Research on at-risk states of psychosis has mainly aimed to predict conversion. Yet as a considerable number of patients does not to progress to this outcome during the investigated observation periods, the course of these non-converters (NC) is of major interest, particularly with regard to preventive interventions and treatment.

To analyze the psychopathological and functional in 18-month non-converters.

Data were derived from the prospective multicenter European Prediction of Psychosis Study with an 18-month follow-up period. Participants had to fulfill ultra-high risk criteria and/or the COGDIS criterion, which is based on a set of cognitive basic symptoms. Psychopathology was assessed with the Structure Interview for Prodromal Syndromes (SIPS), including the Global Assessment of Functioning Scale (GAF) and a short version of the Schizophrenia Proneness Instrument (SPI-A).

All total and subscale scores improved significantly during follow-up. However, a more detailed analysis revealed that a considerable part of the patients showed no improvement or even a worsening of psychopathology and function.

Our first analysis of course on non-converters shows that a high proportion of patients improved. In the light of results from retrospective studies, however, this improvement has to be interpreted with caution, as the observation period does not allow to determine the proportion of outpost syndromes, i.e. precursors of a later prodrome. Furthermore, a considerable portion of our sample worsened functionally and/or symptomatically. With regard to retrospective schizophrenia related results, very long observation periods may be needed to characterize the patterns of course in subpsychotic syndromes.