees give consideration to dose issues within a cohort. For example, evidence in a study shows that a higher risk for prostate cancer is observed in those who estimated to have the largest number of acres sprayed with pesticides.

Congruency of the IOM Process with That of IARC and Other Organizations

The process used by the IOM Agent Orange committees in developing the categorization of the strength of scientific evidence appears to have been adapted from the IARC process. The overall approach differs in that the IOM Agent Orange committees developed findings separately for epidemiological evidence (using criteria adapted from IARC) and biological plausibility based on toxicological and mechanistic studies but do not integrate these, like IARC and other organizations, to make an overall evaluation regarding causality. During the time period in which the IOM Agent Orange committees have conducted their biennial reviews, IARC has also reviewed the evidence concerning dioxins and cancer, as have other organizations including NTP, EPA, and ATSDR. IARC’s 1997 review focused extensively on the experimental literature and an understanding of dose issues (IARC, 1997). Further, the findings in these areas were determinative of the IARC classification of TCDD as a Group 1 carcinogen which is defined as “[t]he agent (mixture) is carcinogenic to humans. The exposure circumstance entails exposures that are carcinogenic to humans” (see http://www.inchem.org/documents/iarc/monoeval/eval.html).

In reaching the determination that TCDD should be classified as a Group 1 carcinogen, IARC particularly relied on mechanistic data and on findings in heavily exposed workgroups. IARC found limited evidence in humans and sufficient evidence in experimental animals for the carcinogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin. This normally would lead to categorizing a compound as a Group 2 carcinogen rather than a Group 1 carcinogen. However, the IARC document explained that

In making the overall evaluation, the Working Group took into consideration the following supporting evidence: (i) 2,3,7,8-TCDD is a multisite carcinogen in experimental animals that has been shown by several lines of evidence to act through a mechanism involving the Ah receptor; (ii) this receptor is highly conserved in an evolutionary sense and functions the same way in humans as in experimental animals; (iii) tissue concentrations are similar both in heavily exposed human populations in which an increased overall cancer risk was observed and in rats exposed to carcinogenic dosage regimens in bioassays.

(IARC, 1997, pp. 8-9)


The approach used by IARC, as well as its focus on dose, lead to cautionary language in considering the epidemiologic evidence:

Overall, the strongest evidence for the carcinogenicity of 2,3,7,8-TCDD is for all cancers combined, rather than for any specific site. The relative risk for all cancers combined in

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

the most highly exposed and longer-latency subcohorts is 1.4. While this relative risk does not appear likely to be explained by confounding, this possibility cannot be excluded. There are few examples of agents which cause an increase in cancers at many sites; examples are smoking and ionizing radiation in the atomic bombing survivors (for which, however, there are clearly elevated risks for certain specific cancer sites). This lack of precedent for a multisite carcinogen without particular sites predominating means that the epidemiologic findings must be treated with caution; on the other hand, the lack of precedent cannot preclude the possibility that in fact 2,3,7,8-TCDD, at high doses does act as a multisite carcinogen. It should be borne in mind that the general population is exposed to levels far lower than those experienced by the industrial populations.

(IARC, 1997, p. 4)


Emphasis on the toxicological database and on dose-response issues is also common to other agencies (e.g., NTP, ATSDR, EPA) reviewing the evidence of adverse health consequences resulting from exposure to dioxins or Agent Orange. The focus by IARC on mechanistic considerations and on cohorts with high levels of exposure as determinative of a cause-and-effect relationship has also been carried through in subsequent reviews. Steenland et al. reviewed the evidence that had accumulated since the 1997 IARC report and summarized additional analyses of the four industrial cohorts whose high-level TCDD exposures had been the focus of the IARC review (Steenland et al., 2004). Steenland et al. also reviewed additional findings in the Seveso cohort and discussed recent toxicological findings supportive of the potential for TCDD carcinogenesis. In the section on new studies they described the Ranch Hand studies of Air Force personnel involved in herbicide spraying, including the increased incidence of malignant melanoma. The authors stated the following regarding elevated prostate cancer findings in these studies:

Akhtar et al. (2004) also found excesses of prostate cancer incidence, but these occurred in both exposed and nonexposed Air Force personnel and may have been due to increased cancer surveillance in both groups; both are subject to intense medical follow-up.

(Steenland et al., 2004, pp. 1266-1267)

Cost Implications of the Presumption for Prostate Cancer

There are limited data regarding the costs to VA associated with the prostate cancer presumption. The commentary on 38 CFR Part 3 includes a brief discussion of the cost of this prostate cancer presumption. It states that

The 6-year benefit cost for prostate cancer based on herbicide exposure is $65.3 million, with an administrative cost of $959,000. Additionally, the medical care cost over 6 years is $38 million. Prostate cancer is a male genitourinary cancer that shows marked increased prevalence with age. Accordingly, costs beyond the 6-year period would likely be substantially higher.

(VA, 1996c, p. 41370)


VA provided the Committee two summary cost estimate documents relating to prostate cancer (VA, 1996a,b). These documents, entitled Cost Estimate for Regulation on Claims Based on Exposure to Herbicides (dated June 10, 1996, and November 6, 1996) summarize the anticipated benefit (i.e., veterans and survivors) and administrative costs for the prostate cancer presumption. Both documents state the “methodology was developed in collaboration with the Office of Man-

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

agement and Budget (OMB).” Caseload was projected by applying DoD mortality rates in military retirees to the estimated number of gross separations of veterans who had service in the Vietnam theater from 1965 through 1981. Age-specific incident and mortality rates for prostate cancer, as found in the Cancer Statistics Review of 1973-1986, were applied to the base population. On June 10, 1996, the total benefit costs estimated for years 1997-2001 were $56.4 million while the administrative costs were estimated at $787,000. On November 6, 1996, the total benefit costs estimated for years 1997-2002 were $65.3 million. The Committee was not provided recent cost estimates with regard to the prostate cancer presumption nor could the Committee confirm the actual costs following the 1996 estimates through year 2002. Taking into account the number of surviving Vietnam veterans who will soon enter into the age range for which there is a steep increase in age-related incidence of prostate cancer, it can be anticipated that a marked rise of prostate cancer in the Vietnam veteran population is virtually certain.

Lessons Learned

This case study offers several lessons that are relevant when considering strategies for improving the current system of presumptions. These lessons relate to the role of Congress in issuing the Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.), VA in its use of scientific evidence and issuance of a presumption related to prostate cancer, and IOM in its evaluation and presentation of the body of evidence on the relationship between Agent Orange and prostate cancer.

Congress

In describing the type of relationship between dioxin and health outcomes necessary for a presumption, Congress used the language both of “association” as well as “causation” in the Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.) (see Chapter 4). Association appears to be the standard set for VA:

An association between the occurrence of a disease in humans and exposure to an herbicide agent shall be considered to be positive for the purposes of this section if the credible evidence for the association is equal to or outweighs the credible evidence against the association.

(Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess. § 2[b][3])


However, the congressional language regarding “Scientific Determinations Concerning Diseases” includes evidence related to causation (Public Law 102-4):

  1. whether a statistical association with herbicide exposure exists, taking into account the strength of the scientific evidence and the appropriateness of the statistical and epidemiologic methods used to detect the association;

  2. the increased risk of the disease among those exposed to herbicides during service in the Republic of Vietnam during the Vietnam era; and

  3. whether there exists a plausible biological mechanism or other evidence of a causal relationship between herbicide exposure and disease.

(Agent Orange Act of 1991. Public Law 102-4. 102d Cong., 1st Sess. § 3[d]; emphasis added)

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

This inconsistency around “causation” and “association” in the congressional language may have allowed differing interpretations of whether sufficient scientific evidence exists for establishing a presumption. The difficulties that result from this lack of clarity in the Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess.) may have been problematic for prostate cancer for which some evidence for association is present from epidemiological data, but limited by chance, bias, or confounding while the current toxicological evidence for causation (particularly evidence from animal studies and on plausible biological mechanisms) differs from that of the epidemiologic data and was not integrated to make an overall conclusion regarding causality.

In addition to the more general confusion between “association” and “causation” in the Agent Orange Act described above, the second charge set for IOM by this act—evaluating evidence for the increased risk of disease among those exposed to herbicides during service in the Republic of Vietnam during the Vietnam era—is particularly difficult to address. Nearly all of the Agent Orange reports issued by IOM (1994, 1996, 1999, 2001, 2003, 2005) comment on the challenge of addressing this second charge noting that the lack of exposure data on Vietnam veterans made the task difficult to fulfill. The intent of Congress in this second charge is unclear, particularly as the lack of exposure data for Vietnam veterans was well known at the time of the Agent Orange Act.

VA

Overall, VA’s process for using the IOM reports to inform presumptive decisions is not transparent. The prostate cancer presumption was based on a “limited/suggestive” classification of the levels of evidence, rather than the highest standard of “sufficient” evidence of an association. This decision was most likely influenced by a variety of considerations (i.e., political, economic, and administrative factors) and not just by the scientific evidence. The interplay of these multiple factors and their relative weighting by VA are not easily characterized because of lack of transparency in the VA process for using scientific evidence to arrive at the final decision of issuing a presumption. This attempt at a balanced critique of the various parties in the Agent Orange and prostate cancer presumption is limited by the nonspecific information provided by VA on this subject, in particular the use of the “limited/suggestive” classification of an association.

VA’s cost projections for the prostate cancer presumption are likely underestimates and did not consider the very large cohort of aging Vietnam veterans past year 2002. Taking into account the number of surviving Vietnam veterans who will soon enter into the age range for which there is a steep increase in age-related incidence of prostate cancer, it can be anticipated that a marked rise of prostate cancer in the Vietnam veteran population is virtually certain.

IOM

There were some questions regarding the responsiveness of the IOM process (via VA) to congressional intent. The Agent Orange Act of 1991 (Public Law 102-4. 102d Cong., 1st Sess. § 2[b][3]) states that a positive association is one for which “the credible evidence of the association is equal to or outweighs the credible evidence against the association.” This statement is not necessarily equivalent to the category of “limited/suggestive” evidence for association used by IOM Agent Orange committees. The IOM “limited/suggestive” category covers a potential range of epidemiologic evidence from relatively weak to strongly suggestive. In the instance of Agent Orange and prostate cancer, VA established a presumption based on the IOM’s 1996 classification of “limited/suggestive” evidence of an association. The evidence at the time was relatively

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

limited, but it did include one study showing a statistically significant excess and a number of other studies showing positive, but weak and nonsignificant associations.


