Effects of family genetic risk scores and environmental factors on risk of schizophrenia and bipolar disorder.

Both genetic and environmental risk factors contribute to the development of schizophrenia (SCZ) and bipolar disorder (BD), but simultaneous investigation of how these factors influence risk has not yet been comprehensively examined in a large population. Therefore, we aimed to investigate and quantify how a multi-generational index of aggregated genetic risk (family genetic risk scores, FGRS) and environmental exposures jointly contribute to risk for SCZ and BD, and whether these relationships differ between the disorders. We conducted a Swedish register-based matched nested case-control study with 3057 SCZ and 15,029 BD cases diagnosed 1988-2013. We used conditional logistic regression to determine individual and joint effects of established environmental risk factors including adverse childhood experiences (ACEs), substance use, adverse perinatal factors, childhood infections, urban birth and longest residence, and FGRS (quintiles) for SCZ and BD and risk of SCZ/BD. We also estimated population attributable fractions (PAFs) for environmental exposures. FGRS were associated with incremental increases in SCZ and BD, with highest risk observed for the highest quintile (SCZ IRR 9.63, 95% CI 7.17-12.94; BD IRR 6.30, 95% CI 5.80-6.84). FGRS and most environmental exposures were independently associated with risk of SCZ and BD. The greatest PAFs were observed for substance use (SCZ 18.3%; BD 13.2%) and ACEs (SCZ 14.1%; BD 19.8%). FGRS are associated with increased risk for SCZ and BD and were largely independent from environmental risk factors. Potentially modifiable factors, ACEs and substance use, accounted for a moderate proportion of all cases and were particularly impactful for BD (PAF 19.8, 95% CI 18.4, 21.2) and SCZ (PAF 18.3, 95% CI 17.6-19.0), respectively.