T83. SUBSTANCE-INDUCED PSYCHOSIS LINKED TO BOTH INFECTIONS AND SCHIZOPHRENIA

Abstract Background Substance-induced psychosis is an under-researched phenomenon, and little is known about its etiology (other than exposure to substances) and long-term prognosis. In this presentation, we aim to present results from two recent studies, one of which was recently published and the other is currently in the process of being analyzed. The first study investigates rates and predictors of conversion from substance-induced psychosis; the second study investigates the association between severe infections and substance-induced psychosis, including the contribution of infections on conversion to schizophrenia. Methods Both studies utilized the nationwide Danish registers. In study 1, we included all people diagnosed with substance-induced psychosis from 1994 to 2014 (n=6,788). These were followed using the Kaplan-Meier method and Cox proportional hazards regression to estimate rates and predictors of conversion to schizophrenia or bipolar disorder. In study 2, we included the entire Danish population born since 1981 (n=2,256,779). These were followed in Cox proportional hazards regression models, linking hospital-requiring infections as time-varying covariates to development of substance-induced psychosis. In further analyses, we followed those who had developed substance-induced psychosis to determine whether infections would influence the risk of converting to schizophrenia. Results Study 1: Overall, 32.2% (95% CI 29.7–34.9) of patients with a substance-induced psychosis converted to either bipolar or schizophrenia-spectrum disorders. The highest conversion rate was found for cannabis-induced psychosis, with 47.4% (95% CI 42.7–52.3) converting to either schizophrenia or bipolar disorder. Young age was associated with a higher risk of converting to schizophrenia. Self-harm was significantly linked to a higher risk of converting to both schizophrenia and bipolar disorder. Study 2: Infections increased the risk of substance-induced psychosis (HR=1.30, 95% CI 1.22–1.39) in the fully adjusted model. Hepatitis was the infection most strongly associated with substance-induced psychosis, at HR=3.42 (95% CI 2.47–4.74). Different sites of infections showed associations with different types of substance-induced psychosis. Finally, hepatitis increased the risk of conversion to schizophrenia with HR=1.87 (95% CI 1.07–3.26). Discussion Substance-induced psychosis is strongly associated with the development of severe mental illness, and a long follow-up peri