Biological Plausibility The IOM Agent Orange committees did not appear to have an explicit and formal methodological protocol for synthesizing evidence and updating the classification of evidence based on new studies. The IOM Agent Orange committees appear to have developed a process which differs somewhat from that used by other Academies’ committees and national and international organizations in their review of the level of evidence between an environmental agent and adverse consequences. In reviewing evidence on Agent Orange, the 2005 review considered biological and toxicological evidence when evaluating the biological plausibility of the association between prostate cancer and exposure to Agent Orange but did not integrate these other lines of evidence with epidemiologic findings to develop an overall evaluation. In seeking to classify the strength of evidence with regard to association only, prior IOM committees have relied almost entirely on epidemiologic findings in classifying the strength of evidence.


Approach to Dose-Response The criterion for reaching “limited/suggestive” evidence for association of Agent Orange used by the IOM committees is that the “(e)vidence is suggestive of an association between herbicides and the outcome but is limited because chance, bias, and confounding could not be ruled out with confidence. For example, at least one high-quality study shows a positive association, but the results of other studies are inconsistent” (IOM, 2005, p. 8). The Agent Orange committees’ evaluations for prostate cancer considered the totality of epidemiologic evidence, but did not take into account the very much higher exposures in certain exposed worker groups and in the population exposed during the Seveso incident. Several studies were of worker groups and measures of exposure to either the chlorophenoxy compounds or dioxin but were not available to enable comparisons across studies. Dose-response relationships could be considered for a few studies; for example, the RR “of prostate cancer in the Ranch Hands study correlated with putative exposure to Agent Orange (high 6.04 versus low 2.17 versus background 1.5; P = 0.01)” (IOM, 2005, p. 282). The lack of accurate estimates of exposure from the military experience has limited the study of the association between prostate cancer and exposure to Agent Orange in veterans.

The IOM reports did not attempt to address the fraction of prostate cancer risk that might be attributed to military service. It nonetheless notes that “(p)rostatic cancer is a common condition in older men, so it is likely that multiple factors are responsible and unlikely that herbicide exposure is a major cause” (IOM, 2005, p. 282) An explicit evaluation of attributable fraction might have been useful to policy makers considering this presumption by placing the scientific evidence on the link between Agent Orange exposure and prostate cancer in context of overall prostate cancer risks, although determining the magnitude of this attributable fraction would remain difficult in the absence of accurate exposure data.

General

The Agent Orange and prostate case study illustrates the challenge—both for scientists and policy makers—of evaluating evidence for an association between exposure and subsequent disease when accurate exposure data are lacking. This challenge is particularly striking for a disease like prostate cancer where multiple factors often contribute to the onset of disease, and the other known risk factors (e.g., race, age, family history of the disease, and a diet high in fats) are common in the general population. In this context, the contribution of exposures incurred during

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

prior military service to overall disease risk is likely to be small as acknowledged in the IOM reports. In the absence of accurate exposure data, determining whether this small increased risk of disease is present and further quantifying the magnitude of this risk is difficult.

Faced with this challenge of identifying a possible small increased risk of disease without accurate exposure data, it appears that policy makers have adopted an approach in the prostate cancer presumption that minimizes the possibility of denying service connection to a veteran whose prostate cancer may have been caused by Agent Orange (maximizing sensitivity). The implicit assumption in the prostate cancer presumption is that if any possibility exists, no matter how small, that a veteran may have been exposed to any amount of Agent Orange (presumption of exposure), and any possibility exists, no matter how small, that Agent Orange may have contributed even the smallest incremental increased risk of prostate cancer (presumption of association), service connection should be granted.

High-quality data for a cohort of veterans are essential for improving this process; ideally such data would include more accurate assessments of exposure during service, evaluation of other risk factors that may have been present during service or have developed after service before the onset of disease, and longitudinal assessments for evaluation of diseases that may have long latency periods. The prostate cancer case study highlights the potential value of such an ongoing cohort study and the missed opportunities when such studies are not continued.

References

Akhtar, F. Z., D. H. Garabrant, N. S. Ketchum, and J. E. Michalek. 2004. Cancer in U.S. Air Force veterans of the Vietnam war. Journal of Occupational and Environmental Medicine 46(2):123-136.

ATSDR (Agency for Toxic Substances and Disease Registry). 1998. Toxicological profile for chlorinated dibenzo-p-dioxins (CDDs). Atlanta, GA: U.S. Public Health Service.

EPA (U.S. Environmental Protection Agency). 2003. Exposure and human health reassessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Part II: Health assessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. NAS review draft. Washington, DC: National Center for Environmental Assessment, Office of Research and Development, U.S. Environmental Protection Agency. http://www.epa.gov/ncea/pdfs/dioxin/nas-review/#part2 (accessed June 9, 2007).

IARC (International Agency for Research on Cancer). 1997. Polychlorinated dibenzo-para-dioxins and polychlorinated dibenzofurans. Summaries of data reported and evaluation. Vol. 69. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. http://monographs.iarc.fr/ENG/Monographs/vol69/volume69.pdf (accessed March 8, 2007).

IOM (Institute of Medicine). 1994. Veterans and Agent Orange: Health effects of herbicides used in Vietnam. Washington, DC: National Academy Press.

IOM. 1996. Veterans and Agent Orange: Update 1996. Washington, DC: National Academy Press.

IOM. 1999. Veterans and Agent Orange: Update 1998. Washington, DC: National Academy Press.

IOM. 2001. Veterans and Agent Orange: Update 2000. Washington, DC: National Academy Press.

IOM. 2003. Veterans and Agent Orange: Update 2002. Washington, DC: The National Academies Press.

IOM. 2005. Veterans and Agent Orange: Update 2004. Washington, DC: The National Academies Press.

Morrison, H., D. Savitz, R. Sernenciw, B. Hulka, Y. Mao, D. Morison, and D. Wigle. 1993. Farming and prostate cancer mortality. American Journal of Epidemiology 137(3):270, 280.

NRC (National Research Council). 1972. The effects on populations of exposure to low levels of ionizing radiation (BEIR I). Washington, DC: National Academy Press.

NRC. 1979. The effects on populations of exposure to low levels of ionizing radiation (BEIR II). Washington, DC: National Academy Press.

NRC. 1980. The effects on populations of exposure to low levels of ionizing radiation (BEIR III). Washington, DC: National Academy Press.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

NRC. 1988. Health effects of radon and other internally deposited alpha-emitters (BEIR IV). Washington, DC: National Academy Press.

NRC. 1990. Health effects of exposure to low levels of ionizing radiation (BEIR V). Washington, DC: National Academy Press.

NRC. 1998. Health risks from exposure to low levels of ionizing radiation (BEIR VII - Phase 1, Letter Report). Washington, DC: National Academy Press.

NRC. 1999a. Arsenic in drinking water. Washington, DC: National Academy Press.

NRC. 1999b. Health effects of exposure to radon (BEIR VI). Washington, DC: National Academy Press.

NRC. 2006a. Health risks from dioxin and related compounds: Evaluation of the EPA reassessment. Washington, DC: The National Academies Press.

NRC. 2006b. Health risks from exposure to low levels of ionizing radiation (BEIR VII - Phase 2). Washington, DC: The National Academies Press.

NTP (National Toxicology Program). 2006. Toxicology and carcinogenesis studies of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (CAS No. 1746-01-6) in female Harlan Sprague-Dawley rats (Gavage Study). Technical Report Series No. 521. NIH Publication No. 06-4468. Research Triangle Park, NC: National Toxicology Program.

Steenland, K. L., P. Bertazzi, A. Baccarelli, and K. Manolis. 2004. Dioxin revisited: Developments since the 1997 IARC classification of dioxin as a human carcinogen. Environmental Health Perspectives 112(13):1265-1268.

VA (Department of Veterans Affairs). 1994. Disease Not Associated with Exposure to Certain Herbicide Agents. Notice. Federal Register 59(2):341-346.

VA. 1996a. Cost estimate for regulation on claims based on exposure to herbicides. Washington, DC: Office of Resource Management, Department of Veterans Affairs.

VA. 1996b. Cost estimate for regulation on claims based on exposure to herbicides. Washington, DC: Office of Resource Management, Department of Veterans Affairs.

VA. 1996c. Diseases associated with exposure to certain herbicide agents (prostate cancer and acute and subacute peripheral neuropathy). Proposed rule. Federal Register 61(154):41368-41371.

VA. 1996d. Diseases associated with exposure to certain herbicide agents (prostate cancer and acute and subacute peripheral neuropathy). Final rule. Federal Register 61(217):57586-57589.

CASE STUDY 9:
AGENT ORANGE AND TYPE 2 DIABETES PRESUMPTION

This case study examines the 2001 decision by the VA to establish presumptive service connection for type 2 diabetes based on herbicide exposure among Vietnam veterans. By exploring the scientific and legislative history of this decision, as well as the consequences of this decision for VA, this analysis seeks to illustrate the challenges and implications of establishing a service connection for a common chronic condition when exposure data are unavailable and evidence for association with the putative causal agent is limited.

Type 2 Diabetes

This presumption establishes service connection for type 2 diabetes developing in veterans of the Vietnam era. Several clinical and epidemiological features of type 2 diabetes are noted as they inform some of the challenges in the epidemiologic investigations of type 2 diabetes, particularly with regard to characterizing the effect of Agent Orange. First, type 2 diabetes is common in the general population, and its prevalence increases with advancing age (Figure I-6). This high background prevalence may pose challenges for determining additional risk attributable to military service in many studies. The high prevalence of type 2 diabetes may also translate into a

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

FIGURE I-6 Prevalence and incidence of type 2 diabetes in the United States.

SOURCE: CDC, 2005, pp. 4, 6.

FIGURE I-7 Age- and sex-adjusted trends in type 2 diabetes prevalence.

