Background: Epidemiological research shows that smoking is associated with psychiatric disorders and psychotic experiences, even after controlling for confounds such as cannabis use and sleep problems. We investigated degree of genetic overlap and tested for causal associations between smoking and psychiatric traits and disorders using genetic data. We tested whether genetic associations existed beyond genetic influences shared with confounding variables.Methods: Genetic correlations were estimated with LD-score regression between smoking behaviours (N=262,990-632,802) and psychiatric disorders (schizophrenia, bipolar disorder and depression; N=41,653-173,005), psychotic experiences in adolescents (N=6,297-10,098) and adults (N=116,787-117,794) and adult schizotypy (N=3,967-4,057). Genomic Structural Equation Modelling was performed to explore the associations while accounting for genetic influences of confounders (cannabis and alcohol use, risk-taking and insomnia). Causal associations were tested using Generalized Summary-based Mendelian Randomization (GSMR).Results: Significant genetic correlations were found between smoking and psychiatric disorders (rg = .10 -.38) and adult PE (rg = .33 -.40). After accounting for covarying genetic influences, genetic associations between most smoking phenotypes and schizophrenia and depression remained but not between smoking behaviours and bipolar disorder or most psychotic experiences. GSMR results supported a causal role of smoking initiation on psychiatric disorders and adolescent cognitive and negative psychotic experiences.
Conclusions: Pleiotropy between smoking behaviours and schizophrenia and depressionexists beyond the common genetic basis of known confounders. Smoking also appears to be causally associated with psychiatric disorders and with cognitive PEs and negative symptoms during adolescence. Exploration of the biological links underlying smoking and psychiatric illness would be well-justified.