Childhood exposure to ambient air pollution and predicting risk of depression onset in UK adolescents

• In Event: Poster Session 07
  In Poster Session: PS 07 Topic - Developmental Psychopathology

Abstract

Objectives: Knowledge about early risk factors for major depressive disorder (MDD) is critical to identify those who are at high risk so that we can intervene early to prevent its onset and a lifetime of suffering. A multivariable model to predict adolescents’ individual risk of future MDD has recently been developed however its performance in a UK sample was far from perfect. The inclusion of additional factors that are involved in the prediction of adolescent MDD may improve its performance. Emerging evidence indicates a possible role for air pollution in the aetiology of depression. Our aim was to investigate whether including childhood exposure to air pollution as an additional predictor in the depression risk prediction model improves the identification of UK adolescents who are at greatest risk for developing MDD.

Methods: We used data from the Environmental-Risk (E-Risk) Longitudinal Twin Study, a nationally representative UK birth cohort of 2,232 children followed to age 18 with 93% retention. Annual exposure to four outdoor air pollutants – nitrogen dioxide (NO2), nitrogen oxides (NOX), particulate matter with aerodynamic diameters <2.5μm (PM2.5) and <10μm (PM10) – were estimated at address-level when children were aged 10. MDD diagnosis was assessed through clinical interviews with participants at age 18. We used binary logistic regression to first check whether age-10 exposure to each air pollutant was associated with age-18 MDD (adjusted for key covariates). Each pollutant that was associated with MDD was then separately included as an additional predictor in the depression risk prediction model. Change in model performance was assessed using the Net Reclassification Improvement (NRI) method. This quantifies the extent to which a new model correctly reclassifies participants as high or low risk for MDD compared to the original model.

Results: Age-18 MDD was associated with childhood exposure to the highest (top quartile vs. lower three quartiles) annualised average level of NOX (adjusted OR=1.43, 95% CIs=0.96-2.13) and PM2.5 (adjusted OR=1.34, 95% CI=0.95-1.92). These two pollutants and a variable capturing high levels of exposure to either NOX or PM2.5 were then separately added to the depression risk prediction model. NRI analyses (Table 1) showed that these new models substantially reduced the rate of false positive predictions of future MDD (by 13%, 17%, and 13% respectively) compared to the original model without pollution. However, their inclusion also reduced the detection of true positive cases (by 9%, 14%, and 8% respectively) such that overall net improvements to the original model were minimal.

Conclusions: Findings suggest a potential role for childhood ambient air pollution exposure in the development of MDD in late adolescence. However, the inclusion of these ambient pollution exposure estimates did not significantly improve the ability of an existing risk prediction model to correctly identify individuals at risk for future development of MDD at age 18. The inclusion of risk factors other than this – such as genetic liability predictors – may therefore be important for improving the performance of the risk prediction model.

Authors

• Rachel Latham, King’s College London
  Presenting Author

• Christian Kieling, Universidade Federal do Rio Grande do Sul (UFRGS)
  Non-Presenting Author

• Louise Arseneault, King's College London
  Non-Presenting Author

• Thiago Botter-Maio Rocha
  Non-Presenting Author

• Andrew Beddows, Imperial College London
  Non-Presenting Author

• Sean Beevers, Imperial College London
  Non-Presenting Author

• Andrea Danese, King's College London
  Non-Presenting Author

• Kathryn De Oliveira, King's College London
  Non-Presenting Author

• Brandon Kohrt, George Washington University
  Non-Presenting Author

• Terrie E Moffitt, King's College London
  Non-Presenting Author

• Valeria Mondelli, Institute of Psychiatry Psychology and Neuroscience, King’s College London
  Non-Presenting Author

• Joanne Newbury, University of Bristol
  Non-Presenting Author

• Aaron Reuben, Duke University
  Non-Presenting Author

• Helen L Fisher, King's College London
  Non-Presenting Author