Importance Polycyclic aromatic hydrocarbons are widespread urban air pollutants from combustion of fossil fuel and other organic material shown previously to be neurotoxic. Objective In a prospective cohort study, we evaluated the relationship between Attention Deficit Hyperactivity Disorder behavior problems and prenatal polycyclic aromatic hydrocarbon exposure, adjusting for postnatal exposure. Materials and Methods Children of nonsmoking African-American and Dominican women in New York City were followed from in utero to 9 years. Prenatal polycyclic aromatic hydrocarbon exposure was estimated by levels of polycyclic aromatic hydrocarbon- DNA adducts in maternal and cord blood collected at delivery. Postnatal exposure was estimated by the concentration of urinary polycyclic aromatic hydrocarbon metabolites at ages 3 or 5. Attention Deficit Hyperactivity Disorder behavior problems were assessed using the Child Behavior Checklist and the Conners Parent Rating Scale- Revised. Results High prenatal adduct exposure, measured by elevated maternal adducts was significantly associated with all Conners Parent Rating Scale-Revised subscales when the raw scores were analyzed continuously (N = 233). After dichotomizing at the threshold for moderately to markedly atypical symptoms, high maternal adducts were significantly associated with the Conners Parent Rating Scale-Revised DSM-IV Inattentive (OR = 5.06, 95% CI [1.43, 17.93]) and DSM-IV Total (OR = 3.37, 95% CI [1.10, 10.34]) subscales. High maternal adducts were positivity associated with the DSM-oriented Attention Deficit/Hyperactivity Problems scale on the Child Behavior Checklist, albeit not significant. In the smaller sample with cord adducts, the associations between outcomes and high cord adduct exposure were not statistically significant (N = 162). Conclusion The results suggest that exposure to polycyclic aromatic hydrocarbons encountered in New York City air may play a role in childhood Attention Deficit Hyperactivity Disorder behavior problems.