Childhood exposure to particulate matter and nitrogen oxides and associations with mental health disorders in early adulthood: testing mediation by cognition in a UK longitudinal cohort study.

Abstract

BACKGROUND Little is known about the mechanisms underlying associations between air pollution exposure in childhood and mental health disorders in adulthood. OBJECTIVE To examine the prospective associations between age-10 air pollution exposure and age-18 mental health disorders and to test potential mediation by impaired cognition at age 12. METHODS We used longitudinal observations of 1969 members of the Environmental Risk Study who were born across England and Wales in 1994-1995. Exposure to nitrogen dioxide, nitrogen oxides (NOx) and particulate matter (PM2.5 and PM10) was modelled for residential addresses at age 10. Past-year prevalence of anxiety, depression, conduct disorder and attention-deficit/hyperactivity disorder was assessed by interview at age 18. Overall cognition (full-scale IQ) and specific domains (fluid ability, crystallised ability and working memory) were assessed at age 12. We employed binary logistic regression to examine pollution-disorder associations and generalised structural equation modelling to examine mediation via impaired cognition. FINDINGS Higher exposure to NOx was associated with greater odds of depression after covariate adjustment (OR=1.25, 95% CI 1.01 to 1.55). No robust associations were evident for the other pollutants or outcomes. Overall cognition (indirect effect (IE): OR=1.00, 95% CI 0.99 to 1.01) and crystallised ability (IE: OR=1.00, 95% CI 0.99 to 1.01) did not mediate the association between NOx and depression. CONCLUSIONS We found no evidence that impaired cognition mediated associations between childhood residential exposure to NOx and depression in early adulthood. CLINICAL IMPLICATIONS Policies to reduce childhood exposure to NOx may help reduce depression in early adulthood. Future research should examine alternative mechanisms.