Reductions in postdischarge suicide after deinstitutionalization and decentralization: a nationwide register study in Finland.

OBJECTIVE: This study analyzed trends in suicides occurring after a psychiatric hospitalization during more than a decade of significant structural changes in mental health services in Finland-that is, deinstitutionalization, downsizing of inpatient care, and decentralization.

METHODS: Retrospective register data on completed suicides and psychiatric inpatient treatments were collected for the periods 1985-1991 and 1995-2001, representing service provision before and after significant structural changes. The data were used to produce an estimate for a change in postdischarge suicide risk.

RESULTS: In both periods, a fifth of suicide victims had been psychiatrically hospitalized within the preceding year. Among persons hospitalized, the risk of suicide was greater in 1985-1991 than in 1995-2001 for both one week after discharge (risk ratio [RR]=1.50, 95% confidence interval [CI]=1.38-1.62) and one year after discharge (RR=1.25, CI=1.19-1.30). When types of disorders were analyzed separately, the relative risk of suicide one year postdischarge for those hospitalized in the earlier period was greater for patients with schizophrenia (RR=1.26, CI=1.17-1.36) and patients with affective disorders (RR=1.60, CI=1.48-1.73). In parallel with general development of inpatient psychiatric services, in 1995-2001 the inpatient treatment periods preceding suicides were significantly shorter (a mean+/-SD of 45+/-340 days in 1995-2001, compared with a mean of 98+/-558 days in 1985-1991), the number of individual patients treated in the hospital for schizophrenia spectrum disorders was lower (26% compared with 36%), and the number treated for affective disorders was higher (45% compared with 35%).

CONCLUSIONS: The restructuring and downsizing of mental health services was not associated with any increase in suicides immediately (one week) or one year postdischarge. Instead, the risk of these suicides decreased significantly between the two time periods among several diagnostic categories. Although the role of psychiatric hospitalization in general may have changed over time, patients who are hospitalized now may be less suicidal after discharge. Our results indicate, in terms of postdischarge suicides, that the downsizing of psychiatric hospitals has been a success. However, there is still a substantial need for better recognition of suicidal risk among psychiatric patients.