tion were reported to be more prevalent) in comparison with nondeployed veterans. The authors then used two alternative methods to derive a case definition: factor analysis and a clinical approach. Since both approaches yielded similar case definitions, the investigators chose the latter for its simplicity of application in research.
The authors defined a case of chronic multisymptom illness as having one or more chronic symptoms from at least two of these three categories: fatigue, mood–cognition (e.g., feeling depressed, difficulty remembering or concentrating), and musculoskeletal (joint pain, joint stiffness, or muscle pain). According to this definition, 39 percent of Gulf War-deployed veterans versus 14 percent of nondeployed veterans had a mild-to-moderate case, whereas 6 percent versus 0.7 percent, respectively, had a severe case. Based on a total of 158 clinical examinations performed on one unit, there were no abnormal physical or laboratory findings among those who met the case definition. Cases reported significantly lower functioning and well-being.
A sizable fraction (14 percent) of nondeployed veterans also met the mild-to-moderate case definition. The investigators therefore concluded that their case definition could not uniquely characterize Gulf War veterans with unexplained illnesses (Fukuda et al., 1998). The study, however, had several limitations, the most important of which was its coverage of only active Air Force personnel (several years after the Gulf War), which limits its generalizability to other branches of service, as well as to those who left the service possibly due to illness.
To assess risk factors, the authors performed clinical evaluations on a subset of veterans (n = 158), all of whom volunteered for the evaluation and came from the index unit of the Pennsylvania Air Force National Guard. Forty-five percent of this unit had been deployed to the Gulf War. Overall, there was a dearth of abnormal findings from blood, stool, and urine testing among those who met the case definition for chronic multisymptom illness. There were no differences between cases and noncases in the proportion that seroreacted to botulinum toxin, anthrax protective antigen, and leishmanial antigens, among other antigens. This was the only study to have assessed exposures (mostly to infectious diseases) via laboratory testing, as opposed to self-reports, but the sample undergoing clinical evaluation was relatively small and restricted to Air Force National Guard members.
A large mortality study of nearly all Gulf War-deployed veterans identified no excess postwar mortality, with the exception of motor vehicle accidents (Kang and Bullman, 1996). The study examined mortality patterns through 1993 using two databases, the VA’s Beneficiary Identification and Records Locator