The aim of this study was to describe the PROD-screen, an instrument for screening prodromal symptoms indicating risk for psychotic conversion in the near future. PROD-screen consists of 29 questions assessing performance and symptoms. Clinical construct validity was tested by comparing scores from the unselected general population (GP, n = 64) with those of general psychiatric patients from a community mental health centre (CMHC, n = 107). The concordant validity of PROD-screen for prodromal symptoms of psychosis was assessed in a large epidemiologically mixed sample of research subjects (n = 132) by comparing PROD-screen scores with the prodromal diagnosis made by Structured Interview for Prodromal Symptoms as a gold standard. Using the cut-off point of 2/12 specific symptoms, PROD-screen gave correct classification of prodromal status in 77% of cases, distinguishing prodromal from non-prodromal subjects with reasonable sensitivity (80%) and specificity (75%) in the epidemiologically mixed sample. According to subsample analysis PROD-screen functions well with first-degree relatives of schizophrenic patients and probably also with general population samples, but not with psychiatric outpatients. In conclusion, PROD-screen is a useful tool for screening prodromal symptoms of psychosis and selecting subjects for more extensive research interviews.