Environmental correlates of infant blood lead levels in Boston.

From a blood lead survey of 11,837 births, 249 newborns were enrolled in a 2-year, longitudinal study. Their blood leads (PbB) were measured semiannually, and their homes were visited for repeated collections of dust, soil, indoor air, tap water, and paint. Recent refinishing activity and the sizes of nearby streets were recorded. Overall mean PbB was 7.2 micrograms/dl (SD = 5.3) at birth. PbB did not vary systematically with age. Each subject's average postnatal PbB correlated highly with the amount of lead in dust (r = 0.4, P less than 0.0001) and soil (r = 0.3, P less than 0.001), and with the lead in paint (r = 0.2, P less than 0.01). Dust, soil and air lead levels correlated with one another. Refinishing activity in the presence of lead paint was associated with elevations of PbB. Water lead, proximate traffic, weight of recovered dust, race, maternal age and education, and sex were not predictive of PbB. Multivariate models of PbB were constructed that become increasingly predictive with age (r2 = 20 to 37%). Indoor dust lead, lead in soil, refinishing activity, and season were the independent variables.