[Discharge curve among psychiatric patients after admission and risk factors associated with long stay based on "patient survey"].

The "Reform Vision of Mental Health Services" (2004) announced the basic policy for the transition from hospital based to community based care, and set up numerical objectives, such as the average proportion remaining hospitalized in the first year after admission and the incidence rate of discharge among psychiatric patients hospitalized for more than one year. Using data from the "Patient Survey" performed in 2002 by the Ministry of Health, Labour and Welfare, we estimated discharge curves for each mental disorder during the first year after admission and assessed the effects of variables, i.e., diagnosis, sex, age, hospital type, and residential area, on remaining hospitalized after one year from admission and the incidence rate of discharge among psychiatric patients hospitalized for more than one year. The estimated number of discharged psychiatric patients was 27,974 in September, 2002, and 86% of them were discharged less than one year after admission. The incidence rate of discharge (per 100 person-year) in the first year was 314.8, but the rate after the second year sharply decreased to 19.9. Patients with dementia, mental retardation, and schizophrenia tended to stay for a long period in hospital, and proportions remaining hospitalized after one year from admission were 27.0%, 16.4%, and 14.6% respectively. Based on multivariate analysis using the weighted Poisson regression model, risk factors associated with an increased chance of remaining hospitalized after the first year included a long length of continuous hospitalization, diagnoses of dementia, mental retardation, and schizophrenia, male, older age, and being in a mental hospital. On the other hand, as to the incidence rate of discharge after one year, a long length of continuous hospitalization and being in a mental hospital were related with a long stay, but other variables were slightly different. Being female, patients aged 45-54 years old, and diagnoses of epilepsy and schizophrenia were associated with a long stay. These results clarify the present situation of discharge among psychiatric inpatients and indicate the important variables associated with discharge to prevent new long hospital-stay cases in Japan.