SOURCE: Adapted, with permission, from Gregg et al., 2005, p. 1871.

larger constituency advocating for this presumption and affect the financial implications of a decision.

Second, there are multiple established risk factors for type 2 diabetes that are common in the general population. These contributors to type 2 diabetes risk include family history and lifestyle factors such as obesity and physical inactivity (Figure I-7). Because these risk factors are common, estimating the incremental risk of type 2 diabetes attributable to herbicide exposure among Vietnam veterans is challenging.

The known link between obesity and type 2 diabetes has particularly important implications for evaluating the observational studies exploring the risk of type 2 diabetes associated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) levels. Adipose tissue is a primary site of distribution for TCDD, and percentage of body fat appears to correlate with serum TCDD levels (IOM, 1994). Increasing body fat may also decrease the rate of TCDD elimination (IOM, 1994). The relationship between body fat and TCDD pharmacokinetics suggests that body mass index and other measures of body fat are important covariates that must be considered in observational studies of dioxins and risk for type 2 diabetes.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Third, type 2 diabetes is a disorder of carbohydrate metabolism that is characterized by hyperglycemia. Many individuals with type 2 diabetes initially exhibit few or no symptoms, making the exact onset of this disease difficult to establish. The diagnostic criteria for type 2 diabetes are based primarily on blood glucose levels, and the level of hyperglycemia that is diagnostic of type 2 diabetes has been adjusted downward over time. The current criteria are fasting blood glucose ≥126 mg/dL or a random blood glucose ≥200 mg/dL in the presence of symptoms (often polyuria or polydipsia) (ADA, 2007). The World Health Organization and American Diabetes Association definitions recognize a state of impaired glucose tolerance (fasting blood glucose of 100-125 mg/dL) that is not diagnostic of type 2 diabetes, but places individuals at high risk for subsequent development of this disease (ADA, 1997, 2003, 2007; Alberti and Zimmet, 1998). These changing definitions of type 2 diabetes make comparisons of studies across time potentially difficult. In addition, many studies evaluate both type 2 diabetes and blood glucose levels as outcomes; although hyperglycemia not meeting diagnostic criteria for type 2 diabetes may signal an important increased risk of type 2 diabetes, the significance of small increases in mean glucose concentrations within the range of normal glucose levels (<100 mg/dL) (ADA, 2007) may be less clear.

Finally, type 2 diabetes is an extraordinarily morbid condition that affects virtually every organ system. Heart disease and stroke account for nearly 65 percent of deaths among individuals with type 2 diabetes (CDC, 2005, p. 6). Type 2 diabetes is also associated with significantly higher rates of nontraumatic amputations, kidney disease, neuropathies, and blindness (CDC, 2005). There are two important implications of this high rate of morbidity associated with type 2 diabetes. First, mortality studies of this disease are limited because complications of type 2 diabetes (i.e., heart disease) are often listed as the underlying cause of death, resulting in an underestimate of type 2 diabetes-related mortality from use of death certificate diagnoses. Second, the multiple complications of type 2 diabetes could have significant consequences for the true costs associated with a service connection for this common and highly morbid chronic condition.

The Need for a Presumption

The history of the Agent Orange Act of 1991 (Public Law.102-4. 102nd Cong., 1st Sess.) establishing a presumption of service connection for diseases associated with herbicide exposure during the Vietnam conflict as well as the procedure of adding additional diseases to this presumption has been reviewed elsewhere in this report (see Chapter 2, Chapter 4, Appendix D). The need for a presumption for Agent Orange-associated diseases generally is based in part on the difficulty of establishing with certainty the degree of exposure to herbicides among Vietnam veterans (IOM, 1994). The Agent Orange Act eliminates the need for an individual Vietnam veteran to provide evidence of herbicide exposure; exposure is presumed for all Vietnam veterans.

Although the need for a presumption is based primarily on the difficulty of establishing exposure, presumptions also address gaps in the evidence for association. An examination of the exact charge to the IOM for determining the diseases that are linked to herbicides is instructive. For each disease, IOM has been asked to determine, to the extent that available data permit meaningful determinations,

  1. whether a statistical association with herbicide exposure exists, taking into account the strength of the scientific evidence and the appropriateness of the statistical and epidemiological methods used to detect the association;

  2. the increased risk of the disease among those exposed to herbicides during Vietnam service; and

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
  1. whether there is a plausible biological mechanism or other evidence of a causal relationship between herbicide exposure and the disease.

Each of the IOM reports that examined herbicides and type 2 diabetes risk (IOM, 1994, 1996, 1999, 2000, 2001, 2003, 2005) has focused almost exclusively on addressing the task described in charge 1 and have included additional discussions of the more limited data to address charge 3. Each of the IOM reports address charge 2 but acknowledge the difficulty of providing risk estimates because of the lack of exposure data:

Although there are data to suggest an association between exposure to the chemicals of interest and type 2 diabetes, the lack of exposure information on Vietnam veterans precludes quantification of any possible increase in their risk.

(IOM, 2005, p. 446)


Some committees additionally note, in response to charge 2, that the epidemiological evidence on known risk factors for type 2 diabetes would suggest that the contribution of exposures during military to overall diabetes risk is likely to be small:

Available data allow for the possibility of an increased risk of type 2 diabetes in Vietnam veterans. It must be noted, however, that studies indicate that the increased risk, if any, posed by herbicide or TCDD exposure appears to be small. The known predictors of diabetes risk—family history, physical inactivity, and obesity—continue to greatly outweigh any suggested increased risk posed by wartime exposure to herbicides.

(IOM, 2003, p. 492)


The presumption of service connection for type 2 diabetes in Vietnam veterans therefore serves more than just to address the gap in evidence of exposure. As the lack of exposure data makes quantification of the magnitude of the association between military service and diabetes difficult, this presumption also implicitly serves to address a gap in the evidence for association between herbicides and type 2 diabetes among Vietnam veterans.

A Review of the Scientific and Legislative Landmarks Leading to the Agent Orange and Type 2 Diabetes Presumption by VA

As a result of Public Law 102-4 (Agent Orange Act of 1991. Public Law 102-4. 102nd Cong., 1st Sess.), the NAS signed an agreement with VA to review and summarize the strength of the scientific evidence concerning the association between herbicide exposure during Vietnam service and diseases or conditions that may be associated with this exposure. The first report from NAS was issued in 1994 (IOM, 1994) and has been updated on a biennial basis since that time. Additionally, a special report on Agent Orange and type 2 diabetes was published in 2000 (IOM, 2000). This section reviews the scientific evidence contained in the IOM Agent Orange reports of 1994-2005 and the legislative events that resulted from the conclusions of these reports. It is not meant to be a comprehensive analysis of the entire body of scientific literature on this topic, but rather to highlight those events and data that inform the conclusions in the reports and changes in legislation.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
1994-1999

The earliest publication reviewed by IOM that suggested a link between the herbicides and type 2 diabetes was a report of the 10-year follow-up of 55 workers who had become acutely ill after exposure to TCDD. Eight percent of these workers developed type 2 diabetes immediately following exposure; after 10 years, 11 of the 55 had developed type 2 diabetes (Pazderova-Vejlupkova et al., 1981, pp. 6, 7). Several small epidemiologic and occupational studies that followed the report of these cases failed to show a statistically significant association between exposure to these compounds and type 2 diabetes risk or type 2 diabetes mortality (Bertazzi et al., 1998; Cook et al., 1987, as referenced in IOM, 1999; Henneberger et al., 1989, as referenced in IOM, 1999; Moses et al., 1984, as referenced in IOM, 1999; May, 1982, as referenced in IOM, 1999; von Benner et al., 1994, as referenced in IOM, 1996; Zober et al., 1994, as referenced in IOM, 1996).

The National Institute for Occupational Safety and Health (NIOSH) studied 281 workers at dioxin plants in New Jersey and Missouri and compared them with 260 unexposed workers. Sweeney et al. found that increasing concentrations of TCDD were associated with a slight, but statistically significant increased risk of type 2 diabetes [odds ratio (OR) 1.1, P < .003] and elevated fasting glucose (≥ 140 mg/dL) (P < .001) (Sweeney et al., 1996, p. 245; Sweeney et al., 1997, as referenced in IOM, 2000, pp. 16, 54). An earlier study of this cohort found that exposed workers had a type 2 diabetes prevalence of 9.2 percent compared with 5.8 percent among unexposed workers, a difference that did not reach statistical significance (Sweeney et al., 1992, as referenced in IOM, 1999, p. 500). The authors of these studies concluded that more analyses were necessary to eliminate the possibility of confounding from other established risk factors for type 2 diabetes, including age and weight, and the IOM Agent Orange 1998 update (IOM, 1999) called for a more detailed analysis of this cohort.

A study of Air Force veterans involved in herbicide spraying—the Air Force Health Study (AFHS)—suggested that serum concentration of TCDD was associated with glucose intolerance or type 2 diabetes in a dose-dependent manner. In this 1997 study (Henriksen et al., 1997, pp. 253-256), 989 Air Force veterans involved in Agent Orange spraying (Operation Ranch Hand) were categorized based on their TCDD levels; initial TCDD levels at the time of wartime exposure were estimated based on their current serum TCDD. In addition to comparing disease outcomes between veterans involved in Agent Orange spraying in different categories of TCDD levels, these subjects were also compared with 1,276 Air Force veterans that were not involved in spraying. Those with the highest level of TCDD had a significantly increased risk of elevated blood glucose (RR 1.4, 95% CI 1.1-1.8) and type 2 diabetes (RR 1.5, 95% CI 1.2-2.0) and a decreased time to onset of disease after adjusting for age and body mass index, compared to those with the lowest level of TCDD. Veterans involved in Agent Orange spraying with intermediate levels of TCDD exhibited an intermediate risk for these outcomes. This study was commended in the IOM Agent Orange 1998 update (IOM, 1999), but two concerns were also noted. First, there were no differences in type 2 diabetes rates overall between veterans involved in Agent Orange spraying and the designated comparison group that was not involved in spraying, despite significant differences in their serum TCDD levels (median 12.2 ppt versus 4.0 ppt [Henriksen et al., 1997]) (IOM, 1999, p. 499). Second, the dose-dependent association with TCDD concentration was also observed within the comparison group that was not involved with spraying (IOM, 1999, p. 500). Taken together these additional observations are concerning for residual confounding by a third factor such as weight or body fat that may be related to both TCDD levels

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

and type 2 diabetes. The IOM Agent Orange 1998 Update (IOM, 1999) called for a more detailed analysis of the AFHS.

