Abstract
Background: The early identification of schizophrenia is important in establishing an adequate therapeutic intervention, especially in emergency situations. Intuitive reasoning is often necessary because a standardized classification according to ICD-10 or DSM-IV criteria is not suitable in these complex, and often changing, settings. The process of recognizing the specific schizophrenic structure in an intuitive way by an experienced psychiatrist was named ‘praecox feeling’ by the Dutch psychiatrist Rümke in 1941. Sampling and Methods: To evaluate the diagnostic quality of this kind of intuitive clinical approach we investigated 67 previously unknown patients. All patients displayed acute psychotic symptoms like paranoid delusions and hallucinations belonging to the schizophrenic spectrum. The first interview by an experienced psychiatrist included only the present state psychopathology. Data referring to the course of the illness or medication or family history of schizophrenia were explored later by other independent staff members who were blind to the aim of the investigation. The intensity of the ‘praecox feeling’ was rated in four degrees: not present – mild – moderate – high. At the end of the inpatient period a standardized diagnostic classification according to ICD-10 and DSM-IV classification was carried out by independent raters. Results: Compared to the standardized diagnostic classification the precision of the intuitive reasoning was remarkably high with a sensitivity of about 0.85, a specificity of about 0.80, a positive predictive power of about 0.90, and a negative predictive power of about 0.65 depending on the standardized system used. Cognitive impairment, affective disturbances, disturbed self-perception, and reduced communication skills all correlated with intensity of ‘praecox feeling’. The single variable ‘affective disturbances’ had the highest impact on the intensity of ‘praecox feeling’ in an ordinal regression analysis. Furthermore a high intensity of ‘praecox feeling’ strongly correlated with a hereditary predisposition to schizophrenia. Neither the severity of the mental illness nor extrapyramidal side effects of the medication correlated with the intensity of ‘praecox feeling’. Conclusion: Our work can possibly help in contributing to reflection on our diagnostic practices and help to make the various factors involved in establishing diagnoses apparent.