cause hospitalization but was significant enough to depress blood cholinesterase levels by 60–70 percent. This study did not find consistent differences between exposed and unexposed populations (Ames et al., 1995). Three of the five populations exhibited increased rates of neurological or psychiatric symptoms in the previously poisoned group (Rosenstock et al., 1991; Reidy and Bowler, 1992; London et al., 1998).
Taken together, these cross sectional studies report a consistent tendency toward poorer neuropsychological performance and increased rates of neurological or psychiatric symptoms among persons with prior acute OP poisoning. The time from poisoning until evaluation in these studies is poorly documented but is typically on the order of years.
In each of these studies, analyses were conducted using exposure measured by previous acute exposure (“poisoning”). The poisonings were documented in most subjects on the basis of symptom reporting, hospitalization, and/or depressed cholinesterase levels. For example, the poisoned group studied by Rosenstock and colleagues (1991) had been hospitalized for OP poisoning and had no previous serious neurological or mental disorders. Steenland and colleagues (1994) obtained subjects from the State of California registry of OP poisoning. Given the severity of OP poisoning, there is likely to be little misclassification on this measure. The neuropsychological tests used in these investigations are varied, but are typically standardized and well defined, so that differential misclassification of the response is also unlikely.
Information bias is of greater concern in self-reported psychiatric and neurological symptoms. Persons previously poisoned who had suffered significant health consequences of OP pesticide exposure might be likely to report current symptoms differentially from persons not previously poisoned. The other major potential source of bias in these studies involves confounders that have not been adequately controlled for. Only one of the four studies (London et al., 1998) controlled for chronic OP exposures since the acute event. Hence, it is not possible to be sure that the longer-term sequelae of OP poisoning were due to the exposure that caused the original “poisoning” rather than to subsequent chronic exposures. London and colleagues (1998) did include prior poisoning and current job status as predictors of neurological symptoms and both were statistically significant—indicating that there may be health effects from chronic exposure (job status) over and above those from an acute exposure. There was little control for confounding in the study by Rosenstock and colleagues (1991) where OP exposure effects were most substantial. However, London and colleagues (1998) had the most detailed control for confounders and still found differences between exposed and unexposed groups.
In summary, the available literature indicates that exposure to OP pesticides at levels sufficient to cause acute health effects requiring medical reporting or treatment is associated with elevated rates of neurological or psychiatric symptoms and poorer performance on standardized neuropsychological tests several years after the acute exposure.