Early predictors for late hospitalizations up to 14 years after first episode psychosis

Purpose New hospitalizations after first episode psychosis (FEP) may be viewed as an indicator of instability in a psychotic disorder. In the current study we wanted to analyse long term risk for psychosis hospitalizations after FEP. We also wanted to analyse predictors for late hospitalizations, with focus on early antipsychotic medication. Methods First episode psychosis cases were recruited to the Swedish Parachute project in 1996-1997. The program offered highly available and continuous psychosocial support and a cautious use of antipsychotic medication for 5 years from inclusion. Longitudinal data from population registers on psychiatric hospitalizations up to 14 years after inclusion were analysed. One hundred and sixty-one cases were included of the original 175 in the project. Associations with possible early predictive factors from the original project data were analysed with COX regression. Results A majority of the cases (67%) had hospitalizations in the first year after inclusion in the study. The cohort then diverged into a group (46%) with new hospitalizations for psychosis after the first year, most of them multiple times, and another group (54%) without new hospitalizations for psychosis, many without any late antipsychotic medication. Forty-two percentage of the cases had antipsychotic medication by month 12, and it was significantly associated with later psychosis hospitalizations (HR = 2.5, p value < 0.001). Conclusions The study demonstrates that a large part of FEP cases have a good outcome as measured by absence of new hospitalizations for psychosis, and that many cases may terminate antipsychotic medication within a year of FEP onset without later relapses needing hospitalizations.