The purpose of the study was to assess the neurocognitive status of 6-month-old infants whose mothers were exposed to low but varying amounts of lead during pregnancy. Lead levels in the cord blood were used to assess environmental exposure and the Fagan Test of Infant Intelligence (FTII) assessed visual recognition memory (VRM). The cohort consisted of 452 infants of mothers who gave birth to babies at 33-42 weeks of gestation between January 2001 and March 2003. The overall mean lead level in the cord blood was 1.42 microg/dl (95% CI: 1.35-1.48). We found that VRM scores in 6 month olds were inversely related to lead cord blood levels (Spearman correlation coefficient -0.16, p=0.007). The infants scored lower by 1.5 points with an increase by one unit (1 microg/dl) of lead concentration in cord blood. In the lower exposed infants (1.67 microg/dl) the mean Fagan score was 61.0 (95% CI: 60.3-61.7) and that in the higher exposed group (>1.67 microg/dl) was 58.4 (95% CI: 57.3-59.7). The difference of 2.5 points was significant at the p=0.0005 level. The estimated risk of scoring the high-risk group of developmental delay (FTII classification 3) due to higher lead blood levels was two-fold greater (OR=2.33, 95% CI: 1.32-4.11) than for lower lead blood levels after adjusting for potential confounders (gestational age, gender of the child and maternal education). As the risk of the deficit in VRM score (Fagan group 3) in exposed infants attributable to Pb prenatal exposure was about 50%, a large portion of cases with developmental delay could be prevented by reducing maternal blood lead level below 1.67 microg/dl. Although the negative predictive value of the chosen screening criterion (above 1.67 microg/dl) was relatively high (89%) its positive predictive value was too low (22%), so that the screening program based on the chosen cord blood lead criterion was recommended.