The new primary studies identified by the committee consisted of hospitalization studies that had discharge diagnoses of some form of musculoskeletal disease. Specific diagnoses were not provided in any of the studies.
Gray et al. (1996) used a retrospective cohort approach to compare hospitalization discharge diagnoses for 547,076 Gulf War deployed and 618,335 nondeployed active-duty personnel at DoD medical facilities. Hospitalizations for 14 ICD-9-CM diagnostic categories, which included “musculoskeletal system diseases,” were assessed across three time periods following the war: August 1, 1991, to December 31, 1991; January 1, 1992, to December 31, 1992; and January 1, 1993, to September 30, 1993. Hospitalizations for musculoskeletal system diseases were not increased among the Gulf War deployed personnel versus nondeployed for 1991 or 1992, and were only marginally increased (OR about 1.01, exact value not given; 95% CI included 1.0) for 1993. This study is limited, however, because of the relatively short follow-up, the lack of outpatient data, restriction to DoD hospitals, restriction to hospitalizations of those who remained on active duty after the war, and limited adjustment for potential confounders.
A later publication expanded the Gray et al. (1996) study to include hospitalizations for reserve and separated military personnel over the same three time periods. This study also included discharge diagnoses for hospital stays from DoD hospitals, the VA system, and the California Office of Statewide Health Planning and Development for the years 1991-1994 (Gray et al., 2000). Because the total number of deployed and nondeployed veterans was not available, the researchers calculated proportional morbidity ratios (PMRs). The PMRs for musculoskeletal system diseases for the DoD hospitals, the VA hospitals, and the California hospitals were 1.01 (95% CI 0.99-1.02), 0.86 (95% CI 0.81-0.91), and 0.79 (95% CI 0.64-0.93), respectively. This analysis is limited since it did not include outpatient diagnoses, it could not determine hospitalization rates, and it did not allow adjustment for confounding.
Musculoskeletal system diseases were examined in an additional study comparing hospitalization rates in DoD hospitals through 2000 in three cohorts of veterans: Gulf War veterans, veterans deployed to southwest Asia after the Gulf War, and veterans deployed to Bosnia (Smith et al., 2006). After adjustment for sex, age, marital status, pay grade, race/ethnicity, service branch, occupation, and predeployment hospitalizations, the rate of hospitalizations for musculoskeletal system diseases (identified according to ICD-9-CM discharge codes) was slightly increased in the southwest Asia deployed veterans compared with the Gulf War veterans (HR 1.06, 95% CI 1.01-1.12) and decreased for the Bosnia deployed veterans compared with the Gulf War veterans (HR 0.78, 95% CI 0.71-0.86).
Gray et al. (1999b) assessed hospitalizations for Gulf War veterans potentially exposed to the nerve agents sarin and cyclosarin following the Khamisiyah demolition. Discharge diagnoses from DoD hospitals between March 1991 and September 1995 were examined for 349,291 Army Gulf War active-duty veterans. Plume estimates were overlaid on military unit locations to classify the veterans as no exposure, uncertain low exposure, and three levels of possible subclinical exposures. There was no increased risk of hospitalization for musculoskeletal system diseases for any of the exposure groups (risk ratios all less than 1.0). A follow-up to this study (Smith et al., 2003) examined DoD hospitalization data for active-duty personnel through December 2000. Comparing 99,614 exposed veterans to 318,458 nonexposed veterans, the