nondeployed veterans of the US Army, Navy, Marine Corps, and Air Force who remained on active duty through September 1993. In the 25-month period following the war (defined as August 1991 through September 1993), total hospitalizations of the deployed group were not increased as compared to the nondeployed group. In the deployed group, there was a small increase in admissions for testicular cancer and for genitourinary problems, ascribed in part to delayed care for problems that developed during the deployment itself. In the 25 months following the war, the deployed cohort had increased numbers of admissions for alcohol and drug abuse, and for adjustment reactions. The authors concluded that veterans of the Gulf War who remained on active duty through September 1993 were not at risk for unexplained disorders severe enough to merit hospitalization. They acknowledge that a potential limitation of this report is that it omits consideration of data both from those who left the military immediately after the war (before September 30, 1993) and from individuals whose symptoms developed after that date.
An extension of this study was subsequently provided by Knoke and colleagues (1998). These authors tested the hypothesis that the study by Gray et al. (1996) may have underestimated admissions for veterans with obscure, undiagnosed disorders, which might be missed by conventional ICD-9 coding. In this follow-up study, Knoke reviewed all admissions that entailed any “illness of unknown cause” as defined by 77 ICD-9 categories used by the CDC Emerging Infections Program to monitor death certificates for unexplained deaths. This study reviewed records for 552,111 deployed and 1,495,751 nondeployed veterans followed through March 1996. Briefly, after excluding a surge of admissions for evaluation of symptoms of unknown causes, this study found no excess of hospitalization for unexplained illnesses in deployed versus nondeployed veteran.
To encompass in this study those veterans who left service prior to September 1993 (for example, reservists and former veterans), Gray et al. (2000) also performed an analysis of all veterans admitted to three health-care systems in California, covering the deployment period 1991-1994. Because of limitations in acquiring data for discharged veterans, this study compared proportional morbidity ratios of discharge diagnoses between deployed and nondeployed veterans in each of the three systems, rather than hospitalization rates. Most major disease categories revealed no differences (for example, infectious diseases and cancer); in general, the deployed Gulf War veterans had overall proportions of hospitalizations that were similar to the nondeployed. The deployed cohort had a slight propensity for more hospitalizations for fractures, soft-tissue injuries, asthma, and other symptoms. The significance of these factors was difficult to assess.
Recently, Smith et al. (2006) have reexamined hospitalization rates using a longer follow-up period and including three comparative groups from three military theaters: Gulf War (lapsed time 10 years, 5 months, n = 445,465), post-Gulf War southwest Asia (9 years, 5 months, n = 249,047), and Bosnia (5 years, 1 month, n = 44,341). The central finding in this study is that individuals deployed to southwest Asia had slightly more hospitalizations that the Gulf War cohort, while the risk of hospitalization was slightly decreased for the Bosnian cohort. The authors conclude “It is unlikely that Gulf War veterans are at greater risk of hospitalization due to specific exposure-related disease.”
In a follow-up study to Kang et al. (2000, 2002), Kang et al. (2009) conducted a survey in 2005 to obtain health information from the 15,000 Gulf War deployed and 15,000 Gulf War era