Retrospective study of the impact of lead-based paint hazard remediation on children's blood lead levels in St. Louis, Missouri.

A retrospective follow-up study was conducted to evaluate the effectiveness of lead-based paint hazard remediation in reducing children's blood lead levels. The authors reviewed existing St. Louis, Missouri, City Health Department records, identified 185 children younger than age 6 years who had blood lead levels > or = 25 micrograms/dl during 1989 or 1990, and compared changes in blood lead levels among children whose dwellings did and those whose dwellings did not undergo remediation. Among 54 children who had not moved or received chelation therapy and whose blood lead levels were measured 10-14 months after diagnosis, the geometric mean blood lead level decreased 23% among children living in remediated dwellings (n = 37) and 12% among children in nonremediated dwellings (n = 17) (p = 0.07, t test). The estimated size of the remediation effect was similar using multiple regression (-13%; 95% confidence interval (CI) -25 to 1; p = 0.06) and an approach based on generalized estimating equations (-16%, 95% CI -25 to -7; p = 0.002), when adjusted for covariates. The effect of remediation was greater among children whose blood lead levels at diagnosis were > or = 35 micrograms/dl (-22%) than among those whose blood lead levels at diagnosis were between 25 and 34 micrograms/dl (-1%). Among lead-poisoned children in St. Louis, children whose dwellings undergo lead-based paint hazard remediation have a greater decline in geometric mean blood lead level than do children whose dwellings do not, but the effect of remediation may be influenced by the blood lead level at diagnosis.