Children can be lead poisoned when leaded paint is disturbed during home renovation or repair. We conducted a case-control study to assess the association between elevated blood lead levels (BLLs) in children younger than 5 years of age and renovation or repair of homes built before 1950 in New York City. In 1998, we interviewed parents of 106 case children (BLLs >/= 10 micro g/dL) and 159 control children (BLLs </= 5 micro g/dL) living in selected New York City neighborhoods. We then used logistic regression methods to estimate odds ratios (ORs) for elevated BLLs among children living in housing that had undergone various renovations or repairs in the 6 months before the blood lead test, and we adjusted for age and test month. Case children were only slightly more likely than control children to live in a house that had undergone any renovation (OR = 1.2, 95% confidence interval [95% CI] = 0.7, 2.1). Case children were more likely to (1) live in housing that had interior surfaces prepared for painting, especially by hand sanding (OR = 3.5, 95% CI = 1.1, 10.9; population attributable risk [PAR%] = 10.4%, 95% CI = 0.5%, 19.3%); and (2) have work-created dust throughout their housing unit (OR = 6.3, 95% CI = 1.2, 32.3; PAR% = 6.8%, 95% CI = 0.0%, 13.1%). The risk for excess lead exposure is increased by home renovation or repair work involving interior paint preparation or reported dispersal of dust beyond the work area. The proportion of cases related to this exposure is high enough to merit preventive measures.