Genetic risk of psychopathology predicts environmental risk

Abstract

Environmental risk factors for psychosis and psychopathology are often treated as non-genetic, although many are partly heritable or correlated with genetically influenced traits. We tested nine polygenic scores against 49 environmental-risk factors in 307,493 UK Biobank participants without recorded schizophrenia, bipolar disorder, or major depressive disorder diagnoses or antipsychotic medication use. Polygenic scores were generated using PRS-CS and tested in 441 standardised linear models with HC1 robust standard errors. Across all models, 212 associations survived within-score false-discovery-rate correction. ADHD, educational attainment, major depression, schizophrenia and bipolar disorder polygenic scores showed the broadest profiles, but associations differed across five environmental-risk categories. In a secondary analysis excluding victimisation and behavioural measures, seven perception/subjectivity factors showed stronger pooled associations than 25 objective/non-perception factors for schizophrenia, major depression, ADHD, bipolar disorder and neuroticism polygenic scores. Four conceptually tighter matched comparisons produced a similar but non-uniform pattern. Effects were small and do not establish causality. Measured environmental risk is partly correlated with inherited liability through behavioural, social, socioeconomic, cognitive, and perception-related pathways.