Effects of organochlorine chemicals on the reproductive outcome of humans who consumed contaminated Great Lakes fish: an epidemiologic consideration.

Three sets of studies of the impacts of human exposure to PCB contaminated fish from the Great Lakes basin--the Michigan Sports Fisherman Cohort, the Michigan Maternal/Infant Cohort, and the Wisconsin Maternal/Infant Cohort-were evaluated using the epidemiologic criteria of Susser (1986). The studies were compared against each other, and against comparable data from other geographic locales. A total of seven major categories of exposure sequelae were evaluated. These ranged from the effects of primary exposure to contaminants upon maternal health status, to effects from secondary intrauterine fetal exposure, including alterations in birth size and gestational age, changes in neonatal health status, and effects persisting into early infancy. Results of the evaluations suggest that the causal hypothesis may be strongly affirmed for the relationship between PCB exposure and alterations in both neonatal health status and in health status in early infancy may be affirmed with reasonable certainty. While the evidence from the Michigan Maternal/Infant Cohort related to maternal exposure to PCB and infant size at birth and gestational age affirms the causal hypothesis, studies from other geographic locales tend only to be supportive. Analytic differences are likely responsible for this variation, but epidemiologically, the composite rating must be regarded as indeterminate. The relationship with observed alterations in maternal health status, composite activity ranking, and McCarthy Memory Scale deficits were also classified as indeterminate. No evidences of obvious negation were seen, although one portion of a study was disqualified because of incoherence.