Associations of prenatal organochlorine and metal exposures with anxiety and depressive disorder diagnoses in the New Bedford Cohort

Abstract

BACKGROUND AND AIM: The extent to which chemical exposures influence mental health has not been well characterized. We evaluated associations of anxiety and depressive disorder diagnoses with prenatal organochlorine and metal exposures, considering effect modification by quality of the home environment. METHODS: Participants were 172 adults (19-25 years), from a prospective cohort recruited at birth (1993-1998) in New Bedford, Massachusetts. Hexachlorobenzene, p,p'-dichlorobiphenyl dichloroethylene (p,p'-DDE), polychlorinated biphenyls (∑PCB4), lead, and manganese levels were analyzed in cord serum or blood. Physician-diagnosis of an anxiety or depressive disorder in the past ten years was ascertained via self-reported medical history. We estimated covariate-adjusted risk ratios, overall and in strata of adolescent Home Observation for Measurement of the Environment (HOME) score, using complete case and inverse probability weighting (IPW). RESULTS:In our socioeconomically diverse cohort (22% Non-White; 30% household income $20,000 at birth), chemical concentrations were consistent with background exposure levels. Higher prenatal p,p’-DDE was associated with increased risk of anxiety and depressive disorder diagnoses, with effect modification by HOME score. For example, a doubling of cord serum p,p’-DDE was associated with 1.51 increased risk [95% confidence interval (CI): 1.08, 2.11)] of anxiety in those with higher quality home environments and 0.67 lower risk (95% CI: 0.45, 1.00; p-interaction=0.01) in those with poorer HOME scores. This effect modification pattern was also present for ∑PCB4, but associations were attenuated with IPW compared to complete case. In IPW models, cord blood lead was associated with higher risk of anxiety and depression in those with a higher HOME score (e.g., depression risk ratio: 2.04; 95% CI: 1.28, 3.26 per doubling lead). Hexachlorobenzene and manganese were not associated with either anxiety or depression. CONCLUSIONS:Prenatal p,p’-DDE and lead exposure may increase susceptibility to anxiety and depression. Varying risk was observed in different HOME strata, demonstrating the importance of considering interactions between chemical and non-chemical stressors. KEYWORDS: Chemical exposures, heavy metals, metal health outcomes, non-chemical stressors