One omission from the otherwise comprehensive review of scientific literature in the IOM reports of this period is an analysis of the Vietnam Experience Study of the CDC. CDC used a random sample of military records to identify a cohort of U.S. enlisted men who had served a single tour in Vietnam between 1965 and 1971 and a comparison cohort of U.S. Army enlisted men who had served elsewhere during this period. TCDD levels were not assessed in this study; rather Vietnam service was used as a proxy for exposure to herbicides. A total of 7,924 Vietnam and 7,364 Vietnam-era veterans completed telephone health surveys in 1985 and 1986 (CDC, 1989b, p. 1). Vietnam veterans did not have an increased risk of self-reported type 2 diabetes compared with the Vietnam-era veterans (adjusted OR = 1.2, P > 0.05) (CDC, 1989b, pp. 109, 114). A random subsample of these individuals (2,490 Vietnam veterans and 1,972 Vietnam-era veterans) underwent complete physical examinations in 1985 and 1986 (CDC, 1989a). The crude geometric mean of fasting serum glucose was slightly higher for Vietnam veterans (93.4 mg/dL) than for Vietnam-era veterans (92.4 mg/dL), a difference that was found to be significant even after adjustment for a variety of factors including age, race, military specialty, body mass index, and medications known to affect glucose metabolism (CDC, 1989c, p. 174). Despite this difference in mean glucose levels, there was no significant difference between the two groups in the number of veterans with abnormal blood glucose levels (> 140 mg/dL in this study) (CDC, 1989c, pp. 172, 175).

Although the Vietnam Experience Study is directly relevant to the IOM charge of determining whether herbicide exposure during the Vietnam era resulted in an increased risk of type 2 diabetes among Vietnam veterans, the IOM reports do not comment on the type 2 diabetes results found in the Vietnam Experience Study. The finding of no association between military service in Vietnam and type 2 diabetes in this study is not included in the summary analyses presented in the various tables of these reports. As participants in the Vietnam Experience Study were not followed beyond this assessment in the 1980s, it is impossible to determine whether the very small but significant increases in blood glucose concentration observed in this younger cohort might have been an indicator of subsequent type 2 diabetes risk.

In reviewing the studies of this period, the IOM Agent Orange reports of 1994, 1996, and 1998 (IOM, 1994, 1996, 1999) all concluded that there was inadequate or insufficient evidence to determine whether an association existed between herbicide exposure and type 2 diabetes. No presumption of service connection for type 2 diabetes among Vietnam veterans exists for this time period.

1999-2001

In 1999 VA asked IOM to convene a special committee to conduct a focused review of the scientific evidence regarding the association between Agent Orange and type 2 diabetes in advance of the regularly scheduled biennial IOM reports (IOM, 2000). VA’s reason for convening this special committee out of the scheduled sequence of IOM reports on Agent Orange is outlined in the Federal Register (VA, 2001a) and reviewed here.

In 1999, NIOSH published a further analysis of exposed and unexposed workers in the New Jersey and Missouri dioxin plants (Calvert et al., 1999). The authors of this study summarize their findings in this way:

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Overall, the prevalence of diabetes mellitus was not significantly different between the workers and referents. Also, there was not a significant positive trend between prevalence of diabetes and increasing serum TCDD concentration. However, diabetes was found in six of 10 workers with current serum TCDD concentrations >1500 pg/g lipid.

(Calvert et al., 1999, p. 270)


In the Federal Register outlining the rationale for the special diabetes report, the NIOSH study is summarized in the following way:

… a report that detects an association, though not a strong association, between type 2 diabetes and dioxin exposure. The study does suggest a dose response relationship because of excess cases of type 2 diabetes found in workers having the highest serum-lipid levels of dioxin.

(VA, 2001a, p. 2378)


The VA Secretary viewed this study as potentially important enough to warrant a full review of the scientific literature on Agent Orange and type 2 diabetes and commissioned a special report (IOM, 2000) on this topic from IOM. As with the prior Agent Orange committees, the charge of the Agent Orange and type 2 diabetes committee was to review evidence from epidemiologic studies, evidence for increased risk among Vietnam veterans, and evidence for biologic plausibility; the majority of this report focused on the epidemiologic evidence (as was the case with prior and subsequent reports). In their summary, the Agent Orange and type 2 diabetes committee (IOM, 2000) commented on two trends that they noted in the literature on this topic published since the IOM Agent Orange 1998 Update (IOM, 1999).

The report stated “[p]ositive associations are reported in many mortality studies, which may underestimate the incidence of diabetes” (IOM, 2000, pp. 2, 36). This statement was based on the data from four mortality studies that had not been reviewed in previous IOM reports:

  • A 1998 report of the 5-year follow-up of individuals living near a 1976 industrial accident site involving dioxin (Pesatori et al., 1998). Individuals were grouped into three categories based on their distance from the site of the accident and compared with an unexposed reference group. Women with substantial exposure (41,391 person-years of follow-up) had significantly higher rates of type 2 diabetes mortality (RR 1.9, 95% CI 1.1-3.2); the same effect was observed among men with substantial exposure (42,219 person-years of follow-up), although the results did not reach significance (RR 1.3, 95% CI 0.6-2.9) (Pesatori et al., 1998, pp. 127-128). Only age and calendar period were included as covariates in this study.

  • A study of 5,172 workers exposed to TCDD at 12 U.S. plants did not find an increase in the standardized mortality ratio (SMR) for type 2 diabetes as the primary cause of death (SMR 1.18, 95% CI 0.77-1.73) or type 2 diabetes as the primary or secondary cause of death (SMR 1.08, 0.87-1.33) (Steenland et al., 1999, p. 782). In fact, analysis of this cohort categorized by septiles of TCDD exposure revealed a statistically significant inverse trend between cumulative exposure and type 2 diabetes risk (those with the highest exposure had the lowest type 2 diabetes risk; p = 0.02), though the trend was not statistically significant when the logarithm of cumulative exposure was considered (p = 0.12) (Steenland et al., 1999). An analysis of the 608 individuals who developed chloracne after the exposure (a condition

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

that may indicate high exposure) failed to reveal an increased risk of type 2 diabetes (Steenland et al., 1999, p. 781). These studies accounted for age and calendar year.

  • A 1998 study that combined data from multiple exposed cohorts internationally (including the U.S. cohort analyzed in the study by Steenland et al. referenced above) found that exposure was associated with an RR of type 2 diabetes of 2.25 (0.53-9.50), although these results did not reach the level of statistical significance (Vena et al., 1998, p. 649). Age, gender, country, employment status, and calendar period were included as covariates in this analysis.

The results of these studies are mixed with a nonsignificant trend toward association between exposure and type 2 diabetes observed in some studies and a significant inverse association between TCDD and type 2 diabetes noted in another study. Additionally, the relevance of the committee’s concern that mortality studies may underestimate the risk of type 2 diabetes is not entirely clear. The report (IOM, 2000, pp. 2, 36), as quoted extensively in the Federal Register (VA, 2001a, p. 2378), notes that (1) type 2 diabetes is not typically fatal, (2) complications of type 2 diabetes are more likely to be listed as a cause of death rather than type 2 diabetes itself, and (3) contributory factors (such as type 2 diabetes) are not routinely listed on death certificates. Although these features are likely all true and would be expected to underestimate type 2 diabetes-associated mortality generally, there is no reason to suspect that this underreporting of type 2 diabetes as a cause of death should differ based on exposure status. The strength of association between dioxin exposure and type 2 diabetes, therefore, would likely be unaffected by overall underreporting of type 2 diabetes on death certificates. However, the IOM report (IOM, 2000) and its summary by VA in the Federal Register (VA, 2001a) appear to suggest that the associations observed in these mortality studies be given additional weight because the type 2 diabetes mortality is underestimated.

The report stated “[p]ositive associations are reported in most of the morbidity studies identified by the committee” (IOM, 2000, pp. 3, 37). Studies of three cohorts are described to support this statement:

  • A survey of male Australian veterans of Vietnam found a statistically significant excess of self-reported type 2 diabetes (2,391 reported cases; 1,780 expected cases with an expected range of 1,558-2,003 (CDVA, 1998, as referenced in IOM, 2000, pp. 32, 37).

  • A further analysis of the NIOSH cohort (the impetus for convening the special Agent Orange type 2 diabetes committee) of 281 workers (exposed to TCDD in chemical plants more than 15 years prior) and 260 controls (from the same residential neighborhood as the exposed individuals and matched on age, race, and sex) found no significantly increased risk of type 2 diabetes among exposed individuals compared with reference individuals (OR 1.49, 95% CI 0.77-2.91) (Calvert et al., 1999, pp. 270-271, 273). Additionally, there was no trend observed between increasing TCDD concentration and type 2 diabetes risk. However, when individuals with type 2 diabetes were excluded from the analysis, individuals in the highest TCDD group had significantly higher glucose concentrations than the unexposed group (geometric mean 5.45 mmol/L [1.02] vs. 5.21 mmol/L [1.01]).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

These analyses were all adjusted for age, gender, race, body mass index, and medications that affect glucose metabolism.

