Prognosis and outcome using broad (DSM-II) and narrow (DSM-III) concepts of schizophrenia.

The classical prognostic indicators of Vaillant and Stephens, the acute onset of psychotic symptoms, and key demographic factors were investigated as predictors of outcome in a prospective study of 153 schizophrenic patients defined using broad (DSM-II) and narrow (DSM-III) concepts of schizophrenia. Findings indicate: Several established prognostic items did not show strong predictive utility in DSM-II or DSM-III schizophrenia when young, nonchronic patients were studied. For both DSM-II and DSM-III schizophrenic patients, longitudinal assessments of psychotic symptoms, work history, and social functioning predicted their respective assessments at followup. However, the predictive utility of some key prognostic indicators shifted for narrow vs. broad concepts of schizophrenia. Most importantly, sex, which was the most powerful predictor of overall outcome among patients with DSM-II schizophrenia, failed to predict outcome among patients with DSM-III schizophrenia, primarily because many women with favorable outcome did not meet the DSM-III criteria for schizophrenia. DSM-III schizophrenia comprises a more homogeneous group of poor prognosis patients in comparison to DSM-II schizophrenia. Educational level, age at first hospitalization, and work history were successful predictors of outcome (p less than .01) when a new, narrow concept of schizophrenia was used.