tionnaire also queried whether the servicemen recalled being exposed to pesticides or had experienced traumatic events during the Gulf War. The authors noted that the scheduling of vaccinations prior to and during the Gulf War was such that routine vaccinations were more likely to be administered before deployment and biological warfare vaccines were more likely to be administered during deployment. All regression analyses controlled for the possible confounding effects of rank, age, branch of service, and education.

The study found that the number of vaccines received prior to deployment was associated with one of the health outcomes (posttraumatic stress reaction). This relationship did not occur when accounting for the number of reported stressors. The number of vaccinations received during deployment was associated with five of the health outcomes (all but posttraumatic stress reaction). Once all vaccinations were taken into account in the analysis, the only two vaccines that showed an association with the CDC multisymptom syndrome were tetanus (adjusted OR = 2.7, 95% CI 1.0–7.2) and cholera (adjusted OR = 2.9, 95% CI 1.0–7.9). However, few records included these vaccinations (3.8 and 3.1 percent, respectively). In an analysis not controlling for all vaccines received, anthrax (OR = 1.4, 95% CI 1.0–1.8) and pertussis vaccination (OR = 1.4, 95% CI 1.0–1.9) were also significant. When the authors further analyzed the data to see if confounders (e.g., number of vaccines received simultaneously, number of years in the military, length of deployment to the Gulf, whether side effects of vaccinations were reported) could account for the association, the association held true.

This study is consistent with the hypothesis that receiving multiple immunizations within a narrow window of time, during a period of presumed stress (deployment to a theater of war), could be associated with the development of multiple symptoms and impaired functional status. One theory for which no direct evidence has been obtained in humans is that such a combination of events could cause alterations in the immune system, in particular a shift in the Th1 to Th2 response (Rook and Zumla, 1997). However, this study was limited by its cross-sectional nature and the fact that it relied on vaccine records that had been retained by only 28 percent of the survey respondents. Other possible confounding factors could be considered, including the timing of the vaccines (servicemen who received multiple vaccines during deployment tended to have been sent to the Gulf earlier, and medical personnel tended to report more vaccines during that period). Despite its limitations, this study was quite large and carefully evaluated the possibility of both confounding and response bias—concluding that neither explained its result.

CDC study. As part of the CDC study (see Chapter 2), Fukuda and colleagues (1998) performed clinical evaluations on a group of Gulf War veterans (n = 158), all of whom volunteered for the evaluation and were a subset of the index unit of Pennsylvania Air Force National Guard (45 percent of this unit had been deployed to the Persian Gulf). Participants in the clinical study were classified as cases (sufferers of the chronic multisymptom condition, as described in



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