PREDICTING FIRST-EPISODE PSYCHOSIS BY BASIC SYMPTOM CRITERIA

Object: Basic symptoms were suggested as a complement to the 'ultra high risk' approach to an early detection of psychosis, and two criteria were derived from it thought to differ in imminence of psychotic breakdown, the risk criterion cognitive-perceptive basic symptoms (APER) and the high-risk criterion Cognitive Disturbances (COGDIS). These were compared with regard to their predictive ability and potentially influencing factors. Method: Criteria were studied prospectively on a sample of 146 putatively prodromal subjects meeting the broader COPER criterion over 19.6+15.4 months with the Schizophrenia Proneness Instrument, Adult version and the PANSS. Results: 124 COPER subjects also met COGDIS; 51 subjects developed first-episode psychosis within 11.019.1 months; 8 of them had only met COPER at baseline. COPER and COGDIS subjects differed significantly on all subscales except the PANSS negative scale (PANSS-N); they did not differ in conversion rate and time or psychotic diagnoses. Comparing converters to non-converters, no differences in psychopathology were found in COPER but in COGDIS subjects; converted COGDIS subjects scored significantly higher on subjective cognitive disturbances, subjective disturbances in experiencing self and surrounding and PANSS-N. Regression models including psychopathology and socio-demographic variables potentially influencing conversion well predicted non-conversion, but failed to predict conversion above chance level. Conclusions: The previous finding of a good predictive ability of certain basic symptoms was supported, yet not the notion that COPER per se delineated a less imminent risk of psychosis than COGDIS. COPER subjects rather appeared as a 'low symptom' subtype of prodromal subjects who might require special prediction.