  • The AFHS again provided important, though perhaps conflicting, evidence related to TCDD and type 2 diabetes. Updated analysis of veterans of Operation Ranch Hand estimated TCDD concentration at the time of exposure based on the elimination kinetics observed in a sample of Ranch Hand veterans with multiple TCDD measures over time. Among Ranch Hand veterans, both measured TCDD concentrations from the start of the study (1987) and estimated time-of-exposure TCDD concentrations were associated with the risk of diabetes. Those Ranch Hand veterans with the highest estimated time-of-exposure TCDD concentration also had higher rates of diabetes compared with the unexposed comparison cohort that had not been involved in Agent Orange spraying, although overall diabetes rates did not differ between Ranch Hand veterans and the unexposed comparison cohort (AFHS, 2000). Among the comparison cohort that had not been involved in Agent Orange spraying, a significant dose-dependent relationship between TCDD concentration and type 2 diabetes prevalence was also observed, with an RR of 1.71 (95% CI 1.00-2.91) for the highest quartile concentration compared with the lowest (Longnecker and Michalek, 2000, p. 46). Veterans in this comparison cohort had very low serum TCDD concentrations, within the range of background exposure typically seen in the United States (≤10 ng/kg lipid) (Longnecker and Michalek, 2000, p. 45). The committee noted that the observations from AFHS, including a graded risk of diabetes associated with very low levels of TCDD, continued to leave open the possibility that residual confounding related to body fat or correlated factors such as triglyceride concentration may interfere with accurate evaluation of the relationship between TCDD and type 2 diabetes (IOM, 2000).

Because of the trends that the committee found in the newly published mortality and morbidity literature, the IOM report on type 2 diabetes concluded that evidence for the association between herbicides and type 2 diabetes had reached the level of limited/suggestive—that is, evidence is suggestive of an association between herbicides and the outcome, but limited because chance, bias, and confounding could not be ruled out with confidence (IOM, 2000).

This “limited/suggestive” finding of the IOM report was submitted to the VA Secretary. In accordance with Title 38 USC 1116(b)(1) (Compensation for Service-Connected Disability or Death. 2006. 38 USC § 1116[b][1]), the VA Secretary was required to review this report and determine whether a presumption was warranted. The VA Secretary determined that evidence of a “positive association” between Agent Orange and type 2 diabetes existed, noting that an association is considered “positive” if the credible evidence for the association is equal to or outweighs the credible evidence against the association. Additional information regarding how data from the IOM report and the IOM summary finding of “limited/suggestive” figured into VA’s recommendation is unavailable. The VA Secretary’s determination of a positive association between herbicide and type 2 diabetes resulted in the May 8, 2001, 38 CFR Part 3 (VA, 2001b) finding of presumptive service connection for type 2 diabetes based on herbicide exposure during Vietnam.

In the Federal Register commentary, the increasing rates of type 2 diabetes associated with the obesity prevalence was noted, and the suggestion was made that more studies be undertaken that control for the background high rates of obesity before the type 2 diabetes presumption be considered. In response to these comments, the VA Secretary noted that IOM “adequately took

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

into consideration the recognized relationship between obesity and type 2 diabetes” (VA, 2001b, p. 23167). This statement appears to conflict with the IOM reports, as the 2000 IOM type 2 diabetes report stated that “the known predictors of diabetes risk—family history, physical inactivity, and obesity—continue to greatly outweigh any suggested increased risk from wartime exposure to herbicides” (IOM, 2000, pp. 3, 37). In the Federal Register commentary, the VA Secretary also notes that the time requirements imposed by section 1116(c)(2) limited the capacity of VA to wait for additional studies on this topic (VA, 2001b).

Title 38 USC 1116(c)(1) (Compensation for Service-Connected Disability or Death. 2006. 38 USC § 1116[c][1]) requires that the Secretary, not later than 60 days after the date on which he receives a report from NAS, determine whether a presumption of service connection is warranted for each disease covered by the report, and if the Secretary determines that a presumption is warranted, issue proposed regulations within 60 days thereafter…. We believe that NAS adequately took into consideration the recognized relationship between obesity and type 2 diabetes, and the existence of additional studies concerning this risk factor does not warrant ignoring the time requirements of section 1116(c)(2).

(VA, 2001b, p. 23167)

2002-2004

IOM reports over this time period (IOM, 2003, 2005) continued to find limited/suggestive evidence for the association between herbicides and type 2 diabetes. One important study of this period includes an attempt to pool the data on participants from the NIOSH study with those from AFHS (Steenland et al., 2001). To match the criteria in AFHS, the cut-off values defining type 2 diabetes in the NIOSH study were adjusted downward (from > 140 mg/dL to > 126 mg/dL) and 55 individuals included in the NIOSH study were excluded from the joint analysis. As a result of these changes, the OR for the NIOSH data was reduced from 1.49 to 1.22 and the dose-dependent association was no longer observed (OR = 0.84, 95% CI 0.4-1.8 for the highest TCDD concentration compared with the lowest). The significant association between serum concentration of TCDD and type 2 diabetes observed in the AFHS cohort was unchanged. Another study explored one possible mechanism for residual confounding in the AFHS studies and found that the significant association between TCDD concentration and type 2 diabetes risk noted in AFHS was not attributable to a third factor that both caused type 2 diabetes and slowed elimination of TCDD (Michalek et al., 2003). The 2002 and 2004 reports again rated the evidence for an association between herbicides and type 2 diabetes as “limited/suggestive.”

Cost Implications of the Presumption for Type 2 Diabetes

There are limited data regarding the costs to VA associated with the type 2 diabetes presumption. The commentary on 38 CFR Part 3 (VA, 2001b, p. 23168) includes a brief discussion of the cost of this type 2 diabetes presumption. Cost projections were estimated by applying a type 2 diabetes prevalence rate of 9 percent to the 2.3 million estimated living Vietnam veterans (VA, 2000). National survey data suggest that 9 percent may be an underestimate, particularly as the population of Vietnam veterans ages and national trends continue to demonstrate increasing rates of incident type 2 diabetes across all age groups (Figure I-8).

The discussion in 38 CFR Part 3 also estimates the average monthly award for type 2 diabetes or its ancillary conditions at $462 (for original claims) (VA, 2001b, p. 23168). Although the

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

bases for this estimate are not described, the multiple possible complications associated with type 2 diabetes suggests that this average payment would likely increase over time for a given veteran with type 2 diabetes. Estimated administrative costs for type 2 diabetes from 2001 through 2005 were $62 million with estimated benefit costs of $3.3 billion during that same time period. VA estimated that there would be 20,399 new type 2 diabetes awards in the first year and 179,000 over the next 5 years. The estimates did not include retroactive payments (McLenachen, 2005). Today, the most frequent service-connected disability for which Vietnam veterans are receiving compensation is type 2 diabetes (VBA, 2006). “At the end of fiscal year 2006, nearly 248,000 veterans were service connected for diabetes. More than 215,000 of these awards were based upon herbicide exposure in Vietnam. As veterans with diabetes reach and move past the 10-year point since initial diagnosis, additional secondary conditions tend to manifest. VA has started to see increasingly complex medical cases resulting in neuropathies, vision problems, cardiovascular problems, and other issues directly related to diabetes” (VA, 2007, pp. 6B-13).

Lessons Learned

This case study offers several lessons that are relevant when considering strategies for improving the current system of presumptions. These lessons relate to the role of Congress in issuing the Agent Orange Act of 1991 (Public Law 102-4. 102nd Cong., 1st Sess.), VA in their use of scientific evidence to both convene the special IOM committee on type 2 diabetes (IOM, 2000) and eventually issue a presumption related to type 2 diabetes, and IOM in its evaluation and presentation of the body of evidence for the relationship between Agent Orange and type 2 diabetes.

FIGURE I-8 Annual incidence of diagnosed type 2 diabetes per 1,000 population aged 18-79 years, by age, United States, 1997-2004.

SOURCE: See http://www.cdc.gov/diabetes/statistics/incidence/fig3.htm.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Congress

In describing the type of relationship between dioxin and health outcomes necessary for a presumption, Congress used the language both of “association” as well as “causation” in the Agent Orange Act of 1991 (Public Law 102-4. 102nd Cong., 1st Sess., Sec. 2[b][3]) (see Chapter 4). Association appears to be the standard set for VA:

An association between the occurrence of a disease in humans and exposure to an herbicide agent shall be considered to be positive for the purposes of this section if the credible evidence for the association is equal to or outweighs the credible evidence against the association.

(Agent Orange Act of 1991. Public Law 102-4. 102nd Cong., 1st Sess.)


However, the congressional language regarding “Scientific Determinations Concerning Diseases” includes evidence related to causation (Agent Orange Act of 1991. Public Law 102-4. 102nd Cong., 1st Sess., Sec. 3[d]):

  1. whether a statistical association with herbicide exposure exists, taking into account the strength of the scientific evidence and the appropriateness of the statistical and epidemiologic methods used to detect the association;

  2. the increased risk of disease among those exposed to herbicides during service in the Republic of Vietnam during the Vietnam era; and

  3. whether there exists a plausible biological mechanism or other evidence of a causal relationship between herbicide exposure and disease. (emphasis added)

This inconsistency in the congressional language may have allowed considerable differences in interpretation of whether scientific evidence exists for the basis for a presumption. The difficulties that result from this lack of clarity in the Agent Orange Act of 1991 (Public Law 102-4. 102nd Cong., 1st Sess.) may have been particularly problematic for type 2 diabetes for which some evidence for association is present, but limited by chance, bias, or confounding.

In addition to the more general confusion between “association” and “causation” in the Agent Orange Act described above, the second standard set for IOM by this act—evaluating evidence for the increased risk of disease among those exposed to herbicides during service in the Republic of Vietnam during the Vietnam era—is particularly vague. Nearly all of the Agent Orange reports issued by IOM (1994, 1996, 1999, 2000, 2001, 2003, 2005) comment on the challenge of addressing this second charge, noting that the lack of exposure data on Vietnam veterans made the task difficult to fulfill. The intent of Congress in this second charge is unclear, particularly as the lack of exposure data for Vietnam veterans was well known at the time of the Agent Orange Act.

VA

In justifying its decision to convene a special IOM panel to evaluate the evidence related to Agent Orange and type 2 diabetes, VA may have overstated the findings of the report that prompted this action. Based on the Federal Register, VA summarized the report by NIOSH on occupational exposure to dioxin as

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

… a report that detects an association, though not a strong association, between type 2 diabetes and dioxin exposure. The study does suggest a dose-response relationship because of excess cases of type 2 diabetes found in workers having the highest serum-lipid levels of dioxin.

