2
EVIDENCE REGARDING AMYOTROPHIC LATERAL SCLEROSIS IN VETERANS

As described in Chapter 1, the committee conducted a literature search to identify studies relevant to amyotrophic lateral sclerosis (ALS) in veterans. The committee reviewed and evaluated all peer-reviewed, published studies on ALS in the veteran population. Those studies served as the basis of the committee’s conclusion, and they are critiqued in this chapter.

STUDIES ON AMYOTROPHIC LATERAL SCLEROSIS IN THE VETERAN POPULATION

In this section, the committee evaluates the studies on ALS in the veteran population that it considered to be the primary studies on which it based its conclusion. The studies are presented chronologically by year of publication. They are summarized in Table 2.1, and complete results are presented in Table 2.2.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature 2 EVIDENCE REGARDING AMYOTROPHIC LATERAL SCLEROSIS IN VETERANS As described in Chapter 1, the committee conducted a literature search to identify studies relevant to amyotrophic lateral sclerosis (ALS) in veterans. The committee reviewed and evaluated all peer-reviewed, published studies on ALS in the veteran population. Those studies served as the basis of the committee’s conclusion, and they are critiqued in this chapter. STUDIES ON AMYOTROPHIC LATERAL SCLEROSIS IN THE VETERAN POPULATION In this section, the committee evaluates the studies on ALS in the veteran population that it considered to be the primary studies on which it based its conclusion. The studies are presented chronologically by year of publication. They are summarized in Table 2.1, and complete results are presented in Table 2.2.

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature TABLE 2.1 Design of Epidemiology Studies of ALS in US Veterans Reference Study Population Comparison Population At Risk N (ALS cases) Reference N (ALS cases) Study Design Case-Ascertainment Methods Analysis Risk Ratios (95% Confidence Intervals) Smith et al. 2000 All active, gulf-deployed military personnel who served during the Gulf War (August 8, 1990–July 31, 1991) Non-gulf-deployed Gulf War-era veterans 551,841 (6) 1,478,704 (12) Cohort followup, August 1991-July 1997 DOD hospitalization records Cox proportional—hazards analysis 1.66(0.62–4.44); adjusted for age Kang and Bullman 2001 Active, gulf-deployed military personnel who served during the Gulf War (August 8, 1990–March 1, 1991) Non-gulf-deployed Gulf War-era veterans 621,902 (NA) 746,248 (NA) Cohort followup, May 1991–December 1997 Database records of VA and Social Security Administration Cox proportional-hazards analysis 0.59 (0.21–1.66); adjusted for age, race, sex, marital status, brach of service, unit component

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Reference Study Population Comparison Population At Risk N (ALS cases) Reference N (ALS cases) Study Design Case-Ascertainment Methods Analysis Risk Ratios (95% Confidence Intervals) Haley 2003 All active, gulf-deployed military personnel who served during the Gulf War (August 1990–April 1991) and were less than 45 years old Age-specific mortality rates from the total US population About 695,000 (17) Cohort followup, 1991–1998 VA data; personal contact and diagnosis SMR; Poisson trend test 1991–1994: 0.94 (0.26–2.41); 1995–1998:2.27 (1.27–3.88); adjusted for age Horner et al. 2003 All active, gulf-deployed military personnel who served during the Gulf War (August 2, 1990–July 31, 1991), Non-gulf-deployed Gulf War-era veterans 696,118 (40) 1,786,215 (67) Cohort followup, August 1990–August 2000 Screening of VA and DOD medical databases and benefit files by ICD-9 code for ALS or riluzole use; toll-free Age-adjusted average annual incidence; attributable risk 1.92 (1.29–2.84); adjusted for age

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Reference Study Population Comparison Population At Risk N (ALS cases) Reference N (ALS cases) Study Design Case-Ascertainment Methods Analysis Risk Ratios (95% Confidence Intervals)   including mobilized reserves and National Guard       telephone enrollment; internet notices; mass mailings to neurologists, VA centers, and veteran service organizations    

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Reference Study Population Comparison Population At Risk N (ALS cases) Reference N (ALS cases) Study Design Case-Ascertainment Methods Analysis Risk Ratios (95% Confidence Intervals) Weisskopf et al. 2005 Male participants in the ACS Cancer Prevention Study II cohort who had self-reported military service Male participants with self-report of no military service at enrollment 281,874 (217) 126,414 (63) Cohort followup, 1989–1998 Automated linkage with National Death Index Cox proportional-hazards analysis 1.53 (1.12–2.09); adjusted for age, smoking, education, alcohol intake, self-reported exposure to pesticides and herbicides, and main occupation as farmer, electrical or welding work, or food preparation NOTE: ACS=American Cancer Society; DOD=Department of Defense; ICD-9=International Classification of Diseases, ninth revision; NA=not available; SMR=standardized mortality ratio; VA=Department of Veterans Affairs.

