might have missed hospitalizations for a new or poorly recognized syndrome(s). Hospital discharge coding might have inconsistently classified such hospitalizations under many different diagnoses so as to mask an effect, if one were present. The second study operationally defined unexplained illnesses as diagnoses falling under several catch-all International Classification of Diseases, Ninth Revision—Clinical Modification (ICD-9-CM) diagnostic categories entailing nonspecific infections and other ill-defined conditions. After adjusting for hospitalizations only for evaluation (as opposed to treatment) under the DoD registry program, the authors found no significant differences between deployed and nondeployed active duty military (Knoke and Gray, 1998).
These hospitalization studies provide some reassurance that excess hospitalizations did not occur among veterans of the Gulf War remaining on active duty through 1993. Like the mortality studies, however, these studies do not capture illnesses that might have longer latency (e.g., cancer) or illnesses in individuals separated from the military and admitted to nonmilitary hospitals (VA and civilian hospitals) (Haley, 1998). The studies did not measure the utilization of outpatient treatment and would not have detected illnesses unless that did not require hospitalization (Gray et al., 1996; Knoke and Gray, 1998).
Several studies have not identified an excess of birth defects in offspring of deployed versus nondeployed veterans. A small study of two Mississippi National Guard units (n = 282) deployed to the Persian Gulf found no excess rate of birth defects in their children compared with expected rates from surveillance systems and previous surveys (Penman et al., 1996). A much larger study of all live births in military hospitals (n = 75,000) from 1991 to 1993 included a comparison population of nondeployed personnel. The risk of birth defects in children of Gulf War personnel was the same as in the control population (Cowan et al., 1997). This important study, the largest to date on birth defects, was limited to military hospitals, thereby excluding those ineligible for care in military hospitals (i.e., members of the National Guard, reserves, and those who left the military over the course of study). National Guard and reserve troops, as noted earlier in this chapter, constituted a relatively high percentage of U.S. troops deployed to the Persian Gulf (Table 2.1). Anecdotal reports of an excess of Goldenhar’s syndrome, a rare congenital anomaly affecting the development of facial structures, prompted another study of birth defects. Since this birth defect is not specifically coded for in reporting birth defects, the study reviewed medical records of all listings in several more inclusive birth defect categories under which this syndrome would have been subsumed. Araneta and colleagues (1997) found too few cases of Goldenhar’s syndrome from which to draw definitive conclusions.
Several ongoing studies are addressing the limitations of previous studies. Population-based studies to capture births in all hospitals—both military and civilian—are under way in the United States and the United Kingdom. A large U.S. study will pool birth defect data across several states using statewide birth