(VA, 2001a, p. 2378)


By contrast, the authors of the NIOSH report summarized their findings in this way:

Overall, the prevalence of diabetes mellitus was not significantly different between the workers and referents. Also, there was not a significant positive trend between prevalence of diabetes and increasing serum TCDD concentration. However, diabetes was found in six of 10 (60%) workers with current serum TCDD concentrations > 1500 pg/g lipid.

(Calvert et al., 1999, p. 270)


Because this study was the primary justification for the special IOM committee, an understanding of VA’s interpretation of the study results is essential. No additional information regarding how new studies come to the attention of VA or the criteria for evaluating new studies and using these to inform the charge to IOM were made available to the Committee.

In the debate regarding this presumption, VA dismissed the link between national increases in type 2 diabetes rates related to obesity and the calls for more studies that control for the background high rates of obesity, noting that the IOM “adequately took into consideration the relationship between obesity and type 2 diabetes” (VA, 2001b, p. 23167). This statement appears to conflict with the IOM reports, as the IOM type 2 diabetes report stated that “the known predictors of diabetes risk—family history, physical inactivity, and obesity—continue to greatly outweigh any suggested increased risk from wartime exposure to herbicides” (IOM, 2000, pp. 3, 37). VA noted that the requirement of timely action on the part of VA imposed by the Agent Orange Act prevented consideration of additional studies; although action is required within 60 days of the finding of a “positive association” on the part of the VA Secretary, it is notable that this VA finding was based on the lesser “limited/suggestive” categorization of evidence in the IOM report.

VA’s cost projections for the type 2 diabetes presumption were likely underestimates. VA failed to consider the likely rise in type 2 diabetes prevalence in the aging veteran population and national trends suggesting increasing rates of type 2 diabetes in all age groups. Furthermore, as type 2 diabetes has many known complications that are also highly morbid, it is likely that the average percentage of service-connected disability for veterans with type 2 diabetes will continue to increase over time.

Overall, the Committee found that VA’s process for using IOM reports to inform presumptive decisions has not been transparent. The diabetes presumption signaled an important trend on the part of VA to assign presumption on the basis of “limited/suggestive” classification of the levels of evidence. This decision could have been influenced by a variety of considerations beyond scientific ones, such as political, economic, and administrative factors. The interplay of these multiple factors and their relative weighting by VA are not easily characterized because of the lack of transparency in the VA process for using scientific evidence to arrive at the final decision of issuing a presumption.

This attempt by the Committee to carry out at a balanced critique of the various parties in the Agent Orange and type 2 diabetes presumption is limited by the availability of information from VA on this subject.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
IOM

All of the IOM committees reviewing the type 2 diabetes evidence did not consider the results of the Vietnam Experience Study. This CDC study is one of the few large studies comparing the health of Vietnam veterans to other veterans of the same age who did not serve in Vietnam. The Vietnam-enlisted veterans surveyed in the Vietnam Experience Study did not have an increased risk of self-reported type 2 diabetes compared with other Vietnam-era veterans. In the subsample of participants that underwent blood test evaluation, Vietnam veterans also did not show higher rates of abnormal blood glucose concentrations compared with their Vietnam-era counterparts, although the geometric mean of their fasting serum glucose was slightly (and significantly) higher.

The IOM committee did not appear to have an explicit and formal methodological protocol for synthesizing evidence and for updating the classification of evidence based on new studies. The IOM type 2 diabetes committee “upgraded” their assessment of the existing literature from “inadequate/insufficient” (studies of insufficient quality, consistency, or statistical power to permit conclusion) to “limited/suggestive” (studies limited because chance, bias, and confounding could not be ruled out with confidence; for example, at least one high-quality study shows a positive association, but the results of other studies are inconsistent). The committee’s conclusion was not reached because of a single high-quality study that provided evidence of a conclusive association between Agent Orange and type 2 diabetes. Rather, the reclassification resulted from the committee’s view of the cumulative weight of the evidence from several smaller studies, each limited in varying ways so that overall chance, bias, and confounding could not be ruled out with confidence and some studies differed in the direction of the association found (i.e., positive or negative).

The IOM report emphasized a particular feature of outcome ascertainment in type 2 diabetes research that may have led to overinterpretation of the conclusions of their report. The IOM committee concluded that “[p]ositive associations are reported in many mortality studies, which may underestimate the incidence of diabetes” (IOM, 2000, pp. 2, 36; emphasis added). The committee correctly pointed out that death certificates routinely underestimate death rates attributable to type 2 diabetes. Because type 2 diabetes is not typically fatal, complications of type 2 diabetes are more likely to be listed as a cause of death rather than type 2 diabetes itself, and contributory factors (such as type 2 diabetes) are not routinely listed on death certificates. However, these features would only be expected to underestimate type 2 diabetes mortality generally and should not lead to differential outcome ascertainment based on exposure status. Although the strength of association between dioxin exposure and type 2 diabetes would likely be unaffected by overall underreporting of type 2 diabetes on death certificates, the VA summary of the IOM findings appeared to suggest that the associations observed in these mortality studies be given additional weight because the type 2 diabetes mortality associated with dioxin exposure is underestimated.

Although determining the extent of type 2 diabetes risk that might be attributable to military service was not explicitly part of their charge, several of the committee reports address this issue and implicitly suggest that this fraction is likely to be small:

It must be noted, however, that these studies indicate that the increased risk, if any, posed by herbicide or TCDD exposure appears to be small. The known predictors of diabetes

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

risk—family history, physical inactivity, and obesity—continue to greatly outweigh any suggested increased risk from wartime exposure to herbicides.

(IOM, 2000, pp. 3, 37)


A more explicit evaluation of attributable fraction might have been useful to policy makers considering this presumption by placing the scientific evidence on the link between TCDD and diabetes in context of overall diabetes risks, although determining the magnitude of this attributable fraction would remain difficult in the absence of accurate exposure data.

General

The Agent Orange and type 2 diabetes case study illustrates the challenge—both for scientists and policy makers—of evaluating evidence for an association between exposure and subsequent disease when accurate exposure data are lacking. This challenge is particularly striking for a disease like diabetes where multiple factors often contribute to the onset of disease, and the other known risk factors (e.g., genetics, obesity) are common in the general population. In this context, the contribution of exposures incurred during prior military service to overall disease risk is likely to be small as acknowledged in the IOM reports. In the absence of accurate exposure data, determining whether this small increased risk of disease is present and further quantifying the magnitude of this risk is difficult.

Faced with this challenge of identifying a possible small increased risk of disease without accurate exposure data, it appears that policy makers have adopted an approach in the diabetes presumption that minimizes the possibility of denying service connection to a veteran whose type 2 diabetes may have been caused by Agent Orange (maximizing sensitivity). The implicit assumption in the type 2 diabetes presumption is if any possibility exists, no matter how small, that a veteran may have been exposed to any amount of Agent Orange (presumption of exposure), and any possibility exists, no matter how small, that Agent Orange may have contributed even the smallest incremental increased risk of type 2 diabetes (presumption of association), service connection should be granted.

High-quality data for a cohort of veterans are essential for improving this process. Ideally such data would include (1) more accurate assessments of exposure during service; (2) evaluation of other risk factors that may have been present during service or have developed after service before the onset of disease; and (3) longitudinal assessments for evaluation of diseases that may have long latency periods. The type 2 diabetes case study highlights the potential value of such an ongoing cohort study and the missed opportunities when such studies are not continued. The VES was the largest study of a representative group of Vietnam veterans exploring whether Vietnam service was associated with a variety of disease outcomes. Although this study did not include an assessment of exposures, the variety of measurements included after military service did provide some ability to address multiple other risk factors that might confound the association observed between military service and type 2 diabetes. Extension of the VES as a cohort study might have provided an opportunity to determine whether the observed increase in mean glucose levels signaled future type 2 diabetes risk, whether this risk was independent of other risk factors present among Vietnam veterans, and what fraction of type 2 diabetes risk is attributable to military service.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

References

ADA (American Diabetes Association). 1997. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 20(7):1183-1197.

ADA. 2003. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care 26(11):3160-3167.

ADA. 2007. Diagnosis and classification of diabetes mellitus. Diabetes Care 30(Suppl 1):S42-S47.

AFHS (Air Force Health Study). 2000. An epidemiologic investigation of health effects in Air Force personnel following exposure to herbicides. 1997 follow-up examination results. Reston, VA: Science Application International Corporation. Doc. F41624-96-C1012.

Alberti, K. G., and P. Z. Zimmet. 1998. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: Diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabetic Medicine 15(7):539-553.

Bertazzi, P. A., I. Bernucci, G. Brambilla, D. Consonni, and A. C. Pesatori. 1998. The Seveso studies on early and long-term effects of dioxin exposure: A review. Environmental Health Perspectives 106(Suppl 2):625-633.

Calvert, G. M., M. H. Sweeney, J. Deddens, and D. K. Wall. 1999. Evaluation of diabetes mellitus, serum glucose, and thyroid function among United States workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Occupational and Environmental Medicine 56(4):270-276.

CDC (Centers for Disease Control and Prevention). 1989a. Health status of Vietnam veterans. Synopsis. Vol. 1. Atlanta, GA: CDC. http://www.cdc.gov/nceh/veterans/default1c.htm (accessed June 7, 2007).

CDC. 1989b. Health status of Vietnam veterans. Telephone interview. Vol. 2. Atlanta, GA: CDC. http://www.cdc.gov/nceh/veterans/default1c.htm (accessed June 7, 2007).

CDC. 1989c. Health status of Vietnam veterans. Medical examination. Vol. 3. Atlanta, GA: CDC. http://www.cdc.gov/nceh/veterans/default1c.htm (accessed June 7, 2007).

CDC. 2005. National diabetes fact sheet. http://www.cdc.gov/diabetes/pubs/pdf/ndfs_2005.pdf (accessed March 7, 2007).

CDVA (Commonwealth Department of Veterans’ Affairs). 1998. Morbidity of Vietnam veterans: A study of the health of Australia’s Vietnam veteran community. Male Vietnam veterans survey and community comparison outcomes. Vol. 1. Canberra, Australia: Department of Veterans’ Affairs.

Cook, R. R., G. G. Bond, R. A. Olson, and M. G. Ott. 1987. Update of the mortality experience of workers exposed to chlorinated dioxins. Chemosphere 16(8-9):2111-2116.

