Lead poisoning remains an important, yet entirely preventable, disease among children worldwide. Children's blood lead levels (BLLs) have been declining in the United States; however, nearly half a million children have BLLs > or = 10 microg/ dl, the level targeted for elimination by 2010. Attainment of this national goal will require translating knowledge into public health practice. The Centers for Disease Control and Prevention (CDC) funds state and local health departments to develop comprehensive prevention programs and surveillance. The Jefferson County, Kentucky Program, which includes Louisville, adopted CDC's recommendation for targeting lead testing to children at highest risk and used knowledge of risk factors for lead poisoning to develop prevention strategies. Blood lead testing was targeted to Louisville neighborhoods at high risk, i.e., those characterized by housing built before 1950 and valued < $50,000, which are known risk factors for BLLs > or = 10 microg/dl among children. We evaluated the impact of these and other interventions. Testing of children aged 9-24 months who were born in high risk housing increased from 64.5% to 73.7% (p-value <0.001) among the 1996 and 2000 birth cohorts. Among the 1996 and 2000 birth cohorts, there was no significant change in testing of children born in low risk housing, i.e., built after 1950 and valued at > or = $50,000 (37.0-37.5%; p-value = 0.649). This report demonstrates that applying scientific knowledge to public health practice and using surveillance and other data to evaluate practice effectively increased testing of children at high risk for lead poisoning, increased lead-safe housing, and empowered communities to protect their children from lead exposure.