Background - Follow-up of children whose initial blood lead level is below the poisoning threshold of 100 μg/L but who have one or more risk factors is recommended. Initially, the frequency of follow-up blood lead testing was calculated among children at risk of lead poisoning. Next, the frequency and determinants of the onset of lead poisoning were estimated in the group of children under follow-up. Methods - The Île-de-France Childhood Lead Poisoning Surveillance System (SSSIILF) has systematically collected initial and follow-up blood lead levels from children in the region since 1992. Children who underwent initial screening before the age of seven, had a blood lead level below 100 μg/L, and presented at least one risk factor were selected. Variables associated with the performance of a follow-up blood lead level test were compared using the chi-square test. Among the children followed up, variables associated with the occurrence of poisoning were compared using the chi-square test; independent risk factors for the occurrence of poisoning were estimated using logistic regression. Results - The frequency of follow-up blood lead level testing was statistically significantly higher when children resided in Seine-Saint-Denis or Paris, when children were initially screened by a maternal and child health service or a private physician, when the initial blood lead level was 50 μg/L or higher, among the youngest children at the time of initial screening, and when certain known risk factors for lead poisoning were present. The incidence of lead poisoning during follow-up decreased from 25.9% for children initially screened between 1992 and 1994 to 5.1% for those initially screened between 2004 and 2005 (p < 0.001). It was lower in Paris and Seine-Saint-Denis compared to other departments in the region (p < 0.01). It increased, independently of known risk factors (p < 0.01), when the initial blood lead level was 50 μg/L or higher, among children screened before the age of 24 months, and among children born to mothers from sub-Saharan Africa (p < 0.01). Conclusion - Regular follow-up of at-risk children with initial screening blood lead levels below 100 μg/L is essential, particularly when initial screening is performed before the age of 24 months. Environmental intervention in the home may also be necessary when the initial blood lead level is 50 μg/L and multiple risk factors are identified. (R.A.)