Gregg, E. W., Y. J. Cheng, B. L. Cadwell, G. Imperatore, D. E. Williams, K. M. Flegal, K. M. V. Narayan, and D. F. Williamson. 2005. Secular trends in cardiovascular disease risk factors according to body mass index in U.S. adults. Journal of the American Medical Association 293(15):1868-1874.

Henneberger, P. K., B. G. Ferris, Jr., and R. R. Monson. 1989. Mortality among pulp and paper workers in Berlin, New Hampshire. British Journal of Industrial Medicine 46(9):658-664.

Henriksen, G. L., N. S. Ketchum, J. E. Michalek, and J. A. Swaby. 1997. Serum dioxin and diabetes mellitus in veterans of Operation Ranch Hand. Epidemiology 8(3):252-258.

IOM (Institute of Medicine). 1994. Veterans and Agent Orange: Health effects of herbicides used in Vietnam. Washington, DC: National Academy Press.

IOM. 1996. Veterans and Agent Orange: Update 1996. Washington, DC: National Academy Press.

IOM. 1999. Veterans and Agent Orange: Update 1998. Washington, DC: National Academy Press.

IOM. 2000. Veterans and Agent Orange: Herbicide/dioxin exposure and type 2 diabetes. Washington, DC: National Academy Press.

IOM. 2001. Veterans and Agent Orange: Update 2000. Washington, DC: National Academy Press.

IOM. 2003. Veterans and Agent Orange: Update 2002. Washington, DC: The National Academies Press.

IOM. 2005. Veterans and Agent Orange: Update 2004. Washington, DC: The National Academies Press.

Longnecker, M. P., and J. E. Michalek. 2000. Serum dioxin levels in relation to diabetes mellitus among Air Force veterans with background levels of exposure. Epidemiology 11(1):44-48.

May, G. 1982. Tetrachlorodibenzodioxin: A survey of subjects ten years after exposure. British Journal of Industrial Medicine 39(2):128-135.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

McLenachen, D. 2005. Presumptive service connection in the VBA. Paper presented to Veterans Disability Benefits Commission. Washington, DC.

Michalek, J. E., N. S. Ketchum, and R. C. Tripathi. 2003. Diabetes mellitus and 2,3,7,8-tetrachlorodibenzo-p-dioxin elimination in veterans of Operation Ranch Hand. Journal of Toxicology and Environmental Health, Part A 66(3):211-221.

Moses, M., R. Lilis, K. D. Crow, J. Thornton, A. Fischbein, H. A. Anderson, and I. J. Selikoff. 1984. Health status of workers with past exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin in the manufacture of 2,4,5-trichlorophenoxyacetic acid: Comparison of findings with and without chloracne. American Journal of Industrial Medicine 5(3):161-182.

Pazderova-Vejlupkova, J., E. Lukas, M. Nemcova, J. Pickova, and L. Jirasek. 1981. The development and prognosis of chronic intoxication by tetrachlorodibenzo-p-dioxin in men. Archives of Environmental Health 36(1):5-11.

Pesatori, A. C., C. Zocchetti, S. Guercilena, D. Consonni, D. Turrini, and P. A. Bertazzi. 1998. Dioxin exposure and nonmalignant health effects: A mortality study. Occupational and Environmental Medicine 55(2):126-131.

Steenland, K., L. Piacitelli, J. Deddens, M. Fingerhut, and L. I. Chang. 1999. Cancer, heart disease, and diabetes in workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Journal of the National Cancer Institute 91(9):779-786.

Steenland, K., G. Calvert, N. Ketchum, and J. Michalek. 2001. Dioxin and diabetes mellitus: An analysis of combined NIOSH and Ranch Hand data. Occupational and Environmental Medicine 58(10):641-648.

Sweeney, M. H., R. W. Hornung, D. K. Wall, M. A. Fingerhut, and W. E. Halperin. 1992. Diabetes and serum glucose levels in TCDD-exposed workers. Abstract of a paper presented at the 12th International Symposium on Chlorinated Dioxins (Dioxin ‘92), Tampere, Finland, August 24-28.

Sweeney, M. H., G. Calvert, G. A. Egeland, M. A. Fingerhut, W. E. Halperin, and L. A. Piacitelli. 1996. Review and update of the results of the NIOSH medical study of workers exposed to chemicals contaminated with 2,3,7,8-tetrachlorodibenzodioxin. Paper presented at the symposium, Dioxin Exposure and Human Health—An Update, Berlin, Germany, June 17.

Sweeney, M. H., G. Calvert, G. A. Egeland, M. A. Fingerhut, W. E. Halperin, and L. A. Piacitelli. 1997. Review and update of the results of the NIOSH medical study of workers exposed to chemicals contaminated with 2,3,7,8-tetrachlorodibenzodioxin. Teratogenesis, Carcinogenesis, and Mutagenesis 17(4-5):241-247.

VA (Department of Veterans Affairs). 2000. Costing of regulation RIN 2900-AK63—Disease associated with exposure to certain herbicide agents: Type 2 diabetes. Washington, DC: Office of Resource Management, Department of Veterans Affairs.

VA. 2001a. Disease associated with exposure to certain herbicide agents: Type 2 diabetes. Proposed rule. Federal Register 66(8):2376-2380.

VA. 2001b. Disease associated with exposure to certain herbicide agents: Type 2 diabetes. Rules and regulations. Federal Register 66(89):23166-23169.

VA. 2007. National Cemetery Administration. Benefits Programs, and Departmental Administration, Congressional Submission, FY 2008. Vol. 2. http://www.va.gov/budget/summary/VolumeIINationalCemeteryAdministrationBenefitsProgramsandDepartmentalAdmin.pdf (accessed June 6, 2007).

VBA (Veterans Benefits Administration). 2006. Annual benefits report. Fiscal year 2005. Washington, DC: VBA.

Vena, J., P. Boffetta, H. Becher, T. Benn, H. B. Bueno de Mesquita, D. Coggon, D. Colin, D. Flesch-Janys, L. Green, T. Kauppinen, M. Littorin, E. Lynge, J. D. Mathews, M. Neuberger, N. Pearce, A. C. Pesatori, R. Saracci, K. Steenland, and M. Kogevinas. 1998. Exposure to dioxin and nonneoplastic mortality in the expanded IARC international cohort study of phenoxy herbicide and chlorophenol production workers and sprayers. Environmental Health Perspectives 106(Suppl 2):645-653.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

Von Benner, A., L. Edler, K. Mayer, and A. Zober. 1994. Dioxin investigation program of the chemical industry professional association. Arbeitsmedizin Sozialmedizin Praventivmedizin 29(1):11-16.

Zober, A., M. G. Ott, and P. Messerer. 1994. Morbidity follow up study of BASF employees exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) after a 1953 chemical reactor incident. Occupational and Environmental Medicine 51:479-486.

CASE STUDY 10:
SPINA BIFIDA PROGRAM

This case study examines the 1996 and 2003 decisions to grant monetary compensation and health benefits to children of Vietnam and Korean War veterans with spina bifida, respectively. These decisions were based on scientific evidence for an association between Agent Orange and this developmental abnormality. This case study illustrates the issues surrounding compensation for reproductive health effects related to exposures incurred during military service. As spina bifida is a condition that affects the children of veterans, it is not a presumptive decision for veterans; however, the children of Vietnam and Korean War veterans are covered by a VA program.

Reproductive Effects of Military Service

Two categories of reproductive effects for exposures to toxic agents have been considered by the IOM committees charged with evaluating the evidence for adverse health effects associated with Agent Orange. The first category relates to the reproductive health of the exposed men and women and includes conditions that affect fertility and impaired ability to conceive and/or to bear live children. The second category relates to developmental effects in the offspring of exposed individuals, including birth defects, growth retardation, and childhood cancers. Exposure to certain toxic agents has long been accepted as leading to developmental abnormalities in the offspring of women. However, the role of paternal exposures in the etiology of developmental outcomes has been more challenging to understand. The Agent Orange reports review the biological plausibility for paternal exposure leading to developmental abnormalities and generally find evidence from both animal and human studies to support the potential for male-mediated developmental toxicity (IOM, 1994, 1996).

The Need for a Spina Bifida Program

The preceding case studies examine presumptions which serve to fill important evidentiary gaps (either gaps for exposure or gaps for association). The program related to spina bifida departs from this pattern in important ways. Strictly speaking, the program for spina bifida is not based on a presumption; however, the program operates in a similar manner as those based on presumptions in the type of evidence that a veteran (and their offspring) is required to produce to claim compensation. The reason for the program for spina bifida is that the existing compensation structure within the Department of Veterans Affairs (VA) does not provide a mechanism for compensating an individual other than the veteran; that is, children with developmental consequences of toxic exposures incurred by the veteran cannot be compensated by existing VA mechanisms for presumptions. This program creates a specific exception, allowing for compensation for one type of developmental effect (i.e., spina bifida) in specific populations (i.e., children of Vietnam and Korea veterans). Therefore, this program exists to fill a gap in legal authority and policy rather than a gap in evidence.

It is important to note that this narrowly tailored program for spina bifida does not address the more general ongoing policy concern that VA is not able to compensate the adverse health

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

consequences to offspring of veterans due to exposures they may have incurred during military service. Through the presumptive process, VA can only compensate a veteran.

A Brief History of the Spina Bifida Program

Reproductive health effects were evaluated by the first IOM Agent Orange committee (IOM, 1994). The Committee concluded that male exposure to toxins could plausibly be linked to adverse developmental consequences in their offspring, stating that

The animal and human data indicate that the exposure of the male to various toxic agents may increase the risk of the full spectrum of adverse developmental endpoints from fetal loss to cancer.

(IOM, 1994, p. 595)


However, the committee found the evidence in support of an association between Agent Orange and a range of birth defects to be “inadequate/insufficient” (IOM, 1994, p. 6). In VA’s Federal Register response to the consideration of developmental toxicities associated with Agent Orange exposure, VA noted that there was no mechanism within the existing VA compensation structure to award benefits to any party other than the veteran; providing compensation to children of veterans with developmental effects attributable to Agent Orange exposure would require additional legislative action (VA, 1994, p. 346).

