64. DEVELOPMENTAL TIMING OF PSYCHIATRIC HOSPITALIZATIONS VARIES BY GENETIC LIABILITY IN PSYCHOTIC DISORDERS

Background Psychotic disorders are characterized by substantial heterogeneity in developmental timing and temporal patterning of symptoms across the lifespan. Age of onset varies considerably, as does the extent to which individuals experience an episodic versus chronic, persistently symptomatic course. Real-world, longitudinal health register data make it possible to capture phenotypic heterogeneity in psychotic disorder progression and estimate genetic influences on those differences. In this study, we tested whether variability in the timing of mental health episodes, as proxied by lifetime inpatient psychiatric admissions, was associated with variation in genetic risk. Methods Genotype and complete, longitudinal inpatient psychiatric hospitalization records were available for 8430 adults from the SUPER-Finland study (49.62% female; 88,174 total admissions). Because hospitalization records began in 1969, we restricted analyses to individuals born in 1954 or later, ensuring that inpatient psychiatric hospitalization data were available from at least age 15 onward for all participants. All participants had a history of at least one psychotic episode in their lifetime. We fit a novel, three-state multistate Cox model in which hospitalization course was parameterized as 1) pre-first admission, 2) hospitalized, and 3) out of the hospital. We jointly tested whether polygenic scores (PGS) for schizophrenia, bipolar disorder, and major depressive disorder were associated with first admission timing, hospitalization duration, and readmission risk. Hazard ratios (HR) represented the likelihood of transitioning to the next state. We adjusted for cohort differences in hospitalization practices, age, sex, and 10 genetic principal components. We also performed a series of sensitivity analyses testing whether associations varied by reason for admission, voluntary versus involuntary hospitalization, and diagnosis. Results Differences in the developmental timing of psychiatric hospitalizations were associated with genetic differences. Higher schizophrenia PGS was associated with earlier first psychiatric admission (HR=1.15, 95% CI: 1.12-1.18), longer inpatient stays (0.96, 0.94-0.97), and longer time until readmission (0.98, 0.96-1.00). In contrast, higher MDD PGS predicted shorter time to readmission (1.03, 1.01-1.05) and shorter inpatient stays (1.03, 1.01-1.05), but not first admission timing. Bipolar disorder PGS was not significantly associated with hospitalization onset or course. Discussion Our results suggest that psychiatric PGS predict not only lifetime case/control status, but also differences in psychotic disorder trajectory. Among adults with psychotic disorders, higher polygenic burden for schizophrenia was associated with a more chronic hospitalization course including earlier first admission and longer stays. Higher MDD PGS, on the other hand, was associated with a more episodic course characterized by shorter, more frequent admissions. Integrating longitudinal, clinically relevant phenotypes and genomics may help parse heterogeneity in psychotic disorder onset and progression.