62. POSTPARTUM PSYCHOSIS: POLYGENIC LIABILITY AND CLINICAL COURSE

Background Postpartum psychosis (PP) is a severe psychiatric disorder occurring shortly after childbirth characterized by a rapid onset of symptoms, a fluctuating course, and often confusion. Although associated with bipolar disorder, its long-term clinical course, and its clinical and polygenic overlap with other psychiatric disorders remain incompletely characterized. We therefore examined both clinical outcomes and polygenic profile of PP in relation to other major psychiatric disorders. Methods Participants were drawn from Swedish studies of bipolar disorder (Swedish Bipolar Collection, SWEBIC), schizophrenia (Swedish Schizophrenia Study, S3), and electroconvulsive therapy (Predictors for ECT, PREFECT) with participants linked to Swedish National Registers. PP was defined as (a) psychiatric inpatient admission within 42 days (6 weeks) postpartum, or (b) an inpatient or outpatient diagnosis of PP (ICD-10 F53.1; ICD-8 294.4) within one year of delivery. Women hospitalized for psychiatric disorders during pregnancy were excluded to avoid misclassification of ongoing illness. The sample included 425 women with PP, and 5,579 parous women with psychiatric disorders without PP (psychiatric controls). Clinical outcomes were obtained from interviews and longitudinal national registers. Polygenic scores (PGS) for bipolar disorder, schizophrenia, major depressive disorder, ADHD, and educational attainment (EA) were calculated using SBayesRC. Associations of PP with clinical outcomes and polygenic scores were examined using regression models within each cohort, followed by random-effects meta-analysis. Results PP was associated with higher psychiatric admissions rates (β=0.35, 95%=CI 0.1–0.6, PFDR=0.014), fewer suicide attempts (OR=0.58, 95% CI=0.41–0.8, PFDR=0.004, and fewer with affective episodes during the past 12 months assessed at outpatient visits (OR=0.53, 95% CI=0.39–0.72, PFDR < 0.001), with no differences in functional outcomes or lithium response. In genetic analyses, PP showed higher PGS for bipolar disorder (OR=1.3, 95% CI=1.16–1.46, PFDR < 0.001), educational attainment (OR=1.2, 95% CI=1.08–1.34, PFDR=0.002), and schizophrenia (OR=1.17, 95% CI=1.04–1.32, PFDR=0.014) compared with psychiatric controls, with no differences for ADHD or major depressive disorder PGS. Discussion PP shows a clinical course marked by severe but remitting episodes and a polygenic profile with elevated bipolar and schizophrenia liability together with enrichment for educational attainment. These findings support positioning PP within the affective–psychotic spectrum, related to bipolar disorder but with partially distinct clinical and genetic characteristics.