Screening housing to prevent lead toxicity in children.

OBJECTIVE: Screening children to identify those with blood lead levels > or = 10 microg/dl fails to protect children from lead-associated cognitive deficits and behavioral problems. To broaden our efforts at primary prevention, screening criteria are needed to identify lead-contaminated housing before children are unduly exposed. The purpose of this study was to identify and validate housing characteristics associated with children having elevated blood lead levels (> or = 10 microg/dl).

METHODS: Two existing studies were used to examine housing characteristics linked with undue lead exposure: a cross-sectional study of 205 children aged 12 to 31 months, and a random sample from a longitudinal study of 276 children followed from 6 to 24 months of age. Logistic regression analysis was conducted to examine the association of children's blood lead levels > or = 10 microg/dl.

RESULTS: The mean age of the 481 children was 17.8 months; 99 (20.6%) had a blood lead concentration of 10 microg/dl or higher. The following characteristics were associated with blood lead concentration > or = 10 microg/dl: floor lead loading > 15 microg/ft2 (odds ratio [OR]=2.2; 95% confidence interval [CI] 1.3, 3.8); rental housing (OR=3.2; 95% CI 1.3, 7.6); poor housing condition (OR=2.1; CI 1.2, 3.6); African American race (OR=3.3; CI 1.9, 6.1); paint chip ingestion (OR=5.8; CI 1.3, 26.5); and soil ingestion (OR=2.2; CI 1.1, 4.2). Housing characteristics including rental status, lead-contaminated floor dust, and housing condition had a range of sensitivity from 47% to 92%; specificity from 28% to 76%; a positive predictive value from 25% to 34%; and a negative predictive value of 85% to 93%.

CONCLUSIONS: Housing characteristics and floor dust lead levels can be used to screen housing to identify lead hazards prior to occupancy, before purchasing a home, or after renovation to prevent children's exposure to lead hazards.