Needleman, NL, C McFarland, RB Ness, SE Fienberg and MJ Tobin. 2003. Bone lead levels in adjudicated delinquents. A case control study. Neurotoxicology and Teratology 24:711-717

This study tightens the scientific links between lead exposure and delinquency in teenagers, finding a strong association between elevated bone lead concentrations and the risk of being arrested for criminal behavior.

The underlying causes for this association are not certain. One possibility is that lead interferes with impulse control and that people who have a harder time controlling impulses are more likely to engage in criminal behavior. Another possibility arises from lead's well-established impact on cognitive function and classroom performance. Teenagers with higher lead levels are more likely to fail in school and drop out. Drop-outs, on average, are more prone to committing crimes.

Both causes could be operating at the same time.

Importantly, Needleman et al. are not asserting that all delinquency is due to lead poisoning. Rather, their results indicate that the loss of impulse control (or increased dropout rate) caused by lead poisoning create circumstances where lead-poisoned teenagers are more likely than they would have been to commit serious criminal offenses.

The good news here is lead poisoning can be prevented, pointing toward an important new opportunity for primary prevention of delinquency.

What did they do? Needleman et al. measured lead levels in the bones of 194 males who had been arrested, found by the Juvenile Court of Allegheny County to be delinquent and assigned to one of two programs within the county for serious delinquents. Both programs serve as alternatives to jail or to provide care after jail. Many of the juveniles in these programs are serious offenders, classified by the court as drug dealers, auto thieves, robbers, etc.

They then compared these lead levels with 142 controls chosed from 6 of Pittsburgh's 11 high schools. Controls were evaluated using a questionnaire designed to eliminate "true delinquents" from the control group who engaged in behaviors similar to the delinquents but somehow simply not been arrested.

Needleman et al. go to great length to control for confounding variables such as race, number of parents in the household, parental education, parental occupation, number of children in the home and neighborhood crime rate.

What did they find? Overall, delinquents had significantly higher lead levels than controls: 11 vs 1.5 ppm. This finding was highly significant (p=0.007). For all subjects, after adjusting for a range of potentially confounding variables, the odds ratio was 3.7 (95% confidence interval 1.4 to 11.1).

Importantly, while race and the number of parents in the home were important factors in predicting delinquency, Needleman et al. showed convincingly that the effect of lead was independent of these two variables.

In addition to analyses with data from both races pooled, Needleman et al. also performed analyses on white and black delinquents separately, comparing them, respectively, to white and black controls. The same pattern held: delinquents had higher lead levels. The odds ratios for white delinquents compared to white controls was 3.8 (1.1 - 13.1), while for black delinquents vs black controls it was 2.2.(0.5 - 10.0).

White delinquents had higher bone lead levels than African American delinquents: 20 vs 9 ppm.

White delinquents were more likely to come from single-parent homes than white controls.

 

What does it mean? As Needleman et al. comment, "most studies of the causes of criminal behavior have focused on social risk factors; less attention has been given to the influence of brain dysfunction." Their results shine a spotlight on this alternative influence, revealing that an acknowledged developmental neurotoxicant, lead, is found at significantly elevated levels in the bones of juvenile delinquents compared to controls.

What might drive this association? Needleman et al. review a series of studies demonstrating that lead-exposed children are more likely to be impulsive and to have attention problems. The detail mechanism by which lead exerts this influence is not known, but lead's ability to affect a large number of sites in the central nervous system (CNS) is well-established, including brain centers that affect impulse control. These effects are developmental: exposure in the womb and in early childhood disrupts proper brain development, in ways that can be permanent.

How important is this observation? Citing prior work by T.E. Moffitt, Needleman et al. observe:

  Abnormal CNS function is frequently found in the most serious delinquents. Those delinquents who display anti-social behavior early in life persist in their behavior, while those who start later tend to give it up in their late teens. Those offenders in the group of life-persistent delinquents have a higher rate of impaired neuropsychological function when compared to those begin later. This early onset/life persistent group, approximately 6% of the population, is responsible for 50% of the crime.  

 

One of the details of Needleman et al.'s results warrants further comment. They found that the effect of lead was stronger in white delinquents than in black. According to Needleman et al.: "This apparent paradox occurs because non-lead social factors that raise the risk of delinquincy increase the number of subjects in the low lead/delinquent case group." In other words, other factors pushing kids toward delinquency are more frequently encountered by black teenagers than by white in this study population. As a result, the pool of delinquent black subjects include a larger proportion of teenagers whose delinquency was facilitated by non-lead factors. This statistically dilutes the apparent strength of lead's effect.

Most importantly: "if other studies find a similar association between lead and delinquency, a sizable segment of this important societal problem would become accessible to primary prevention," i.e., reducing lead exposures.