In the 1996 Agent Orange report (IOM, 1996), the committee included a specific evaluation of the evidence for the association between Agent Orange and spina bifida as “limited/suggestive.” This classification of the strength of evidence was based on the review of three studies in Vietnam veterans that, although limited in their ability to completely control the effects of bias, were deemed by the committee to be of high quality and demonstrate a consistent pattern of results (IOM, 1996). The three studies are summarized as follows.

The CDC VES surveyed Vietnam and non-Vietnam veterans and found that Vietnam veterans were more likely to report central nervous system defects in their offspring than non-Vietnam veterans (OR 2.3; 95% CI 1.2-4.5) (CDC, 1989, p. 23). A substudy that attempted to validate these findings with birth records failed to confirm the results, but the substudy was limited by differential participation between Vietnam and non-Vietnam veterans and by the difficulty in validating negative responses (CDC, 1988).

The CDC Birth Defects Study was a case-control study utilizing a population-based birth defects registry in the Atlanta, Georgia area (Erickson et al., 1984a,b). Service in Vietnam was not associated with risk of spina bifida among the offspring of veterans; however, when an exposure opportunity index was used (based on interviews that evaluated which types of activities the veteran engaged in during military service), those veterans with the highest estimated level of exposure to Agent Orange had the highest risk of having children with spina bifida (OR 2.7; 95% CI 1.2-6.2) (Erickson et al., 1984a,b, as referenced in IOM, 1996, p. 9). This study was limited by the low response rates among both cases and controls and the lag between the birth of the offspring and exposure assessment.

The Ranch Hand study of Air Force personnel involved in herbicide spraying found excess cases of neural tube defects among offspring of the Ranch Hands, with two cases of spina bifida occurring among those with the highest level of exposure, and one case of spina bifida and one of anencephaly occurring among the low-exposure group. No cases of neural tube defects were observed in the nonexposed group (P = .04) (IOM, 1996, p. 9; Wolfe et al., 1995).

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

In 1996, VA noted the findings in the IOM report and again stated that providing compensation to anyone other than the veteran would require enabling legislation by Congress (VA, 1996). Public Law 104-204 (Departments of Veterans Affairs and Housing and Urban Development, and Independent Agencies Appropriation Act, 1997. 104th Cong., 2d Sess.) was passed in 1996 and authorized benefits for children born to Vietnam veterans with spina bifida. Additional legislation in 2000 established benefits for “children of women Vietnam veterans with certain birth defects” (Veterans Benefits and Healthcare Improvement Act of 2000. Public Law 106-419 § 401. 106th Cong., 2d Sess.). This law provided benefits to children of female veterans that covered a broad range of defects potentially attributable to maternal exposure during Vietnam service; however, the law excluded defects that were the result of familial predisposition or of injury suffered at birth. In 2003, these benefits were extended to children of veterans of the Korean War (Veterans Benefits Act of 2003. Public Law 108-103. 108th Cong., 2d Sess.).

Lessons Learned

With the exceptions of the legislative actions to establish the spina bifida program as well as the program for the children of female Vietnam veterans, there continues to be no overall mechanism for compensating the offspring of veterans for health consequences attributable to maternal and paternal exposures incurred during military service. Toxic exposures that occur during military service have the potential to cause adverse developmental effects, and each of the IOM Agent Orange reports (IOM, 1994, 1996, 1999, 2001, 2003, 2005) has described biologically plausible mechanisms for these effects in the offspring of both exposed female and male veterans. Given VA’s interest in compensating veterans for adverse health effects incurred as a result of military service and the possibility that such effects may extend to the health of veterans’ offspring, the absence of a clear and consistent mechanism and policy on compensating potentially affected offspring is notable. The need for a clear policy statement will continue to grow as VA considers the health effects of military service in the large population of reproductive-aged female and male veterans, especially with the growing number of women who serve in the military.

Although the public laws providing compensation for particular categories of offspring with birth defects may have been expedient for these affected individuals, the approach of addressing the more general policy gap described above with these VA programs runs counter to principles of consistency and equity that should inform the approach for presumptions. Any new adverse reproductive consequences of Agent Orange exposure identified in the IOM reports would again require legislative action for these specific effects in order for compensation to be granted to the offspring of veterans; the administrative route that has applied to all other Agent Orange presumptions is not available for reproductive consequences of exposure at present.

It is worthy of note that the evidence standard for establishing the program for spina bifida was “limited/suggestive evidence of an association” not the more rigorous “sufficient” classification. The challenges in using this lower evidence classification as the basis for VA’s presumptions have been described in case studies related to Agent Orange and prostate cancer and type 2 diabetes.

References

CDC (Centers for Disease Control and Prevention). 1988. Health status of Vietnam veterans: Reproductive outcomes and child health. Vol. 3. Journal of the American Medical Association 259(18):2715-2719.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×

CDC. 1989. Health status of Vietnam veterans: Reproductive outcomes and child health. Vol. 5. Atlanta, GA: Centers for Disease Control. http://www.cdc.gov/nceh/veterans/default1c.htm (accessed July 1, 2007).

Erickson, J. D., J. Mulinare, P. W. McClain, T. G. Fitch, L. M. James, A. B. McClearn, and M. J. Adams, Jr. 1984a. Vietnam veterans’ risks for fathering babies with birth defects. Atlanta, GA: Centers for Disease Control.

Erickson, J. D., J. Mulinare, P. W. McClain, T. G. Fitch, L. M. James, A. B. McClearn, and M. J. Adams, Jr. 1984b. Vietnam veterans’ risks for fathering babies with birth defects. Journal of the American Medical Association 252(7):903-912.

IOM (Institute of Medicine). 1994. Veterans and Agent Orange: Health effects of herbicides used in Vietnam. Washington, DC: National Academy Press.

IOM. 1996. Veterans and Agent Orange: Update 1996. Washington, DC: National Academy Press.

IOM. 1999. Veterans and Agent Orange: Update 1998. Washington, DC: National Academy Press.

IOM. 2001. Veterans and Agent Orange: Update 2000. Washington, DC: National Academy Press.

IOM. 2003. Veterans and Agent Orange: Update 2002. Washington, DC: The National Academies Press.

IOM. 2005. Veterans and Agent Orange: Update 2004. Washington, DC: The National Academies Press.

VA (Department of Veterans Affairs). 1994. Disease not associated with exposure to certain herbicide agents. Notice. Federal Register 59(2):341-346.

VA. 1996. Disease not associated with exposure to certain herbicide agents. Federal Register 61(154):41442-41449.

Wolfe, W. H., J. E. Michalek, J. C. Miner, A. J. Rahe, C. A. Moore, L. L. Needham, and D. G. Patterson, Jr. 1995. Paternal serum dioxin and reproductive outcomes among veterans of Operation Ranch Hand. Epidemiology 6(1):17-22.

Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 592
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 593
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 594
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 595
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 596
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 597
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 598
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 599
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 600
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 601
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 602
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 603
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 604
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 605
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 606
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 607
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 608
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 609
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 610
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 611
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 612
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 613
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 614
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 615
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 616
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 617
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 618
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 619
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 620
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 621
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 622
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 623
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 624
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 625
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 626
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 627
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 628
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 629
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 630
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 631
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 632
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 633
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 634
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 635
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 636
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 637
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 638
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 639
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 640
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 641
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 642
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 643
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 644
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 645
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 646
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 647
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 648
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 649
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 650
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 651
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 652
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 653
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 654
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 655
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 656
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 657
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 658
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 659
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 660
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 661
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 662
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 663
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 664
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 665
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 666
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 667
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 668
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 669
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 670
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 671
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 672
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 673
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 674
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 675
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 676
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 677
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 678
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 679
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 680
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 681
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 682
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 683
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 684
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 685
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 686
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 687
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 688
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 689
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 690
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 691
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 692
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 693
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 694
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 695
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 696
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 697
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 698
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 699
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 700
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 701
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 702
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 703
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 704
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 705
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 706
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 707
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 708
Suggested Citation:"Appendix I: Case Studies." Institute of Medicine. 2008. Improving the Presumptive Disability Decision-Making Process for Veterans. Washington, DC: The National Academies Press. doi: 10.17226/11908.
×
Page 709
Next: Appendix J: Causation and Statistical Causal Methods »
Improving the Presumptive Disability Decision-Making Process for Veterans Get This Book
×
Buy Paperback | $85.50 Buy Ebook | $69.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The United States has long recognized and honored the service and sacrifices of its military and veterans. Veterans who have been injured by their service (whether their injury appears during service or afterwards) are owed appropriate health care and disability compensation. For some medical conditions that develop after military service, the scientific information needed to connect the health conditions to the circumstances of service may be incomplete. When information is incomplete, Congress or the Department of Veterans Affairs (VA) may need to make a "presumption" of service connection so that a group of veterans can be appropriately compensated.

The missing information may be about the specific exposures of the veterans, or there may be incomplete scientific evidence as to whether an exposure during service causes the health condition of concern. For example, when the exposures of military personnel in Vietnam to Agent Orange could not be clearly documented, a presumption was established that all those who set foot on Vietnam soil were exposed to Agent Orange. The Institute of Medicine (IOM) Committee was charged with reviewing and describing how presumptions have been made in the past and, if needed, to make recommendations for an improved scientific framework that could be used in the future for determining if a presumption should be made. The Committee was asked to consider and describe the processes of all participants in the current presumptive disability decision-making process for veterans. The Committee was not asked to offer an opinion about past presumptive decisions or to suggest specific future presumptions.

The Committee heard from a range of groups that figure into this decision-making process, including past and present staffers from Congress, the VA, the IOM, veterans service organizations, and individual veterans. The Department of Defense (DoD) briefed the Committee about its current activities and plans to better track the exposures and health conditions of military personnel. The Committee further documented the current process by developing case studies around exposures and health conditions for which presumptions had been made. Improving the Presumptive Disability Decision-Making Process for Veterans explains recommendations made by the committee general methods by which scientists, as well as government and other organizations, evaluate scientific evidence in order to determine if a specific exposure causes a health condition.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!