Validation of Danish case register diagnosis for schizophrenia

The ABC schizophrenia study aims at investigating sex differences in age of onset, symptoms and course of schizophrenic and paranoid disorders. For this purpose, we used case register data from Denmark and Mannheim and a directly examined sample of first admissions (ABC sample). The Danish case register sample included less clinical diagnoses of schizophrenia and more schizophrenia-related disorders (acute paranoid reaction, paranoid states and borderline schizophrenia) than the Mannheim data (case register and ABC sample). The problem therefore was whether the two datasets are comparable and the results are valid. For this reason a randomized, stratified sample of 116 patients was drawn from the Danish case register sample. The case notes of these 116 patients were requested from the hospitals where the patients had been treated and analyzed by means of a scoring sheet based on the Interview for the Retrospective Assessment of the Onset of Schizophrenia (IRAOS). The use of operationalized diagnoses of the CATEGO program, based on PSE items, which are integrated in IRAOS, demonstrated that the samples of the Danish and the Mannheim case registers and the directly investigated ABC sample have comparable diagnostic distributions. Possible explanations for the differences between the clinical and the CATEGO diagnoses in the Danish case register may be the frequent use of diagnoses of borderline schizophrenia and reactive psychoses (previously called psychogenic psychoses), and above all a more narrow concept of schizophrenia; in Denmark, schizophrenia is diagnosed relatively late, i.e., after the presence of enduring negative symptoms, and thus mostly after the appearance of residual state. These diagnostic preferences may help to explain the fall in first admission rates for schizophrenia – above all in women – in Denmark and the low incidence rates of schizophrenia by first contact within the WHO determinants of outcome study. The earlier hospitalization of men could be replicated as well as the course of treatment (readmissions and discharges) of schizophrenic men and women over 10 years after first admission.