Risk architecture of postpartum psychosis

Postpartum psychosis is a severe mood disorder, one of the most severe psychiatric conditions, with high risks of suicide and infanticide if untreated, and should be treated as a medical emergency. It is generally considered a bipolar spectrum disorder, yet, this disorder has not been classified in current disease classification systems because the underlying neurobiology and risk architecture is unclear. In particular, it is unknown how postpartum psychosis fits within the bipolar spectrum, and thus prevention and treatment guidelines are lacking. The long- term goal of this project is to identify the distinct risk architecture of postpartum psychosis. The overall objectives in this application is to use unique large epidemiological data from the Swedish national registers to characterize and compare the genetic risk architecture of postpartum psychosis with postpartum depression, bipolar disorder, and schizophrenia. Our central hypotheses are: (1) postpartum psychosis shows significant evidence of additive genetic effects; and (2) postpartum psychosis has a high but not complete (not equal to one) genetic correlation with bipolar disorder. The rationale is that by understanding the cross-disorder genetic risk of postpartum psychosis and related disorders, we know more about the neurobiology and how to classify this disorder. Using family-based designs and data for a cohort of 271,303 mothers from the Swedish national registers, we will carry out the following Specific Aims: (1) evaluate the familial risk of postpartum psychosis; (2) quantify the impact of additive genetic effects on risk of postpartum psychosis and (3) estimate the genetic correlation between postpartum psychosis, bipolar disorder, postpartum depression, and schizophrenia. The proposed research is innovative in five main areas: (1) it is built around a large homogeneous population-based resource covering an entire nation, minimizing selection bias; Studies thus far provide evidence for the familiality of postpartum psychosis, but these studies have mainly focused on exploring if there is a familial association between postpartum psychosis and other mental disorders by interviewing probands. They did not include controls and, therefore, were not able to provide an estimate of the risk relative to the general population and these studies did not explore familiality across disorders; (2) linking the Swedish population-based register provides a unique opportunity to determine specific risk effects, such as mother's age at pregnancy, psychiatric history, socioeconomic factors, and previous births; (3) the study uses cutting-edge survival analysis methods to study novel aspects of additive genetics while taking into account censoring; (4) the proposed model delivers the estimates of the specific risk factors for postpartum psychosis beyond genetic confounding; and (5) it provides a novel framework for understanding shared risk across postpartum psychosis and bipolar disorder, postpartum depression, and schizophrenia. Our approach is a new and substantially different approach compared to other studies and will allow us to overcome the challenges of modeling additive genetic effects and genetic correlation, thereby opening new horizons for understanding the risk profile of postpartum psychosis.