Association of Environmental Toxicants and Conduct Disorder in U.S. Children: NHANES 2001-2004

Joseph M. Braun; Tanya E. Froehlich; Julie L. Daniels; Kim N. Dietrich; Richard Hornung; Peggy Auinger; Bruce P. Lanphear


Environ Health Perspect. 2008;116(7):956-962. 

In This Article

Abstract and Introduction


Objective: The purpose of this study was to examine the association of tobacco smoke and environmental lead exposure with conduct disorder (CD).
Methods: The National Health and Nutrition Examination Survey (NHANES) 2001-2004 is a nationally representative cross-sectional sample of the noninstitutionalized U.S. population. We examined the association of prenatal tobacco, postnatal tobacco, and environmental lead exposure with CD in children 8-15 years of age (n = 3,081). We measured prenatal tobacco exposure by parent report of cigarette use during pregnancy, and postnatal tobacco using serum cotinine levels. We assessed lead exposure using current blood lead concentration. Parents completed the Diagnostic Interview Schedule for Children to determine whether their children met criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) for CD.
Results: Overall, 2.06% of children met DSM-IV criteria for CD in the past year, equivalent to 560,000 U.S. children 8-15 years of age. After adjustment, prenatal tobacco exposure was associated with increased odds for CD [odds ratio (OR) = 3.00; 95% confidence interval (CI), 1.36-6.63]. Increased blood lead levels (fourth vs. first quartile) and serum cotinine levels (fifth vs. first quintile) were associated with an 8.64-fold (95% CI, 1.87-40.04) and 9.15-fold (95% CI, 1.47-6.90) increased odds of meeting DSM-IV CD criteria. Increasing serum cotinine levels and blood lead levels were also associated with increased prevalence of CD symptoms (symptom count ratio, lead: 1.73; 95% CI, 1.23-2.43; symptom count ratio, cotinine: 1.97; 95% CI, 1.15-3.40).
Conclusions: These results suggest that prenatal tobacco exposure and environmental lead exposure contribute substantially to CD in U.S. children.


Conduct disorder (CD) is characterized by persistent behavioral patterns that violate social rules and the rights of individuals (American Psychiatric Association 2000). Children with CD display aggression toward people and animals, intentionally destroy others' property, and chronically steal or deceive. Children with CD are at increased risk for drug and alcohol abuse, antisocial personality disorder, and anxiety-related disorders (Goldstein et al. 2006; Gunter et al. 2006). Public expenditures related to general and mental health care, school services, and juvenile justice for children with CD exceed $10,000 per child per year (Foster and Jones 2005).

The national prevalence of CD is unknown. It is estimated from community-based regional samples that 0.4-3.3% of children and adolescents have CD, with males two to three times more likely than females to receive a CD diagnosis (Lahey et al. 2000; Loeber et al. 2000; Maughan et al. 2004). Prior variation in CD prevalence estimates may be related to the informant, the diagnostic instrument, the version of Diagnostic and Statistical Manual of Mental Disorders (Dsm) used (3rd ed., 3rd ed. rev., or 4th ed.), participants' age, socioeconomic status, and degree of urbanicity (Loeber et al. 2000). To date, no studies have provided an estimate of the national prevalence of CD in U.S. children using criteria of the 4th edition of the DSM (DSM-IV; American Psychiatric Association 2000).

Questions also persist about the underlying risk factors for the development of CD (Burke et al. 2002). Prenatal tobacco smoke exposure has been consistently associated with disruptive behavior disorders, such as oppositional defiant disorder and CD, even after controlling for potential confounders, including sociodemographic factors, prenatal insults, and parental psychopathology (Fergusson et al. 1993, 1998; Wakschlag et al. 1997, 2002; Weissman et al. 1999). Results from a prospective cohort in New Zealand indicate that children exposed to tobacco smoke in utero had CD symptom rates approximately two times higher than unexposed children (Fergusson et al. 1998). Another prospective study of approximately 6,000 Finnish males found that prenatal tobacco smoke exposure was associated with a 1.74-fold increased odds of committing a delinquent act in late childhood or early adulthood (Rantakallio et al. 1992).

Although prior studies have documented associations between prenatal tobacco smoke exposure and CD, the relationship between postnatal environmental tobacco smoke (ETS) exposure and CD is less clear (Fergusson et al. 1998; Weitzman et al. 1992). Weitzman et al. (1992) reported significant increases in the number of behavior problems among children whose mothers smoked only after pregnancy. Compared with children of nonsmoking women, children exposed to postnatal ETS had a 2-fold increased odds of having extreme behavior problems on the Behavior Problem Index of the Child Behavior Checklist (CBCL) (Weitzman et al. 1992). Similarly, in a New Zealand cohort, Fergusson et al. (1998) found an increased number of CD symptoms among children exposed to postnatal ETS. To date, none of these studies has used a biomarker of tobacco smoke exposure to measure the association between ETS and behavior problems. This limitation can result in exposure misclassification, because a substantial proportion of women who report no ETS exposure have measurable cotinine levels (DeLorenze et al. 2002).

The relationship between environmental lead exposure and violent, aggressive, and oppositional behavior was first reported by Byers and Lord (1943). They observed that among 20 lead-poisoned children, 19 who had “recovered” from lead poisoning failed high school or had behavioral problems. Since then, elevated bone and blood lead levels have been associated with an increased risk of juvenile delinquency in late childhood and early adulthood in case-control and prospective cohort studies (Dietrich et al. 2001; Needleman et al. 1996, 2002). Earlier studies observed children with higher blood lead levels than the levels currently seen; thus, inferences may not be directly relevant to contemporary children with lower blood lead levels (Dietrich et al. 2001; Needleman et al. 1996, 2002). Additionally, recent studies have observed associations between blood lead levels below the current Centers for Disease Control and Prevention recommended action level of 10 μg/dL and cognitive and behavioral deficits (Braun et al. 2006; Lanphear et al. 2005). It is unclear whether lower blood lead levels are associated with more severe behavioral problems such as CD in children.

The purpose of this study was to provide an estimate of the national prevalence of CD defined using DSM-IV criteria and to test the hypothesis that exposures to ETS and childhood lead exposure were associated with CD in a large nationally representative sample of U.S. children.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: