A linear structural equation modeling procedure was used to explore th
e mechanisms and pathways for lead intake among urban children and the
relative contribution of various lead sources to lead-contaminated ho
use dust. Dust lead levels were significantly associated with children
's blood lead levels, both indirectly and directly via hand lead, Both
soil and paint lead contributed to dust lead levels, but paint contri
buted significantly more lead to house dust than soil (P < 0.001). Bla
ck race and income level both directly affected children's blood lead
levels. Finally, time spent outdoors was associated with children putt
ing soil or dirt in their mouths which was, in turn, associated with c
hildren's blood lead levels. These data indicate that mouthing behavio
rs are an important mechanism of exposure among urban children with lo
w-level elevations in blood lead and that lead-based paint is a more i
mportant contributor of lead to house dust than is lead-contaminated s
oil. (C) 1997 Academic Press.