The Canadian Department of National Defense commissioned Goss Gilroy, Inc., to assess the prevalence of health outcomes in Canadian forces deployed to the Gulf War (Goss Gilroy, 1998). In 1997, a questionnaire was administered to 3113 Gulf War deployed veterans and 3439 active but nondeployed Canadian veterans. Among the respondents 20-44 years of age, 0.8% of the deployed veterans and 0.5% of the nondeployed reported any cancer, and among the 45-65 year olds, the prevalence of any cancer was 4.2% in the deployed veterans and 2.5% in the nondeployed veterans.

Ishoy et al. (1999a) conducted a cross-sectional clinical examination study during 1997. Participants included Danish servicemembers deployed to the Gulf (n = 686) and eligible but nondeployed Danish servicemembers (n = 231). Skin cancer was assessed; however, results were not presented since the p-value comparing the two groups was greater than 0.05.

Kang et al. (2000) conducted a health survey comparing self-reported health outcomes for a population based sample of 15,000 deployed and 15,000 nondeployed US Gulf War veterans. Based on the responses from 11,441 deployed and 9476 nondeployed veterans, they estimated the population prevalence rates of various medical conditions. The estimated population prevalence for skin cancer was 1.5% for the deployed and 1.4% for the nondeployed veterans, with a significant difference (rate difference 0.15, 95% CI 0.11-0.19).

A population based survey of Kansas veterans deployed to the gulf (n = 1548) and nondeployed (n = 482) examined the health outcomes of deployment through automated telephone interviews to define and establish prevalence of “Gulf War illness” symptom complex (Steele, 2000). The incidence of disease diagnosed or treated by a physician was tracked through 1998. The OR for cancers, not including skin cancer (n = 18) was 1.21 (95% CI 0.40-3.69), and the OR for skin cancer (n = 23) was 1.17 (95% CI 0.47-2.90), adjusting for sex, age, income, and level of education.

In 1999, McCauley et al. (2002) conducted a telephone survey of 2918 active or reserve Army or National Guard veterans resident in Oregon, Washington, California, Georgia, or North Carolina. Among the 1263 deployed subjects, 21 incident cancers were reported, compared to 3 among the 516 nondeployed comparison group (OR 3.0, 95% CI 1.0-13.1). Details on year of diagnosis and type of cancer were obtained for 20 of the 24 cases. When all skin cancers (n = 7) and cases that were not confirmed at the time of telephone follow-up (n = 4) were excluded, the OR was 4.94 (95% CI 0.6-38.1) and there was no apparent trend for any specific type of cancer. Among the Gulf War veterans, there was no indication of an association of cancer risk with exposure to nerve agents at Khamisiyah, defined as being within a 50-km radius of Khamisiyah between March 4, 1991, and March 13, 1991 (OR for exposed vs unexposed 0.4, 95% CI 0.1-1.4).

A cohort study of Australian service personnel who had (n = 1456) or had not (n = 1588) been deployed to the gulf was designed to investigate the association of symptoms and medical conditions with immunizations and other Gulf War exposures (Kelsall et al., 2004a). This study was included in Volume 4 as a major cohort study and specifically in the reviews of diseases of the circulatory system, diseases of the respiratory system, and symptoms in general. However, skin cancer, other than malignant melanoma, was also one of the 15 most frequently doctor-diagnosed medical conditions reported by the participants. After medical record review, 92 deployed veterans and 110 nondeployed veterans had a possible or probable diagnosis of the malignancy. The OR, adjusted for service type, rank, age, education, and marital status was 1.0 (95% CI 0.7-1.3).

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