sex was not statistically significant, indicating that the Hg effects were similar among boys and girls. In general, children's test scores were more strongly associated with cord-blood Hg concentration than with maternal-hair Hg concentration or with Hg concentrations in samples of children's hair collected at 1 and 7 years of age. Five tests were selected, on the basis of high psychometric validity, to represent key domains of cognitive function: motor, attention, visual-spatial ability, language, and memory. For the tests selected to represent attention, language, and memory, the percentages of children with adjusted scores in the lowest quartile increased significantly as cord-blood Hg concentration increased (see Figure 5-9).
In an additional set of analyses (Grandjean et al. 1998), the investigators compared the neuropsychological scores of two groups of children: a case group of 112 children with maternal-hair concentrations of 10 to 20 ppm (median, 12.5 ppm) at parturition, and a control group of 272 children with maternal-hair Hg concentrations less than 3 ppm (median, 1.8 ppm) matched to cases on age, sex, year of examination, and care-giver intelligence. Median cord-blood Hg concentrations also differed substantially (59.0 µg/L in the case group versus 11.9 µg/L in the control group). On 6 of the 18 end points, the case group scored significantly lower than the control group (one-tailed p value of 0.05). Those end points were finger tapping (both hands), hand-eye coordination (average of all trials), WISC-R block design, Boston Naming Test (no cues, cues), and California Verbal Learning Test — Children (long-term reproduction). The results of those analyses differ in certain respects from those of the main analyses. First, the set of end points on which the cases and controls differed is similar but not identical to the set of end points that were significantly associated with cord Hg concentration found in the main analyses. Moreover, in contrast to the main analyses, a term for the interaction between Hg and sex was statistically significant for several scores, including the Bender Gestalt Test error score, short-term reproduction on the California Verbal Learning Test — Children, all three finger-tapping conditions, continuous-performance-test reaction time, and average hand-eye coordination score. For all scores, adverse Hg effects were noted for boys but not girls.
Grandjean and colleagues assembled an additional study cohort of 351 children 7 to 12 years old from four riverine communities in Amazonian Brazil (Rio Tapajos) with increased exposures to MeHg due to the