Mortality in Kraepelinian schizophrenic patients: a prospective study with a median follow-up of 8.4 years.

Kraepelinian schizophrenia is a subtype of schizophrenia defined by its very poor outcome. Several studies have demonstrated the validity of this distinction, but the expected excess mortality has not been investigated. The aim of the present study was to test the hypothesis that Kraepelinian schizophrenic subjects have a higher mortality rate, notably due to natural causes, than non-Kraepelinian schizophrenic subjects. The results of a prospective study with a median follow-up of 8.4 years are reported. Three hundred ten schizophrenic subjects meeting the International Classification of Diseases, 10th Revision, criteria for schizophrenia were classified on admission as Kraepelinian schizophrenia (n = 31) and non-Kraepelinian schizophrenia (n = 279); 31 deaths (16 from unnatural causes, 13 from natural causes, and 2 from undetermined causes) were observed during follow-up. Univariate and multivariate Cox regression analyses were performed using schizophrenia subtype (Kraepelinian vs non-Kraepelinian), age, and duration of illness as predictors. Schizophrenia subtype was a significant predictor of mortality in univariate analysis only for death from natural causes, and a trend was observed after adjusting for age and duration of illness. A log-rank test was used to compare mortality rates between Kraepelinian and non-Kraepelinian schizophrenic subjects and showed a significantly higher mortality rate from natural causes in Kraepelinian schizophrenia (16.7%) than in non-Kraepelinian schizophrenia (4.1%).