Background. There are suggestions that mortality, especially that due to suicide, increases among schizophrenia patients during a period of declining psychiatric beds. We investigated the mortality of schizophrenia patients in the general population of Finland during the reduction of psychiatric beds during 1980–1996.

Method. Patients hospitalized for schizophrenia before 31 December 1996, and alive on 1 January 1980 (n=58761) were identified via the National Hospital Discharge Register. General population data came from the National Population Register, and mortality data from the National Causes of Death Register. We calculated relative risks (RR) for total mortality, mortality due to natural causes (cancer, ischaemic heart disease, respiratory disease), unnatural causes (accident, homicide, suicide), and suicide.

Results. Patients with schizophrenia had an increased mortality both from natural causes (RR 2·59, 95% CI 2·55–2·63) and from suicide (RR 9·9, 95% CI 9·43–10·30). The RR for both natural and unnatural deaths was highest among patients with <5 years since onset of schizophrenia. Among them all-cause mortality rose in the 1990s, but decreased among patients with >10 years from onset. Otherwise no major changes or linear trends were found in mortality during deinstitutionalization.

Conclusions. Reduction of psychiatric beds did not generally increase the mortality of patients with schizophrenia. However, patients in their early years of illness experienced increased mortality after the steepest bed reduction. Improved recognition and treatment of somatic illness would benefit patients with schizophrenia.