Modeling environment through a general exposome factor in two independent adolescent cohorts

Tyler M Moore; Elina Visoki; Stirling T Argabright; Grace E Didomenico; Ingrid Sotelo; Jeremy D Wortzel; Areebah Naeem; Ruben C Gur; Raquel E Gur; Varun Warrier; Sinan Guloksuz; Ran Barzilay

doi:10.1093/exposome/osac010

Modeling environment through a general exposome factor in two independent adolescent cohorts

Exposome. 2022 Dec 14;2(1):osac010. doi: 10.1093/exposome/osac010. eCollection 2022.

Authors

Tyler M Moore^{1

2}, Elina Visoki², Stirling T Argabright², Grace E Didomenico², Ingrid Sotelo², Jeremy D Wortzel², Areebah Naeem², Ruben C Gur^{1

2}, Raquel E Gur^{1

2}, Varun Warrier³, Sinan Guloksuz^{4

5}, Ran Barzilay^{1

2

6}

Affiliations

¹ Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
² Lifespan Brain Institute of the Children's Hospital of Philadelphia (CHOP) and Penn Medicine, Philadelphia, PA, USA.
³ Autism Research Centre, Department of Psychiatry, University of Cambridge, Cambridge, UK.
⁴ Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA.
⁵ Department of Psychiatry and Neuropsychology, School for Mental Health and Neuroscience, Maastricht University Medical Centre, Maastricht, The Netherlands.
⁶ Department of Child and Adolescent Psychiatry and Behavioral Science, Children's Hospital of Philadelphia (CHOP), Philadelphia, PA, USA.

Abstract

Exposures to perinatal, familial, social, and physical environmental stimuli can have substantial effects on human development. We aimed to generate a single measure that capture's the complex network structure of the environment (ie, exposome) using multi-level data (participant's report, parent report, and geocoded measures) of environmental exposures (primarily from the psychosocial environment) in two independent adolescent cohorts: The Adolescent Brain Cognitive Development Study (ABCD Study, N = 11 235; mean age, 10.9 years; 47.7% females) and an age- and sex-matched sample from the Philadelphia Neurodevelopmental Cohort (PNC, N = 4993). We conducted a series of data-driven iterative factor analyses and bifactor modeling in the ABCD Study, reducing dimensionality from 348 variables tapping to environment to six orthogonal exposome subfactors and a general (adverse) exposome factor. The general exposome factor was associated with overall psychopathology (B = 0.28, 95% CI, 0.26-0.3) and key health-related outcomes: obesity (odds ratio [OR] , 1.4; 95% CI, 1.3-1.5) and advanced pubertal development (OR, 1.3; 95% CI, 1.2-1.5). A similar approach in PNC reduced dimensionality of environment from 29 variables to 4 exposome subfactors and a general exposome factor. PNC analyses yielded consistent associations of the general exposome factor with psychopathology (B = 0.15; 95% CI, 0.13-0.17), obesity (OR, 1.4; 95% CI, 1.3-1.6), and advanced pubertal development (OR, 1.3; 95% CI, 1-1.6). In both cohorts, inclusion of exposome factors greatly increased variance explained in overall psychopathology compared with models relying solely on demographics and parental education (from <4% to >38% in ABCD; from <4% to >18.5% in PNC). Findings suggest that a general exposome factor capturing multi-level environmental exposures can be derived and can consistently explain variance in youth's mental and general health.

Keywords: allostatic load; child adolescent psychiatry; mental health; obesity; stress.

Grants and funding

R01 MH119219/MH/NIMH NIH HHS/United States