To the Editor.
—Dr Needleman and colleagues1 report that lead exposure is associated with increased risk for delinquency in boys. However, their findings are far from definitive. Although the sample was adequate in size (212 after exclusions), and high-risk cases were oversampled, the measures of delinquency, including self-reports at ages 7 and 11 years, parent and teacher Child Behavior Checklist (CBCL) T scores greater than 70, and CBCL delinquency cluster scores, all failed to reach statistical significance. There were, however, some significant associations between lead level and parent and teacher ratings of aggression and other behavioral disturbances. The results of a replication 2 years later were not reported, and there was no indication as to whether they were confirmatory.A major concern in research on lead effects is the control of confounding factors. These variables typically contribute appreciably more to analytic models than does lead. Thus, it was startling