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature TABLE 2.2 ALS and Military Service—Cohort Studies Reference Study Population Cases Estimated Relative Risk (95% Confidence Intervals) Smith et al. 2000 Non-gulf-deployed Gulf War-era veterans 12 1.0 (ref)   Gulf War-deployed military personnel (1990–1991) 6 1.66 (0.62–4.44) Kang and Bullman 2001 Non-gulf-deployed Gulf War-era veterans NA 1.0 (ref) Gulf War-deployed military personnel (1990–1991) NA 0.59 (0.21–1.66) Haley 2003 Gulf War-deployed military personnel (1990–1991), less than 45 years old       Year of diagnosis:       1991–1994 4 0.94 (0.26–2.41)a   1995–1998 13 2.27 (1.27–3.88)a Horner et al. 2003 Non-gulf-deployed Gulf War-era veterans 67 1.0 (ref)   Gulf War-deployed military personnel (1990–1991) 40 1.92 (1.29–2.84)   Unit       Nondeployed active duty 61 1.0 (ref)   Deployed active duty 31 2.15 (1.38–3.36)   Nondeployed reserves and National Guard 6 1.0 (ref)   Deployed reserves and National Guard 9 2.50 (0.88–7.07)   Service branch       Nondeployed Air Force 24 1.0 (ref)

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Reference Study Population Cases Estimated Relative Risk (95% Confidence Intervals)   Deployed Air Force 9 2.68 (1.24–5.78)   Nondeployed Army 20 1.0 (ref)   Deployed Army 21 2.04 (1.10–3.77)   Nondeployed Marine Corps 5 1.0 (ref)   Deployed Marine Corps 3 1.13 (0.27–4.79)   Nondeployed Navy 18 1.0 (ref)   Deployed Navy 7 1.48 (0.62–3.57) Weisskopf et al. 2005 Self-reported no military service 63 1.0 (ref)   Self-reported military service 217 1.53 (1.12–2.09)   Service branch       No military service 63 1.0 (ref)   Army and National Guard 97 1.54 (1.09–2.17)   Navy 65 1.87 (1.28–2.74)   Air Force 34 1.54 (0.99–2.39)   Marine Corpes 4 0.64 (0.23–1.78)   Coast Guard 3 2.24 (0.70–7.18)   Not given 14 1.08 (0.60–1.94)   Years of military service (median years)       No military service 63 1.0 (ref)   1st quintile (2) 21 1.60 (0.95–2.70)

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Reference Study Population Cases Estimated Relative Risk (95% Confidence Intervals)   2nd quintile (3) 65 1.80 (1.23–2.63)   3rd quintile (4) 54 1.49 (1.01–2.21)   4th quintile (5) 37 1.47 (0.95–2.25)   5th quintile (9) 36 1.47 (0.95–2.27) p trend=0.26   Number of wars during service       No military service 63 1.0 (ref)   0 (no wartime service) 36 1.34 (0.87–2.06)   1 156 1.57 (1.14–2.17)   2 15 1.74 (0.97–3.14)   >2 6 1.97 (0.83–4.70) p trend=0.004   Service during only one war       No military service 63 1.0 (ref)   World War II 116 1.60 (1.12–2.30)   Korean War 36 1.54 (0.92–2.60)   Vietnam War 4 1.44 (0.47–4.47) aReported as standardized mortality ratios. NOTE: ACS=American Cancer Society; CPS=Cancer Prevention Study; NA=not available, ref=reference. Smith et al. 2000 Smith et al. (2000) relied on hospitalization records and reported an imprecisely estimated increase in relative risk of ALS

