Whether there was a pattern to the multitude of symptoms reported by Australian male Gulf War veterans was studied by Forbes et al. (2004) also using factor analysis. The occurrence of 62 symptoms was assessed in 1322 deployed and 1459 nondeployed veterans. Three factors were identified—psycho-physiological distress, somatic distress, and arthroneuromuscular distress—in both deployed and nondeployed veterans. The authors report that the results did not suggest a unique pattern of self-reported symptoms in the deployed veterans.
To assess the persistence of multisymptom problems in Gulf War veterans, Blanchard (2006) performed a follow-up study of about 1061 deployed and 1128 nondeployed veterans who had participated in the initial studies of Gulf War CMI in the mid-1990s. Participants underwent examinations that included detailed medical and psychiatric histories, physical examinations (general and neurological), pulmonary function and nerve conduction tests, and both laboratory and neuropsychology studies. These investigators concluded that a decade after they were first studied, the Gulf War veterans showed a persistant prevalence of CMI (28.9%) compared with the nondeployed controls (15.8%). Moreover, the difference was exaggerated when only the most severely affected individuals were considered (7.0% vs 1.6%). In the deployed group, CMI correlated with higher combat exposure, while in the nondeployed group it correlated with full-time military service. CMI was associated with more fibromyalgia syndrome, chronic fatigue syndrome, arthralgias, dyspepsia, and metabolic syndrome in both the deployed and nondeployed cohorts. Among the deployed CMI cases, chronic fatigue syndrome was higher than in the nondeployed. In both groups (deployed and nondeployed), prewar non-PTSD anxiety disorders and depression were highly associated with CMI. The authors of this study conclude that “Ten years after the 1991 Gulf War, CMI is twice as prevalent in deployed veterans but still affects 15 percent of nondeployed veterans.” They also note that at 10 years, the prevalence in the deployed group nonetheless seems to have decreased slowly over time; in the deployed group the 10-year prevalence (28.9%) is lower than it had been at 4 and 7 years (44.7% and 47%). Moreover, the authors argue that the critical predictor of CMI is stress. Blanchard and colleagues (2005) concluded with the caution that “Poor mental and physical functioning and metabolic syndrome in veterans with CMI portend a substantial future health-case burden.”
Data from multiple factor studies of veterans in the United Kingdom and the United States were reanalyzed with a different statistical methodology involving dichtonous analysis of multiple lists of symptoms (Nisenbaum et al., 2004). This was undertaken to address the concern that biased findings may result when linear analysis of symptom sets is applied to data that are essentially binary (that is, a symptom is or is not present). The study’s conclusions are broadly similar to those from other factor analyses: similar symptom categories are detected in Gulf War deployed and nondeployed veterans,
except that the gastrointestinal factor in gulf veterans included other symptom types. Correlations among factors raise the question as to whether there is a general illness, even if not unique to gulf veterans, representing the common pathway underlying the identified factors.
Another approach to gauging the impact of Gulf War service is to assess the rates of hospitalizations of deployed versus nondeployed veterans in the years following deployment. Four studies have taken this approach. Gray et al. (1996) performed a retrospective multivariate, logistic regression analysis of the hospitalization rates of 547,076 deployed and 618,335