The committee restricted its review to major controlled epidemiologic studies that evaluated neurological, neuropsychological, and/or psychiatric symptoms and conditions in exposed versus unexposed individuals.³ The committee excluded papers with only neurophysiological outcome measures, such as nerve conduction velocity or vibrotactile thresholds, because these appeared to be further removed from the symptoms typically reported by Gulf War veterans at the time of the committee’s literature review. It was only later that Roland and colleagues (2000) reported on vestibular dysfunction in a small group of Gulf War veterans. This review is limited to publications after 1980, but earlier research supports the basic findings described here (Tabershaw and Cooper, 1966; Rodnitzky et al., 1975; Levin and Rodnitzky, 1976). A complete review of the health effects of OP pesticide exposure will be undertaken by the Institute of Medicine (IOM) in the second phase of this study.

Tables E.1 and E.2 summarize the major features of 15 published studies since 1980 that compare exposed and unexposed individuals to estimate the health effects of OP pesticide exposure. Table E.1 summarizes six studies that demonstrate the longer-term sequelae of OP pesticide poisoning. Table E.2 summarizes 10 studies, one of which is also included in Table E.1, that provide evidence of the longer-term health effects of chronic exposure to OP pesticides at levels insufficient to cause acute effects. Nearly all such studies are cross sectional, observational studies, which by nature are subject to common epidemiologic biases, including selection bias and reporting or information bias (see Chapter 3).

Table E.1 presents findings from six studies reporting on five distinct populations. All the studies were cross sectional, with neurological, neuropsychological, and/or psychiatric outcomes and with previous poisoning by an organophosphate as the main exposure variable. The time at which the health outcome was typically measured was years after the reported poisoning.

In four of the five populations, neuropsychological performance was significantly poorer in the group with previous poisoning (Savage et al., 1988; Rosenstock et al., 1991; Reidy and Bowler, 1992; London et al., 1998). In the fourth population (Steenland et al., 1994), the trend for neuropsychological performance was in the same direction as in the other four studies, but the difference did not achieve statistical significance. The fifth population was actually a subgroup of the fourth. In this subgroup, the poisoning was not sufficient to