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature in deployed Gulf War veterans compared with non-gulf-deployed Gulf War-era veterans (relative risk [RR], 1.66; 95% confidence interval [CI], 0.62–4.44). The study population was defined as regular active-duty military personnel deployed to the Persian Gulf War theater for one or more days from August 8, 1990-July 31, 1991. The authors obtained information on deployment dates from the Defense Manpower Data Center, Seaside, CA. The study included only 18 cases of ALS in total (six among deployed veterans) and was also limited by its exclusive use of Department of Defense (DOD) hospitalization records, inclusion of only active-duty military personnel, and no more than 6 years of followup (through July 1997). The mean duration of active service during the followup period was only 3.4 years for Gulf War veterans and 3.8 years for nondeployed veterans. This study has been superseded by others with more inclusive populations and longer periods of followup. Kang and Bullman 2001 Kang and Bullman (2001) performed a study to investigate causes of death of Gulf War veterans compared with concurrently serving veterans who were not deployed. Vital status and cause of death were determined from a variety of databases for 621,902 veterans who served in the Persian Gulf before the termination of hostilities and for 746,248 non-gulf-deployed Gulf War-era veterans, stratified by branch of service. The Gulf War veterans included in the study arrived in the Persian Gulf prior to March 1, 1991. That date was chosen in order to exclude those who arrived in the Persian Gulf theater after hostilities had ended and, therefore, might not have received “exposures of concern” (for example, chemical and biologic warfare agents, certain vaccines, and pyridostigmine bromide). The authors obtained information on deployment dates from the Defense Manpower Data Center. For disease-specific death determination, cause of death was obtained from death certificates. Followup terminated at death or on December 31, 1997, providing a maximum followup of slightly less than 7 years. The investigators found no specific cause of death that occurred with greater frequency in the veteran group that actively served in the Gulf War Theater. With respect to ALS, no excess risk was observed in the Gulf War veterans; an imprecisely

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature estimated risk ratio of 0.59 (95% CI, 0.21–1.66) adjusted for age (in years), race, sex, branch of service, unit component, and marital status was reported. The authors did not provide the number of cases of ALS that they identified. It could be argued that the risk reduction (the point estimate was less than 1) is of the same order of magnitude as the increases in relative risk reported in other studies described in this chapter. However, this study was uninformative because of its wide confidence limits and the short followup period (it encompassed only about 7 years after the war). Haley 2003 Haley (2003) found an excess incidence of ALS among deployed Gulf War veterans in comparison with the expected incidence based on US vital statistics. The study population was defined as the approximately 695,000 members of the Army, Navy, Air Force, Marines, and Coast Guard who served in the Southwest Asia Theater of Operations from August 1990-April 1991. The analysis spanned 1991–1998. In the first half of that period, the increase in incidence was not apparent (standardized mortality ratio [SMR], 0.94; 95% CI, 0.26–2.41), but from 1995 to 1998 the increase was more than double (SMR, 2.27; 95% CI, 1.27–3.88). Incidence peaked in the final year of analysis (1998: SMR, 3.19; 95% CI, 1.03–7.43) and increased markedly from 1991 to 1998 (p=0.05). Although the study used passive and active means of case ascertainment similar to those of Horner et al. (2003) (described below), it differed in several key respects: it restricted cases to those below the age of 45 years (instead of all ages); accepted only clinically definite cases of ALS (instead of clinically definite, probable, and suspect cases); used 8 years of followup (instead of 10); and used as a comparison population the age-adjusted rates from US mortality statistics (instead of age-adjusted rates in non-gulf-deployed Gulf War-era veterans). The major criticism of this study is its use of mortality statistics from the general population to estimate the “expected” incidence (Armon 2004c). Mortality may underestimate incidence of ALS (Armon 2004a). In addition, age at onset of ALS was defined as age at death minus 2 years. The calculations for the expected number of cases were age-adjusted but were not sex-adjusted. Because the majority of the deployed Gulf War veterans were men

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature and mortality from ALS is higher in men (Armon 2004a), those calculations may have resulted in an overestimation of the SMR. In summary, Haley’s analyses may have underestimated the expected rates in the comparison population, thereby making SMRs appear higher. Because of the limitations discussed above, the meaning of Haley’s study results remains uncertain. Horner et al. 2003 Coffman et al. 2005 Horner and colleagues conducted a nationwide epidemiologic study with the goal of identifying all new occurrences of ALS among deployed and nondeployed veterans in the 10 years after the end of the Gulf War (Horner et al. 2003). All active-duty and mobilized Reserves and National Guard personnel were eligible if they had served for at least 1 month at any time during the period August 2, 1990-July 31, 1991 (defined as the Gulf War period). Military personnel were considered deployed if they were in the gulf region or received hazardous duty pay during the Gulf War. All other individuals in the study population were defined as nondeployed, although they may have been stationed outside of the US. Information on deployment status and dates was obtained from the Defense Manpower Data Center. The authors used active and passive methods of case ascertainment; active methods included screening of inpatient, outpatient, and pharmacy medical databases of the Department of Veterans Affairs (VA) and DOD. For the passive methods, they established a toll-free telephone number and conducted solicitations through relevant Internet sites and mass mailings of study brochures to practicing VA neurologists and to members of the American Academy of Neurology. For all cases originally identified, they determined eligibility; and for all cases included in the analyses presented, the diagnosis of ALS was verified with medical-record or mortality-record review. Among nearly 2.5 million eligible military personnel, they identified and confirmed 107 cases of ALS, which yielded an age-adjusted average annual incidence of 0.43 per 100,000 persons. Most of the cases (about 86%) were found through active ascertainment methods. Nearly 700,000 of the 2.5 million military personnel had been deployed to the gulf region and are referred to as deployed military personnel hereafter. Comparing deployed

