Table 3 Associations between dentine lead levels and crime, after adjustment for (a) covariate factors and (b) intervening education measures
OutcomeLead levels (μg/g)Regression parameters*
0–23–56–89–1112+BSEp Value
Adjusted mean number of violent/property convictions, ages 14–21†0.160.230.350.520.790.410.180.02
Adjusted mean number of self-reported/violent/ property offences, ages 14–21‡7.407.938.499.109.740.070.110.52
  • *Negative binomial regression.

  • †Covariate factors: family socioeconomic status at birth (B = 0.37, SE = 0.14, p = 0.01); ethnicity (B = −1.5, SE = 0.53, p = 0.004); parental criminal offending (B = 1.7, SE = 0.51, p<0.001); parental alcoholism (B = 1.1, SE = 0.52, p<0.001). Intervening factors: leaving school without qualifications (B = 1.89; SE = 0.50, p<0.001); grade point average, ages 11–13 (B = −0.23, SE = 0.25, p = 0.35).

  • ‡Covariate factors: maternal education (B = 0.41, SE = 0.13, p = 0.002); ethnicity (B = −0.52, SE = 0.32, p = 0.11); family conflict (B = 0.004, SE = 0.001, p = 0.02); exposure to childhood physical abuse (B = 0.44, SE = 0.15, p = 0.004); maternal smoking during pregnancy (B = 0.03, SE = 0.02, p = 0.05). Intervening factors: leaving school without qualifications (B = 1.03; SE = 0.28, p<0.001); grade point average, ages 11–13 (B = −0.16, SE = 0.13, p = 0.21).