Purpose: To evaluate the health status of Gulf War veterans who reported receipt of anthrax vaccination and a small group of Gulf War veterans for whom documentation of anthrax vaccination exists.
Methods: Among the 11,441 Gulf War veterans who completed a health survey, 4601 reported receiving the anthrax vaccine during the war; 2979 veterans reported not receiving it; 3861 were uncertain. Also, 352 of these respondents were documented by the Department of Defense as having received anthrax vaccination. We compared the medical history of these groups of veterans using multivariate analyses. Finally, we analyzed perception of exposure and its relation to reporting bias.
Results: There were statistically significant differences in prevalence for almost all outcomes studied between those who reported having received anthrax vaccination and those who did not so report. However, when we compared the veterans for whom vaccination records exist to the group who self-reported that they had not received the vaccine, the significant differences in prevalence for almost all of the outcomes disappeared.
Conclusions: The extent of a reporting bias should be carefully considered when one evaluates the health consequences of anthrax vaccination based on self-reported data.