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature with nondeployed personnel, Horner et al. observed an almost doubled age-adjusted risk of ALS (RR, 1.92; 95% CI, 1.29–2.84). The risk of ALS was increased among deployed active-duty veterans (RR, 2.15; 95% CI, 1.38–3.36). There was also an increase in the risk of ALS among deployed veterans who served in the reserves or National Guard (RR, 2.50; 95% CI, 0.88–7.07), although the estimate for this smaller group was less precise. Further subgroup analyses by service branch confirmed increased risks for Air Force personnel (RR, 2.68; 95% CI, 1.24–5.78) and Army personnel (RR, 2.04; 95% CI, 1.10–3.77); for other service branches, the numbers of ALS cases observed were small and rendered the comparison of deployed with nondeployed personnel less informative. Overall, the authors estimated the excess risk attributable to deployment to be 18% (95% CI 4.9%–29.4%), and the risk difference to be 0.32 per 100,000 persons per year. The foremost limitation of this study lies in the potential for underascertainment of cases, which, if different between deployed and nondeployed groups, may have resulted in overestimation or underestimation of the risk. For example, underascertainment might occur if nondeployed veterans with ALS were less likely to be listed in the record systems used for active case identification or had less incentive to participate in the study than deployed personnel. Because of the rarity of ALS, differential underascertainment of even a small number of cases might exaggerate differences between groups considerably. The concern that ALS would more likely be underascertained in nondeployed than in deployed personnel, thus creating the impression that nondeployed veterans suffered a lower rate of the disease, was raised in a letter to the editor of Neurology by Carmel Armon, an ALS researcher (Armon 2004b). In a followup of the study of Horner et al., Coffman and colleagues (2005) formally assessed the potential impact of such a bias by using three statistical modeling methods known as capture-recapture methods: log-linear modeling, the sample coverage approach, and ecologic modeling. On the basis of the three modeling approaches, the investigators concluded that a modest underascertainment of cases among nondeployed military personnel might have occurred, whereas little or no underascertainment was apparent in deployed personnel. However, even after correcting the ALS rates among

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature the nondeployed for underascertainment, the investigators still estimated an increased age-adjusted risk of ALS in the deployed military personnel (RR, 1.77; lower bound, 1.21 with the log-linear model). Those results confirmed the original findings of Horner et al. (2003). However, their validity relies on the assumption that the number of underascertained cases in nondeployed military personnel derived from the models is a correct approximation of the number of cases missed. Armon (2004b) pointed out that the Horner et al. estimated rates for nondeployed personnel were lower than the age-adjusted rates reported in a high-quality population-based study of ALS conducted in Washington state by McGuire et al. (1996). Armon presented a table with ALS incidence rates standardized to the 1990 US population for deployed and nondeployed personnel (3.6 and 1.4 per 100,000 people per year, respectively; reported by Horner et. al.) and for the population from McGuire et al. (2.1 per 100,000 people per year; calculated by Armon). However, dividing the incidence rates in deployed personnel by the incidence rates in the McGuire et al. population (serving as the comparison group), the RR is 1.7. This result is similar to the RR estimates using the capture-recapture analyses (Coffman et al. 2005). Using the McGuire et al. rates lends further support to the Horner et al. results because even if there was underascertainment of ALS among the nondeployed military personnel (resulting in lower rates in that comparison group), the rates among deployed personnel were still higher than those reported by McGuire et al. The committee discussed the implications of the decision by Horner et al. to use persons rather than person-time to calculate risk ratios. In order for that approach not to have resulted in bias, the investigators had to make the following assumptions: (1) that the disease is rare, and removing cases from the denominator will affect followup time minimally at worst, and (2) that the cohort is young and healthy enough not to experience competing risks or loss, which would remove subjects from the person-time denominator during followup (that is, that loss to followup or right censoring is minimal). The committee believes that those assumptions are reasonable for this cohort and the 10-year followup time after the Gulf War.

