Haley, R.W. et al. Is There a Gulf War Syndrome: Searching for Syndromes by Factor Analysis of Symptoms. JAMA 1997; 227: 215-222.

Authors attempted to study 606 GW veterans from the 24th Reserve Naval Mobile Construction Battalion (RNMCB-24), a unit which had been studied several times previously by other investigators. 249 (41%) participated. Participants were administered surveys of symptoms and possible risk factors and psychological testing. 175 (70%) of participants reported having had a serious health problem since returning from the GW. Of these 175, 91% felt that their illnesses were certainly or probably due to exposures received during the GW. Through a complicated mathematical technique called factor analysis, the authors were able to derive 6 different groupings of symptoms which they termed "Syndromes" in 63 (25%) of the participants. Participants with Syndromes 2 ("confusion-ataxia") and 4 ("phobia-apraxia) were more likely to be unemployed than those with no health problems. Unemployment among those with Syndromes 1, 3, 5, and 6 and among those with serious health problems (but none of the syndromes) was not significantly different from that of veterans with no serious health problems. Psychological testing disclosed similar profiles across the participants in each of the six syndromes. The authors conclude that the reported symptom Syndromes may represent generalized neurologic injuries possibly due to exposures to neurotoxins occurring during the GW. Limitations of the study are the low response rate (41%), the prior knowledge that this group had a high prevalence of reported symptoms (representativeness), the fact that the participants’ average age (41) was significantly greater than GW veterans as a whole (29), the self-reporting of all health outcomes and the lack of information about diagnosed illnesses, lack of definition of "serious illness", the failure to exclude patients with diagnoses clearly unrelated to the GW, and the employment of factor analysis, a mathematical technique which has debatable relevance to the diagnosis and treatment of disease. Other limitations revolve around the failure to report various types of clinical data that may be relevant to the veterans’ symptoms.