veterans compared with the controls (total cancers: males, OR 0.90, 95% CI 0.81-1.01; females, OR 1.11, 95% CI 0.78-1.57).

Macfarlane et al. (2003) conducted a cohort study among 51,721 UK servicemembers deployed to the Persian Gulf and 50,755 nondeployed servicemembers using the National Health Service Cancer Register to identify first diagnoses of malignant cancer through July 31, 2002. The rate ratio for unspecified cancer was 0.99 (95% CI 0.83-1.17), after adjusting for sex, age, service branch, and rank. In subgroups of the cohort who participated in morbidity surveys and provided information on smoking and alcohol use, the adjusted rate ratio for all cancers was 1.12 (95% CI 0.86-1.45).

Updated and Supplemental Literature

Primary Studies

The Update committee identified one new primary study of brain cancer mortality. It also identified three studies of hospitalization or incidence, one mortality study, and one combined study where cancer was specifically assessed.

Brain Cancer

In continued mortality follow-up through 2004 of the 621,902 Gulf War deployed veterans and 746,248 nondeployed veterans originally studied by Kang and Bullman (2001), Barth et al. (2009) focused on mortality from neurological causes, that is, amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS), Parkinson’s disease, and primary brain cancer. The cases included in the Bullman et al. study (2005) described above, also were identified from this cohort and were included in the Barth et al. study. A total of 144 cases of brain cancer were identified among the deployed veterans and 228 among the nondeployed for a mortality rate ratio (RR) of 0.90 (95% CI 0.73-1.11), adjusted for race, branch of service, type of unit, age, marital status at entry to follow-up, and sex. Within the Gulf War cohort, exposure to nerve agents from the Khamisiyah explosion for 2 or more days (Winkenwerder, 2002) and exposure to oil-well fire smoke (Rostker, 2000) were both positively associated with risk of brain cancer mortality when modeled simultaneously (adjusted rate ratios 2.71, 95% CI 1.25-5.87 and 1.81, 95% CI 1.00-3.27, respectively). Of the 43 brain cancer cases exposed to oil-well fires, 20 were also exposed for at least 1 day at Khamisiyah (S. Barth, Department of Veterans Affairs, personal communication, November 30, 2009).

Medical records were obtained for 236 of the 372 cases of brain cancer (63%). The record review resulted in 204 confirmed cases, 13 probable cases, and 19 misclassified cancers. The risk of dying from brain cancer did not change with the removal of the 19 misclassified cancers.

Hospitalization Studies

Gray et al. (2000) conducted an expanded analysis of their original 1996 study of hospitalizations to include US Gulf War veterans (n = 652,979) and nondeployed veterans (random selection of n = 652,922 from 2,912,737 total) who had separated from the armed services and those who served in the National Guard or reserve. Hospitalization data from the DoD, the VA, and California Office of Statewide Health Planning and Development over the period August 1, 1991, through December 31, 1994, were assessed separately. There was no evidence of increased hospitalization from neoplasms among the Gulf War veterans compared to the nondeployed veterans in any of the three hospital systems.

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