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Weisskopf et al. 2005 Weisskopf et al. (2005) published the results of an analysis of a cohort study in which they examined the relationship between self-report of military service and death from ALS. The study subjects were drawn from male participants in an ongoing cohort, the American Cancer Society’s Cancer Prevention Study II (CPS II) cohort. The CPS II cohort, assembled in 1982, included a total of 508,334 men who completed a four-page questionnaire at baseline. The CPS II cohort was originally designed as a prospective cohort study to examine environmental and lifestyle risk factors in the etiology of cancer. The information collected also included data on military service. In the self-administered questionnaire, military service was determined by the question “Were you in the US armed services?” For those who reported military service, followup questions were included to collect information on the branch, the location, and the years of service. At baseline, persons who did not respond to the military-service questions were excluded (N=6,854). Because monitoring for the occurrence of ALS during followup was not included as part of the CPS II, the authors used ALS mortality as the outcome. To ascertain ALS mortality, the CPS II cohort records were linked to the National Death Index and death certificates were obtained. The CPS II cohort was established in 1982, but ALS deaths were not coded separately on death certificates before 1989. Therefore, the statistical analysis included followup (from 1989 to 1998) of only the cohort members alive in 1989 (N=408,288). During the followup period, 280 deaths of which ALS was the underlying or contributory cause were identified in the cohort of more than 400,000 participants. Of the 280 deaths, 217 were among 281,874 men who reported military service, and 63 were among 126,414 men who reported no military service. Cox proportional-hazards analysis with adjustment for differences in age, smoking history, education, alcohol consumption, self-reported exposure to pesticides and herbicides, and several main lifetime occupations (farming, jobs involving electrical work or welding, and food preparation) showed that persons who reported any military service were 1.5 times more likely to have died with a notation of ALS on their death certificates as those who reported no military service. For different branches of the military, risk

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature appeared higher among those who reported serving in the Army and National Guard (RR, 1.54; 95% CI, 1.09–2.17) and in the Navy (RR, 1.87; 95% CI, 1.28–2.74). Increases in risk (although not reaching conventional levels of significance) were also found in those who served in the Air Force and Coast Guard. There was no evidence of an increased risk for those people who served in the Marine Corps. The authors also examined ALS mortality according to the number of years of military service; there was an increase in risk regardless of the number of years of service. Finally, they examined the risk in relation to service during periods of war (that is, World War II, Korea, or Vietnam) as well as the number of wars during the period of service. Increase in ALS mortality was found to be similar across the three wars, with the RR reaching conventional levels of statistical significance only for service during World War II, the subgroup with the largest sample size. There was some evidence of a trend toward increasing risk as the number of wars during the service period increased from zero to more than two. Although the study was large, the interpretation of the results is limited by the use of self-report of military service and the use of ALS mortality as a proxy for ALS incidence. The first limitation is mitigated somewhat because most people who have been in the military are likely to remember their service, and the report of military service took place some time prior to the onset of disease and is thus independent of subsequent ALS status. However, the use of mortality data rather than incidence data is a weakness. Nevertheless, the fact that the median survival in ALS is 3 years means that mortality data for ALS can be a good proxy for incidence data. The use of death certificates to ascertain death from ALS, however, is a potential problem because of inaccuracy of death certificate reporting of ALS. However, given the inevitable progression of ALS, it is unlikely that a large number of cases would be miscoded on death certificates. Even if some miscoding occurred, it probably was not related to military service inasmuch as the data were recorded before the putative link between service in the Gulf War and ALS had been widely reported. Weisskopf et al. have reported on the first study of ALS and military service in which sufficient data to adjust for lifestyle factors and occupation

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature were available. Such adjustments did not materially change the effect estimates. A letter to the editor of the journal Neurology after publication of the Weisskopf et al. study raised a concern about the use of a cohort of subjects originally recruited for a cancer study. The authors of the letter suggested that the volunteer cohort might have been healthier than nonparticipants and the occurrence of ALS might have been underestimated (Horner et al. 2005). Weisskopf et al. replied that they were not concerned that the association that they found between military service and ALS mortality could have been due to the self-selection of healthy volunteers. They supported their view by showing that the age-specific death rates from ALS in the cohort were similar to those in the US population. The study by Weisskopf et al. is the first to suggest a relationship between military service before the Gulf War and ALS mortality. While the study does have limitations inherent in an analysis of a cohort assembled for other purposes, the findings are intriguing. The implication is that military service in general—not confined to exposures specific to the Gulf War—is related to the development of ALS. The findings, if validated in other studies, suggest that exposures during military service, even among those with no wartime service, might be responsible. CONCLUSION The committee identified one high-quality cohort study that examined the relationship between serving in the military and development of ALS (Weisskopf et al. 2005). The study authors reported an increased risk of ALS among men who served in the military in the period 1910–1982, regardless of which war or wars the men served in. Risk was also increased among men who served in the military during nonwartime periods. Although the study has some limitations (such as lack of confirmation of ALS diagnosis because the data came from death certificates), overall it was a well-designed and well-conducted study. It adequately controlled for confounding factors (age, cigarette use, alcohol consumption, education, self-reported exposure to pesticides and herbicides, and several main lifetime occupations).

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Three other studies corroborate the findings of the Weisskopf et al. study regarding service in the military and development of ALS. Horner et al. (2003), Haley (2003), and Smith et al. (2000) reported associations between deployment to the Gulf War and ALS. Although the Horner et al. study was well conducted and provides some evidence regarding military service and development of ALS, it is limited by the potential for underascertainment of cases in the comparison group. The Haley study had additional methodologic limitations. Mortality statistics from the general population were used to estimate the expected incidence, and the calculation of expected incidence was age-adjusted but not sex-adjusted. Both limitations may have underestimated the expected incidence of ALS. A study using hospitalization records of Gulf War veterans still on active duty suggested an increased risk of ALS (Smith et al. 2000). Several problems are related to the design of the study, and the committee found it to be of little value in its assessment. Finally, a mortality study of Gulf War veterans did not report an increase in the risk of ALS (Kang and Bullman 2001); again, multiple methodologic problems limit its value in the committee’s assessment. On the basis of its evaluation of the literature, the committee concludes that there is limited and suggestive evidence of an association between military service and later development of ALS. REFERENCES Armon C. 2004a. Amyotrophic lateral sclerosis. In: Nelson LM, Tanner CM, Van Den Eeden SK, McGuire VM, Editors. Neuroepidemiology: From Principles to Practice. New York: Oxford University Press. Pp. 162–187. Armon C. 2004b. Occurrence of amyotrophic lateral sclerosis among Gulf War veterans. Neurology 62(6):1027. Armon C. 2004c. Excess incidence of ALS in young Gulf War veterans. Neurology 63(10):1986–1987.

OCR for page 19
Amyotrophic Lateral Sclerosis in Veterans: Review of the Scientific Literature Coffman CJ, Horner RD, Grambow SC, Lindquist J. 2005. Estimating the occurrence of amyotrophic lateral sclerosis among Gulf War (1990–1991) veterans using capture-pecapture methods. Neuroepidemiology 24(3):141–150. Haley RW. 2003. Excess incidence of ALS in young Gulf War veterans. Neurology 61(6):750–756. Horner RD, Feussner JR, Kasarskis EJ. 2005. Prospective study of military service and mortality from ALS. Neurology 65(1):180–181. Horner RD, Kamins KG, Feussner JR, Grambow SC, Hoff-Lindquist J, Harati Y, Mitsumoto H, Pascuzzi R, Spencer PS, Tim R, Howard D, Smith TC, Ryan MA, Coffman CJ, Kasarskis EJ. 2003. Occurrence of amyotrophic lateral sclerosis among Gulf War veterans. Neurology 61(6):742–749. Kang HK, Bullman TA. 2001. Mortality among US veterans of the Persian Gulf War: 7-year follow-up. American Journal of Epidemiology 154(5):399–405. McGuire V, Longstreth WT Jr, Koepsell TD, van Belle G. 1996. Incidence of amyotrophic lateral sclerosis in three counties in western Washington state. Neurology 47(2):571–573. Smith TC, Gray GC, Knoke JD. 2000. Is systemic lupus erythematosus, amyotrophic lateral sclerosis, or fibromyalgia associated with Persian Gulf War service? An examination of Department of Defense hospitalization data. American Journal of Epidemiology 151(11):1053–1059. Weisskopf MG, O’Reilly EJ, McCullough ML, Calle EE, Thun MJ, Cudkowicz M, Ascherio A. 2005. Prospective study of military service and mortality from ALS. Neurology 64(1): 32–37.